# **Family Business and Regional Development**

This book explores the relationship between families, frms, and regions and the extent to which these relationships contribute to regional economic and social development.

Although family business participation in economic activities has been a common phenomenon since pre-industrial societies, and its importance has evolved throughout time and across spatial contexts, the book suggests that these factors have often been neglected in family business and regional studies. Taking this research gap into account, the book aims to deepen our understanding of the role family frms play in the regional economy. In particular, it explores two seldom studied questions. Firstly, what role do family frms play in regional development? Secondly, how do different spatial regional contexts shape family frm operations and performance?

*Family Business and Regional Development* presents a model of "spatial familiness" and uses themes such as productivity, networks and competitiveness to shed new light on family businesses. Moreover, it approaches the juxtaposition between family business and regional studies to encourage the cross-fertilisation of ideas, theories, and research methods between the two felds.

Bringing together leading experts in entrepreneurship, regional economics, and economic geography, this book will be a valuable reading for advanced students, researchers and policymakers interested in family frms, regional studies and economic geography.

**Rodrigo Basco** is Associate Professor and Sheikh Saoud bin Khalid bin Khalid Al-Qassimi Chair in Family Business at the American University of Sharjah, United Arab Emirates.

**Roger Stough** was the Northern Virginia Endowed Chair and Professor of Public Policy at George Mason University, USA.

**Lech Suwala** is Professor of Urban and Regional Economics at the Technical University of Berlin, Germany.

# **Routledge Advances in Regional Economics, Science and Policy**

**Economic Clusters and Globalization**  Diversity and Resilience *Edited by Francisco Puig and Berrbizne Urzelai* 

**Transnational Regions in Historical Perspective**  *Edited by Marten Boon, Hein A.M. Klemann and Ben Wubs* 

**Urban Development in China under the Institution of Land Rights**  *Jieming Zhu* 

**The Geography of Mobility, Wellbeing and Development in China** 

Understanding Transformations Through Big Data *Edited by Wenjie Wu and Yiming Wang* 

#### **Traveling Expertise and Regional Development**

*Andreas Öjehag-Pettersson and Tomas Mitander* 

**Governing Cities**  Asia's Urban Transformation *Edited by Kris Hartley, Glen Kuecker, Michael Waschak, Jun Jie Woo, and Charles Chao Rong Phua* 

**Mental Health and Wellbeing in Rural Regions** 

International Perspectives *Edited by Sarah-Anne Munoz and Steve F. Bain* 

**Family Business and Regional Development**  *Edited by Rodrigo Basco, Roger Stough and Lech Suwala* 

For more information about this series, please visit: www.routledge.com/ series/RAIRESP

# **Family Business and Regional Development**

**Edited by Rodrigo Basco, Roger Stough and Lech Suwala** 

First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

and by Routledge 52 Vanderbilt Avenue, New York, NY 10017

*Routledge is an imprint of the Taylor & Francis Group, an informa business* 

© 2021 selection and editorial matter, Rodrigo Basco, Roger Stough and Lech Suwala; individual chapters, the contributors

The right of Rodrigo Basco, Roger Stough and Lech Suwala to be identifed as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.

The Open Access version of this book, available at www.taylorfrancis.com, has been made available under a Creative Commons Attribution 4.0 license.

The authors acknowledge the generous support by the Open Access Publication Fund of Technische Universität Berlin.

*Trademark notice*: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

*British Library Cataloguing-in-Publication Data* A catalogue record for this book is available from the British Library

*Library of Congress Cataloging-in-Publication Data* Names: Basco, Rodrigo, editor. | Stough, Roger, editor. | Suwala, Lech, editor. Title: Family business and regional development / edited by Rodrigo Basco, Roger Stough and Lech Suwala. Description: 1 Edition. | NY: Routledge, 2020. | Series: Routledge advances in regional economics, science and policy | Includes bibliographical references and index. Identifiers: LCCN 2020027916 Subjects: LCSH: Family-owned business enterprises. | Regional economics. | Regional planning. Classification: LCC HD62.25 .F3726 2020 | DDC 338.9—dc23 LC record available at https://lccn.loc.gov/2020027916

ISBN: 978-0-367-17861-1 (hbk) ISBN: 978-0-429-05809-7 (ebk)

Typeset in Times New Roman by codeMantra

# **Contents**


vi *Contents* 



*Index* 273

# **Figures**



# **Maps**


# **Tables**


xii *Tables* 


# **Boxes**


# **Contributors**


background where she has published several articles related to entrepreneurship, self-employment, and frm development.


# xvi *Contributors*

Entrepreneurship at IPAG Business School, Nice, France. He is currently the Global Academic Director of the STEP (Successful Transgenerational Entrepreneurship Practices) Project Global Consortium. He has published journal articles on family frms, internationalization, and corporate governance in leading international peer-reviewed journals such as: *Strategic Management Journal*, *Entrepreneurship Theory & Practice*, *Harvard Business Review* (Case Study Collection), *Journal of Business Research*, *Journal of Business Ethics*, *International Journal of Management Reviews*, and *Family Business Review*.


 Geography from Umeå University in 2010 and is editor for *Geografska Annaler Series B*. His research interests include uneven regional development, labour market externalities, post-redundancy mobility, and the role of entrepreneurial efforts for regional development.


### xviii *Contributors*

His research interests focus on SMEs, family business strategy, business performance and prediction, managerial and entrepreneurial behavior, business and neuroscience.


# **Abbreviations**


# **Introduction**

*Rodrigo Basco, Roger Stough and Lech Suwala* 

### **Family Business and Regional Development**

Family business participation in economic activities has been a common phenomenon since pre-industrial societies, and its importance has evolved throughout time and across spatial contexts. These factors have often been neglected in family business and regional studies. Taking this research gap into account, this edited collection aims to deepen our understanding of the role family frms play in regional economies. Another objective is to recognise the effect of regional contexts on family frms by encouraging the cross-fertilisation of ideas, theories, and research methods between the two felds. In particular, our collection explores the relationship between families, frms, and regions and the extent to which these connections contribute to regional economic and social development. We therefore invite readers on this collective journey as we explore two seldom studied questions*: (1) What role do family frms play in regional development? (2) How do different spatial regional contexts shape family frms' operations and performance?* 

The main rationale is to evaluate what we already know about the juxtaposition and cross-fertilisation between family business and regional studies. This will enable us to discover new lenses and research paths that can strengthen interdisciplinary methods and to envision and propose future lines of research. This edited collection intends to contribute to theory and practice by linking economic, social, and political players, whose interactions at local and regional levels are believed to play a role in their success.

Firstly, by unveiling the correlation between family business and regional studies, this volume explores the merging of families, frms, and regions from an interdisciplinary perspective at different levels of analysis (micro, meso, and macro levels). The connections between family and business have been extensively explored and studied and there is no doubt about the infuence family has on frm decision-making. However, less is known about the intensity of these relationships across contexts, territories, and places. At the same time, regional studies have not recognised the role and importance of family frms as idiosyncratic economic and social players in regional development. Therefore, as an added value, this book addresses an

### 2 *Rodrigo Basco et al.*

interdisciplinary view of the nexus between family frms and regions. By proposing new theoretical reasoning, this collection challenges existing theories to enlighten researchers, family businesses, and policymakers and help them understand and prepare for the phenomenon.

Secondly, this collection, which brings together researchers across family business studies, regional studies, and policymaking felds, is practical in nature. Connecting both theory and practice is our fundamental goal, so as to add value to our explorative and predictive research. Given that family frms are the most common organisational form around the world in developed, emerging, and transitional economies, our interests transcended basic research to include the sole interaction between family and business. In addition, we included the regional and spatial perspectives to unveil the role that family frms play in boosting or hindering local and regional socioeconomic development. From this perspective, research efforts undertaken here connect business families, family frms, regional developers and planners, and policymakers from both angles, to improve the world we live in. If this book can focus attention on the topic of family frms by initiating political debate at local, regional, national, and multinational levels, our initial goal will be accomplished.

This volume is organised into fve segments to accommodate its interdisciplinary interpretation and to guide our readers in approaching the nexus of family business and regional studies. Part I ('Family Business and Regions') aims to introduce the topic of 'Family Business and Regional Development'. *Chapter 1*, by Basco and Suwala, introduces the discussion by examining theoretical fertile grounds between family business and regional studies. Research efforts are carefully approached in terms of both disciplines in the form of a literature review. This review explains and elaborates on the two most integrated models dealing with the nexus of family frms and regions and proposes future lines of research. In *Chapter 2*, Basco and Ricotta present quantitative exploratory research by analysing the persistence of family frms across European regions. In this chapter, the authors address two fundamental research questions: (1) Why are family frms unevenly distributed across regions, and to what extent is this related to the territories in which they exist? (2) Does the presence of family frms affect regional competitiveness, and if so, how and when?

Part II ('Micro-foundation channels') discloses the micro-foundations of the nexus between family businesses and regional development to our readers. In *Chapter 3*, Basco, Amato, Gómez-Anson, and Calabrò theorise following the regional familiness approach and taking into account the fact that family business research is contextless—that family frms are more advantageous than their non-family counterparts when operating in small municipalities because of their emotional and social connections. The authors empirically demonstrate, in terms of productivity, that large urban settings are sources of diseconomies for Spanish family-managed frms. Hence, they are better suited to exploit highly embedded contexts such as small municipalities. In *Chapter 4*, Adjei and Eriksson explore the connection between family co-occurrence in frms and the relationship with frm performance in longitudinal quantitative research conducted in Sweden. They demonstrate that family co-occurrence in workplaces infuences productivity and that the positive or negative impact of familial relationships on productivity is contingent on the type of family ties, members' skills, and the regional context. This section ends with *Chapter 5*, authored by Amato, Patuelli, and Lattanzi. This chapter, which draws on a sample of frms located in fve different Italian industrial districts, suggests that family frms are more engaged in digital innovation than non-family frms. The authors justify their fndings due to the stronger frm-specifc family social capital and embeddedness in local district networks.

Part III ('Meso-foundation Channels') investigates family frms at the regional level by considering the connections activated and triggered by family frms to stimulate (or not) regional development. In *Chapter 6*, Lenz presents qualitative research that focuses on the regional disembedding process of family frms across generations. Exploring several Basque family frms in Spain, Lenz confrms that the differences between incumbents and successors in terms of personal characteristics, network connections, and beliefs on how to conduct business lead successors to experience weaker regional embedding than their predecessors. In *Chapter 7*, Amato, Backman, and Peltonen answer the question of whether family frms in Finland are more locally embedded than non-family ones. They conclude that frms in Finland rely greatly on enduring and spatially proximate stakeholders, and therefore indicate a stronger link with the socioeconomic milieu in which they dwell. This section concludes with *Chapter 8*, in which Graffenberger and Görmar present the role of hidden champions located in small towns in peripheral regions in Germany from a corporate responsibility perspective. The authors indicate that family frms' commitment to supporting corporate activities is higher than that of non-family frms because of familyrelated specifcities and socio-emotional dimensions of local embeddedness.

Part IV ('Evidence around the World') presents three chapters on family business and regional development in the African, European, and Chinese contexts. In *Chapter 9*, Murithi and Woldesenbet explore the self-perception of family and non-family frms in terms of their contribution to regional development in Kenya. Because of the strategic behavioural differences between family and non-family frms, their contribution varies. In *Chapter 10*, Cappelli, Cucculelli, and Peruzzi further develop our understanding of the topic by analysing whether family frm density and industry scale affects entrepreneurship at regional level. Interestingly, the authors fnd that entrepreneurship reacts positively to the prevalence of family frms and negatively to the presence of large companies. However, these results only occur in Western European as opposed to Eastern European regions, in which entrepreneurial initiatives are driven signifcantly by the average frm size. Finally, in *Chapter 11*, Zhang, Dou, and Fang shed light on the connection

### 4 *Rodrigo Basco et al.*

between family frms and regional development in China. This exploratory investigation exposes readers to the unique Chinese context to better understand the role of family frms in regions and, conversely, the effects of regional economic and development on family frms.

We conclude the edited collection with Part V ('A Policymaker Perspective') which provides a practical orientation to connect research and practice. *Chapter 12*, authored by Albers and Suwala, explores the link between family businesses and regional development from the spatial corporate responsibility lens in a German context. This practical view presents implications for urban and regional planning on how to perceive the role that family frms could play in regional development. Finally, in *Chapter 13*, Basco and Bartkevičiūtė present and explore the role of family frms in a place-based programme in Lithuania. The authors conclude that the local embeddedness of family members empower family frms to better utilise such place-based programme actions by reconfguring existing regional resources, developing local social capital, and embracing a sense of regional belonging.

At this point, we would like to inform readers that our editor family shrank during the process of assembling chapters and reviewing manuscripts for this collection. Sadly, our esteemed colleague Roger Stough passed away in September 2019. During his exceptional academic career, Roger not only inspired us, but many other colleagues over the last fve decades. He also produced numerous remarkable publications that later set the agenda for all disciplines. We are very honoured to have had the opportunity to work with this prestigious scholar and to form a mutual editorship with him on his probably last edited collection. Naturally, this volume is dedicated to Roger.

# **Part I Family business and regions**

# **1 Spatial familiness and family spatialities—searching for fertile ground between family business and regional studies**

*Rodrigo Basco and Lech Suwala* 

# **Introduction**

Although family business and regional studies remained unconnected until a few years ago, key thinkers in both felds reminded us early on about the value added by considering space in family frms and the embeddedness of families and their frms in territories. Indeed, just in the second issue of *Family Business Review* in 1988, referring to family frms and community culture, Joseph Astrachan emphasised that space can act as an integrative factor in the family frm's success: 'Family businesses acquired in a manner that is at odds with the local culture will suffer, while frms that are acquired and managed in harmony with the local culture will have a higher level of morale and long-run productivity' (Astrachan, 1988, 165). Walther Isard (considered the father of regional science) formulated thoughts on future directions for the discipline by saying that a general theory of human society (as a response to Masahisa Fujita) should consider

family, social group, and political decision-making and policy formulation. The optimization type of decision-making involving the family as a basic social organization and the behavior of political groups (parties) [,] which I explored in my *General Theory: Social, Political, Economic and Regional* (1969) from an economist's standpoint would need to be extended greatly to be made much more realistic.

(Isard, 1999, 388)

 Family frms are the most common form of organisation around the world, existing in different sizes, sectors, and locations (Basco & Bartkevičiūtė, 2016). Regardless of whether they were investigating gigantic multinational conglomerates in North America, the Middle East, and far-East Asia; small and medium Mittelstand family frms in Germany; or the vast number of family-based micro-businesses in Africa, researchers have observed that family frms (as legal, social, and economic entities) are characterised by family involvement in ownership, governance, and management, which in turn affects frm behaviour and performance (Basco, 2013). In the last

few decades, research in family business studies has extended beyond the aforementioned classical internal variables and has begun considering external and/or surrounding variables (Discua Cruz & Basco, 2018; Gomez-Mejia, Basco, Müller, & Gonzalez, 2020; James et al., 2020; Krueger, Bogers, Labaki, & Basco, 2020). For instance, context plays an important role when understanding the idiosyncrasies of family frms, including their economic positions (Steier, Chua, & Chrisman, 2009), the cultural imprints of society (Astrachan, 1988), their embeddedness in wider social networks (Le Breton-Miller & Miller, 2009), and their integration into institutional and political frameworks (Berrone, Cruz, Gomez-Mejia, & Larraza-Kintana, 2010). Although every economic entity is somehow situated or embedded in different contexts (Granovetter, 1985) and those different contexts (e.g., organisational, cultural, social, institutional) have been analysed by family business scholars in a variety of ways, the spatiality aspect of *context* has received little academic attention in this realm (with some exceptions, such as Seaman, 2012, 2013, 2015; Basco, 2015; Stough et al., 2015; Basco, Stough, & Suwala, 2020; Basco & Suwala, 2020).

On the other hand, in regional studies (including regional science, urban and regional economics, economic geography, urban and regional planning and management, etc.), research on frms has experienced a renaissance in the last 25 years (e.g., Dicken & Malmberg, 2001; Taylor & Asheim, 2001; Taylor & Oinas, 2006). The origins of this interest can be traced back to Robert B. McNee's (1958) seminal contribution 'Functional Geography of the Firm'. The following years were characterised by studies on the increasingly global geography of large inter- and multinational conglomerates (e.g., in the petroleum industry) (Krumme, 1969; Taylor, 1975; Dicken, 1976). The crisis of the Fordist formation in the 1970s and 1980s brought research on frms and regional decline to the fore (Hayter & Watts, 1983; Laulajainen & Stafford, 1984; Malecki, 1985), while paving the way for the resurgence and re-examination of small and medium enterprises in (mature) industrial districts of the Third Italy (Becattini, 1978). Later, frms were seen as a forge of innovation situated in new industrial spaces and technology parks with an accompanied interest in regional entrepreneurship and a new wave of high-tech activities. In this vein, scholars from regional studies also focused on specifc types of frms (e.g., new-born, small, medium, large, and foreign frms) when dismantling the role they play in regional and economic development (Scott, 1986; Giaoutzi, Nijkamp, & Storey, 1988; Sternberg, 1989; Fritsch, 1992). Then, the network paradigm took over, which dealt with the increased complexities of horizontal and vertical (dis)integration and the rise of digital frms (Yeung, 2000; Taylor & Oinas, 2006). However, the phenomenon of family frms has not been seriously considered by regional study scholars, who have mainly observed the family (business) as a supporting condition of *spatial contexts* with no systematic approach until recently (Stough et al., 2015; Suwala, 2019).

Considering the aforementioned evolution in each feld of research, this chapter takes different theoretical perspectives and searches for a common fertile conceptual ground between family frm and regional studies. To address this aim, we frst provide an overview of studies from the former, dealing with spatial entities, as well as investigations from the latter, dealing with family business issues. Second, we present two conceptual models the spatial family management model (Suwala & Oinas, 2012; Suwala, 2019) from regional studies and the regional familiness model (Basco, 2015) from family business studies—which are the seedbeds for that fertile ground between both disciplines. Finally, we juxtapose both felds according to different types of *spatial entities*.

# **Searching for fertile ground between family business studies and regional studies**

# *Family business scholars dealing with spatial entities*

Even though mainstream research in the feld has focused on the internal aspects of the family-business relationship, family business studies explored topics related to space (context) relatively early in the discipline's formation. For instance, early studies tried to invoke location or spatial distribution as a determining factor when defning family frms. Westhead and Cowling (1998), in a bivariate analysis, found that family frms are over-represented in rural locations and under-represented in resource-rich core regions in the United Kingdom (Westhead & Cowling, 1998, 54).

Probably, the frst attempt to deal with spatial entities, or more precisely *spatial factors*, was Kahn and Henderson's (1992) work investigating family frms and locational factors. The general idea underlying this study is that family frms have to balance both the family and business perspectives in strategic decision-making, such as when choosing their location (see also Hollander & Elman, 1988). In a sample comparing family frms and nonfamily frms, Kahn and Henderson (1992) found only mixed support for the assumption that family frms prefer soft locational factors (e.g., quality of life and amenities, among others) in relation to non-family frms, highlighting the importance of rational hard locational factors in frms' decision-making (e.g., proximity to markets and facility costs among others). However, they found that family frms are more concerned with spatial proximity to their residence than non-family frms (Kahn & Henderson, 1992). In general, family ownership seems to affect business decisions; whether this infuence is adverse or benefcial will require further investigation (Scranton, 1993; Getz & Petersen, 2004; Ingstrup, Jensen, & Lüthje, 2016). Today, this research stream mostly investigates locational factors for family businesses in specifc countries (e.g., Heinemann et al., 2019).

Apart from family frms' locational preferences, some studies have focused on the *spatial structure and distribution* of family frms and their impact on local and regional economies (Pérez & Raposo, 2007; Spiegel & Block, 2011; Scholes, Wilson, Wright, & Noke, 2012). Exploring West Germany at the district level (NUTS-3), Spiegel and Block (2011) emphasised that family frms tend to be located in rural areas in proximity to regional metropoles. They found that North Rhine-Westphalia and Baden-Württemberg (two economic powerhouses in Germany) have the highest densities of family frms, confrming the assumption that they are located in economically strong regions (Spiegel & Block, 2011). Taking the spatial distribution of 245 large Spanish family frms into account, Pérez and Raposo (2007) confrmed the hypothesis that family frms are located in the most important economic centres (traditional industrial areas) with headquarters in Catalonia, Madrid, Aragon, the Basque Country, and Valencia.

The third and probably most popular research stream up to today focuses on family frms' internationalisation, which could be considered a *spatial processes*. Starting with Gallo and Sveen's (1991) seminal study on factors hindering and favouring this process, the topic of internationalisation has become popular among family business scholars. Most studies have focused on either searching for key determinants or ideal pathways (process view) to explain it (e.g., Gallo & Pont, 1996; Okoroafo, 1999; Graves & Thomas, 2008) or analysing the infuence of family resources, ownership, heterogeneity, and networks (Calabrò, Campopiano, Basco, & Pukall, 2017) (for a detailed literature overview, see Kontinen & Ojala, 2010; Pukall & Calabrò, 2014, and for a review of theoretical frameworks, see Kraus et al., 2016; Reuber, 2016; Arregle, Duran, Hitt, & Van Essen, 2017) on family frms' internationalisation.

The fourth research stream, which is a bit far-fung, focuses on family frms across various *spatial contexts*. Studies on spatial contexts encompass different perspectives by considering emerging markets (Basco, 2018; Suwala, Kulke, & Gade, 2018; Rienda, Claver, Quer, & Andreu, 2019), transition economies (e.g., Donckels & Lambrecht, 1999; Banalieva, Eddleston, & Zellweger, 2014; Stangej & Basco, 2017), home regions (Banalieva & Eddleston, 2011; Pongelli, Calabrò, & Basco, 2019), urban versus rural locations (Backman & Palmberg, 2015; Baù et al., 2019), specifc countries (e.g., Kowalewski, Talavera, & Stetsyuk, 2010; Daszkiewicz & Wach, 2014), and specifc regions (e.g., Müller, Botero, Cruz, & Subramanian, 2018). The main rationale behind these investigations is that context can be related to family frms' wellbeing, functional logic, and success factors. Context leads to idiosyncratic practices. Therefore, it is not surprising that family frms are often tied to a specifc spatial entity, such as their home region (e.g., Hennart, Majocchi, & Forlani, 2019). For example, Donckels and Lambrecht (1999) found that the re-emergence of family frms in East Central Europe in the 1990s was mostly driven by frms' implementation of crucial management functions and their growth aspirations. In other words, these studies account for spatial varieties of family frms.

The ffth research stream highlights the *spatial scales* of family frms, such as local (Seaman, McQuaid, & Pearson, 2017; Baù et al., 2019), regional (Chang, Chrisman, Chua, & Kellermanns, 2008; Bird & Wennberg, 2014), and global (De Massis, Frattini, Majocchi, & Piscitello, 2018). Most newer studies have considered these different scales not solely as passive surroundings but rather as active spatial frames that provide unique embeddings enabling family frms to thrive. The recursive relationship between family frms and location can have many dimensions: local embeddedness; family corporate local responsibility; family frm-led place leadership (Albers & Suwala, 2020a–c; Graffenberger & Görmar, 2020); family frm-led regional economic growth and development (Basco, 2015); and family frm-led local persistence, resilience, and sustainably (Brewton, Danes, Stafford, & Haynes, 2010; Ljungkvist & Boers, 2016). Therefore, local (i.e., socio-spatial) embeddedness is a particularly important feature of family frms (Basco, 2018; Selcuk & Suwala, 2020). In this realm, Bird and Wennberg (2014, 424) conclude that 'family businesses are more embedded within the regional community than their non-family counterparts'.

The sixth research stream deals with family frms' *spatial settings*. This refers to the spatial confgurations that make places unique. For instance, the territorial innovation models (i.e., industrial districts, local networks, clusters, or regional innovation systems) (e.g., Block & Spiegel, 2013; Cucculelli & Storai, 2015; Lopes, Branco, Parejo, & Rangel, 2016; Basco & Calabrò, 2016; Seaman et al., 2017; Konsti-Laakso et al., 2019; Amato, Basco, Backman, & Lattanzi, 2020). All territorial innovation models 'emphasize the importance of spatial proximity … in generating production, knowledge, learning processes and/or innovation in certain locales' (Brinkhoff, Suwala, & Kulke, 2015, 129). With regard to industrial districts in the Italian manufacturing industry, Cucculelli and Storai (2015) concluded that the family effect and the district effect both act as substitutes and complements depending on the size of Italian manufacturing frms. Block and Spiegel (2013) investigated the effect of family frms and regional innovation systems, fnding evidence that the higher the density of family frms, the higher the level of innovation outputs, which in turn enhances regional patent outputs. Apart from that, Seaman et al. (2017) showed that family-owned frms make considerable contributions towards local economies through manifold linkages manifested via family, business, and friendship networks.

The seventh research stream comprises thoughts about *spatial concepts*. Seaman (2012 and 2013) developed an interesting framework of four types of spaces for business development that combines three types of inner space and one type of outer. Conceptual space refers to a cognitive superstructure that includes the idea of the family frm as a desirable activity as well as initial ideas supporting business development. Cultural space refers to the space created within the family by social and fnancial capital. Contingency space is about the help the family provides to the frm during the start-up phase, including both hands-on assistance and a pool of individuals with tacit knowledge and emotional commitment to the business. As a result of these different types of spaces, families and businesses create idiosyncratic knowledge that is deeply interwoven and mediated through networks in an additional space—called community space—and thus facilitates (regional) development.

Additionally, different topics related to *spatial policies and planning* have been investigated with less research intensity (Glassop & Graves, 2010; Basco & Bartkevičiūtė, 2016). Even though family frms play an important role in regional and national economies, studies on regional policy have generally only focused on policies aimed at supporting family business, such as tax benefts or advice about ownership and management succession (Basco & Bartkevičiūtė, 2016). This poor state of scientifc studies exists despite policymakers' knowledge of family frms' idiosyncratic needs and challenges in terms of intergenerational business transfer, fnancial obligations, business-family balance, lack of specifc education, access to fnance for growth, and ability to maintain a skilled workforce (Glassop & Graves, 2010). Ricotta, Golikova, and Kuznetsov (2017) found no notable differences affecting the innovative performance of family frms *versus*  non-family frms in seven European countries and Russia despite different development phases of the economy and institutional environments among countries (as a proxy for spatial policies). For South Asia, Samphantharak (2019) reported that ownership and control are concentrated among a handful of prominent families that have formed business groups. These family business groups maintain extensive connections with politicians and bureaucrats and therefore indirectly exert power over relevant local and regional policies in favour of family businesses (e.g., 'guanxi capitalism' and 'bamboo networks') (Samphantharak, 2019).

### *Regional studies scholars dealing with family businesses*

Although regional studies scholars emphasised early on that the family is a crucial factor when an individual frm's locational choice is at stake (Townroe, 1969), the phenomenon of the family frm was mostly disconnected from spatial scales and from any internal and idiosyncratic spatialities (Taylor & Asheim, 2001). Since space (and time) are abstract variables that dominate the discourse in regional studies, it is not surprising that the family has been a less important and often neglected dimension. The contributions of regional studies scholars in relation to family frms are scarce and fragmented in term of *zeitgeist*, research methods, theoretical concepts, and policy interventions.

The frst research stream we can highlight, probably the strongest attempt to broach the subject of the family in regional studies, was the (re-)discovery of the Marshallian industrial districts (MIDs)—a particular *spatial setting* (Bagnasco, 1977; Becattini, 1978). Industrial districts are 'clusters of small family- and craft-based frms in the rural areas of the Third Italy … with petit bourgeois traditions, community-wide social and economic rules, and municipal mercantilist traditions, which are historically sedimented in particular areas' (Trigilia, 1990, 199). These family frm–based spatial settings were viewed as a new regional remedy in the Fordist crisis of the 1970s and 1980s (Paci, 1980; Pyke, Becattini, & Sengenberger, 1990). In this sense, the link between economic spaces and family frms emerges when one considers that 'the spatial integration of productive and reproductive spheres also permits artisans to rely on the casual labour of family members, particularly women and pensioner[s]' (Lazerson, 1990, 121). Family in business is the crucial argument for family-based spatial settings, with Marshallian theory initially being extended to analyse the successful performance of regionally concentrated systems of production based on family-owned and highly specialised small and medium-sized enterprises. Moreover, industrial districts orchestrate market mechanisms and social institutions, such as families, kinship networks, and local communities, capitalising on external economies of scale (Trigilia, 1986). However, the family as an important ingredient in spatial settings was picked up randomly in studies investigating the spatial organisation patterns of post-Fordist formation in relation to new business formation (Garofoli, 1994) and the advantages of small and medium-sized enterprises (Bryson, Wood, & Keeble, 1993) due to cooperation, trust, and reciprocity among spatially proximate economic entities (Hansen, 1992).

Apart from the Italian industrial district, the Wenzhou model (Yeung, 2000; Wei et al., 2007) and the Gnösjo Spirit (Wigren, 2003) became popular research frameworks for exploring family frm–driven regional development, wherein business, family, and spatial context were heavily intertwined. In these frameworks, family frms or kinship ties serve as the 'glue' for socio-spatial proximity, holding together spatial arrangements and localized social relationships in socalled territorial innovation models (innovative milieus, clusters, regional innovation systems, and new industrial spaces) (Sforzi, 1989; Pypłacz, 2013; Brinkhoff et al., 2015). Therefore, in regional studies, the frm is frequently considered 'an organisational unit bringing together diverse social relations in which actors are embedded … [and] these relationships may be inter-personal relationships, family linkages or simply social ties' (Yeung, 2000, 311). In other words, the frm and its (family) management orchestrate spatial locations and networks, all of which are stabilised by family ties (Suwala & Oinas, 2012; Suwala, 2019).

A second research stream considers the family and family frms as (un-)favourable *spatial factors or structures* in local decisions. While Malecki (2000) subsumed the family as a soft factor in regional science, Zhou (1996) and Suwala and Kitzmann (2019) emphasised the benefts of spatially-concentrated ethnic and migrant family frms. Moreover, research has shown that founders/owners tend to locate near their families and friends in general (Schamp, 2005; Stam, 2007; Martyniuk & Gierusz, 2016) as these soft locational factors are important for frm performance and survival (Martyniuk-Pęczek, Martyniuk, Gierusz, & Pęczek, 2017; Suwala, 2019). Based on a study of 251 Polish frms, Martyniuk and Gierusz (2016) confrmed that in the case of family business, when deciding on the location for their business activity, the majority of owners choose a location near their home/residence.

The increased availability and accessibility of family business-specifc and regionally dis-aggregated data has enabled new research on the *spatial structures* of family-based economic activities and their recursive relationship with the context (e.g., Adjei et al., 2016; Berlemann & Jahn, 2016; Majocchi, D'Angelo, Forlani, & Buck, 2018; Amato, 2019). Amato (2019) provided a fne-grained look into the generally positive association between family-managed frms and employment growth within a large panel dataset of Spanish manufacturing frms during the economic recession between 2007 and 2008. He concluded that when location is considered, municipality size positively affects employment growth within family-managed frms. Moreover, the results reveal that during the economic crisis, the reduction in employment level was less pronounced for family-managed frms located in small municipalities due to their stronger socioeconomic ties than for those in larger urban settings (Amato, 2019). The rationale here is that spatial structures make a difference for family frms.

A third research stream takes the availability of new data to explore the phenomenon of family frm at different *spatial scales*, ranging from local (e.g., Yanagisako, 2002; Baschieri, Carosi, & Mengoli, 2017) to global (e.g., Yeung & Soh, 2000; Majocchi et al., 2018). Interestingly, the widely acknowledged local home bias of family frms could not be confrmed in the study by Baschieri et al. (2017). Using a dataset of Italian frms (half of them family frms) over the 1999–2011 period, they concluded that local home bias was not a common phenomenon among the entire sample and mainly occurred in family frms in which the founder led the business. Local home bias did not occur in non-family frms and in family frms in which the owner had acquired control through a market transaction. Moreover, the results suggest that locally committed family frms elicit investor preference for local stocks and, in doing so, exploit local clientele to lower the cost of funding. This means that family frms' social contributions to their local communities may have an opportunistic effect (Baschieri et al., 2017).

A fourth research stream links family frms to *spatial processes* such as regional learning, innovation, and internationalisation (Yeung, 2000; Wei et al., 2007; Berlemann & Jahn, 2016; Adjei, Eriksson, Lindgren, & Holm, 2019; Amato, 2019). This stream mostly investigates different dimensions of social capital and proximities that infuence frm performance and thus regional prosperity (Karakayaci, 2013; Pucci, Brumana, Minola, & Zanni, 2017; Suwala & Micek, 2018). For example, studies have shown that the family and different forms of entrepreneurial family relationships can be considered sources of effective learning or skills and thereby contribute to differences in frm performance across regions (Adjei et al., 2016, 2019). With regard to innovation, for Italy, Pini (2019) showcased that while external management signifcantly affects frms' propensity for innovation in the more advanced area (centre-north), in the less developed area (south), it requires an additional simultaneous investment in research and development to drive frms' innovation in family management. However, it is unclear whether a strong presence of family frms in certain spatial settings always leads to prosperity. Schamp (2005) concluded that the lack of fexibility in regions dominated by traditional family branches, which often have a high density of family frms, might result in regional cognitive lock-in, regional sclerosis, and/or regional decline.

A ffth research stream comprises *spatial policy and planning* approaches, where family frms have been considered complementary additional actors in the development of local and regional economies (Albers & Suwala, 2018, 2020a–c; Graffenberger & Görmar, 2020). Albers and Suwala (2018 and 2020) differentiated between different intensities of so-called 'corporate spatial responsibility'—a spatial extension of the traditional concept of corporate social responsibility—and the impact on regional development. As shown in some cases, family frms' local commitment became so intensive that they coined these family frm initiatives as family frm–led local and regional economic development. In these cases, family frms took over former sovereign and administrative districts as well as county and state tasks necessary for local and regional governance and exerted genuine place leadership through holistic local future plans and master plan initiatives (Albers & Suwala, 2018, 2020c; Suwala et al., 2018).

# **Spatial familiness and family spatialities**

The contemporary re-discovery of space in family business research studies and families within territories have connected family business studies with regional studies. Not until recently did both these research streams start the cross-fertilisation process. While scholars like Basco (2015) and Stough et al. (2015) from family business studies raised the question of whether the mere presence of family frms is good or bad for regional economic development, Suwala and Oinas (2012) from the regional studies feld almost simultaneously examined the micro-foundations of the spatial management of (family) frms.

To explore each of the aforementioned streams, we frst present the conceptual model of spatial family management, which aims to interpret how different spaces affect family management (Suwala & Oinas, 2012). Second, we re-visit the regional familiness model (Basco, 2015), which aims to capture family frms' embeddedness in social, economic, and productive structures within the spatial context to explain family frms' role in regional development. For both approaches, we shed light on the spatial entities we discussed in the previous sections—namely, factors and structures, processes, contexts, scales, settings, policies, and concepts.

### *Spatial family management model*

Despite being a central topic in regional studies, neither the frm nor the manager have been properly acknowledged, due to the rather macro perspective used to develop models of spatial economies (e.g., von Thünen, Weber, Christaller, Lösch, Isard). Moreover, these models assume a passive interpretation of the individual as *homo economicus* (Suwala, 2020). Despite novel and promising developments in the feld, the (family) manager is still considered rather lonesome or, more precisely, a fragmented maverick (Suwala & Schlunze, 2019). The management geography1 research stream (Schlunze, Baber, & Agola, 2012; Suwala & Oinas, 2012) attempts to explain managerial agency across and within spaces when businesses have to decide on appropriate locations, local or international operations, strategic relationships with suppliers and customers, and internal and external images. In other words, this research stream aims to analyse the management of economic, social, and cognitive spatial domains in multi-scalar confgurations and their infuence on corporate performance through concentration, interaction, and perception (Suwala & Oinas, 2012). Family managers have to address three spatialities—economic, social, and cognitive spaces spatialy-infuenced self-reinforcing (dis-) and each of them can be seen as a continuum between different types of economies versus diseconomies of scale i.e., spatialy-infuenced self-reinforcing (dis-) economies (Suwala, 2014).

First, the economic space can be thought of as a location. Space retains economic meaning as a location due to the simple fact that transport and transaction expenses occur over distance; hence, location is relative (against other locations), and spatial costs (e.g., transaction and transportation costs) can be calculated (Suwala, 2020). Within this domain, the family manager has to address different problems related to location, such as the frm's locational strategy, the optimal spatial choice regarding locational factors, and internationalisation-related decision making (e.g., liabilities of foreignness) (Suwala & Kulke, 2017). Managerial tasks have to optimise self-reinforcing

 *Figure 1.1* Spatial family management model (adopted and refned further from Suwala and Oinas [2012], Suwala [2019]).

spatial mechanisms arising from agglomeration economies that are internal (e.g., economies of scale and scope) and external (e.g., localisation and urbanisation economies) to the frm. Here, *spatial factors* (importance of locational factors), *spatial structures* (relative location of the business), and partly *spatial processes* (expansion and internationalisation of the frm) are important. Managers have to optimally balance the concentration and dispersion of economic activities within space (Suwala, 2019).

Second, the social space is perceived as a place. A place is constructed through social ego-centric relationships between individuals; hence, it is relational. In other words, 'social space involves the network of functional relationships and social interactions' (Trip & Romein, 2010, 5). Places are about context, and they can only be understood through actors' social and relational embeddedness (Selcuk & Suwala, 2020). Within this domain, the family manager has to balance family relationships, internal and external frm ties (suppliers, customers, etc.), and friendship networks (Suwala, 2019) to provide solutions to the fundamental problem of coordinating relationships between economic actors. These relationships are characterised by different dimensions of proximities (e.g., spatial, cognitive, organisational, institutional, and cultural) (Knoben & Oerlemans, 2006). Proximities may become proximity economies when fuelled by power, trust, and reciprocity, resulting in place-based knowledge and learning processes (by means of face-to-face meetings) (Storper & Venables, 2004). However, not only is being there (spatial proximity, physical presence) important but so is being aware (cognitive proximity) (Grabher et al., 2018). Managers have to orchestrate the frm's networks its interaction and isolation while being situated in particular *spatial contexts* and *spatial settings* (e.g., industrial districts, migrant communities) (Suwala, 2019).

 Finally, the cognitive space can be visualised as a landscape. Landscapes are the result of individuals' topo-centric relationships with space. Individuals (e.g., employees) are bound together not only by (ego-)relationships among themselves (relational view) but also through topical-relationships directly with space (Suwala, 2020). The result is a topical feld in which individuals collectively share similar cognitive images, spatial mindsets, and particular atmospheres without necessarily knowing each other personally (Brinkhoff et al., 2015). Within this domain, the family manager lays out frm ethics or a family charter in order to establish a long-lasting frm culture that is reinforced by manifold images related to the family name, house symbols, and logos and in which the environment or workplace design plays a crucial role. This cognitive space is about a 'stockpile of knowledge, traditions, memories and images' (Scott, 2010, 123), which helps create a (mutual) atmosphere to stimulate particular perceptions and holistic experiences. The cognitive space as landscape may result in experience economies (e.g., entertainment, educational, aesthetic, and escapist experiences) (Suwala, 2014) depending on the intensity of attention and the memorability of extraordinary events and may arise from overlapping perceptions among individuals (Lange, Power, & Suwala, 2014; Pfeufer & Suwala, 2020). Therefore, managers have to set up and balance between perceptions and misconceptions of economic actors' experiences with family frms creating a certain *spatial context—spatial processes* like local learning, creativity, and innovation.

Extending these ideas from management geography to the family business feld, managerial decisions in the context of family frms have to balance three different spatialities—economic, social, and cognitive spaces—and each can be interpreted as a continuum between different types of (dis-) economies arising from concentration/dispersion, interaction/isolation, and perception/misconception. Each of these spatialities comprises different managerial roles, such as the concentration role as a locational explorer (relative view, economic space, concentration), the interactional role as an embedded gatekeeper (relational view, social space, interaction), and the informational role as an experienced preceptor (topic view, cognitive space, perception) (Figure 1.1).

### *Regional familiness model*

Since family involvement in the frm affects the way an organisation is owned, governed, and managed, consequently, it is a source of heterogeneity among family frms and between them and non-family frms. The microfoundations of the spatial family management model should be linked to the meso- and macro-foundations in order to theoretically and empirically interpret the effect of family frms at different *spatial scales* (e.g., local or regional). In this sense, the regional familiness model aims to connect the meso- and macro-foundations of family frms in regional studies.

Following Basco (2015), we defne regional familiness as the consequences of family businesses' embeddedness in the *spatial context* that alter regional or *spatial factors* (i.e., tangible and intangible factors) and regional or *spatial processes* (e.g., spill-overs, information exchange, learning processes, social interactions, competition dynamics, and institutional dynamics) through proximity dimensions (i.e., relational, institutional, organisational, social, and cognitive proximities). In this sense, the family frm is not only an important actor due to its mere presence but also a driver of heterogeneity for regional factors and processes and their associated proximity dimensions.

The frst connection that the regional familiness model proposes is the effect of family frms on regional factors (*spatial factors)*. Regional or *spatial factors* are the aggregate resources (tangible or intangible, endogenous or exogenous) in an adopted spatial entity. They include not only traditional neoclassical resources related to capital and labour but also human factors (knowledge embedded in the labour force), social factors (networks and access to networks through which information fows), and entrepreneurial factors (ability and willingness to discover and exploit opportunities). The distinctive characteristics of family involvement in economic activities develop and create aggregate factors that, because of their endogenous characteristics, represent basic elements that may determine the quantity, quality, and pace of economic and social development. It is important to recognise that the aggregate effect of family frm activities could alter existing regional factors, imprinting them with specifc properties, such as patient capital, local re-investment, and long-term commitment, but could also create additional unique and diffcult-to-imitate regional factors, such as labour commitment, collective knowledge transfer across generations, stock of kinship and social relationships, and collective entrepreneurial spirit (see the literature on territorial innovation models—Brinkhoff et al., 2015).

However, understanding the effect of family frms on regional factors is necessary to unpack family frms' connection to the meso-level; regional factors need regional processes that are responsible for exploiting and allocating them. The main regional or *spatial processes* include spillovers, information exchange, learning processes, social interactions, competitive dynamics, and institutional dynamics. The quality of regional processes can accelerate or slow the productivity of regional factors and can thus have consequences for regional economic and social development. By recognising the existence of regional processes, it is possible to move the concept of space from an absolute and relative perspective to a relational perspective.

The effcient functioning of regional processes lies on the dimension of proximity. Proximity refers to the state, quality, sense, or fact of being near or next to in space, time, or relationship (Torre & Wallet, 2014; Basco, 2015). However, just 'being there' is not suffcient for regional processes to function effectively. According to Boschma (2005), proximity is more than simple geographical proximity related to the physical distance between economic actors and between economic actors and regional factors. The concept of proximity has different dimensions. Cognitive proximity refers to 'the similarity of the subjective mental framework of actors and the tacit and codifed knowledge owned by actors' (Westlund & Adam, 2010, 112). Social proximity can be defned as the socially embedded relationships among agents based on trust derived from friendship, kinship, and experience (Brinkhoff et al., 2012). Finally, while organisational proximity refers to the individual relationships within the boundaries of an organisation itself and the relationships among organisations, institutional proximity is the general macro-level (political) framework.

Because family businesses are locally embedded and have historical roots in certain places, the regional familiness model proposes that the thickness and quality of proximity—in other words, the *spatial context*—are affected by the aggregate effect of family businesses. For instance, family frms generally intend to stay where they dwell (belong) even during diffcult times like crises (Zhou, He, & Wang, 2017). In this sense, family frms stabilise geographical proximity and therefore *spatial structures* across generations. The most promising and necessary aggregate effect of family frms is on cognitive and social proximity due to the intrinsic relationship between the family, the frm, and the local community (see already Astrachan,

1988; Brinkhoff et al., 2012; Seaman et al., 2017). This relationship is pronounced in family frms since they are active actors in the regional socialisation process, in which economic, social, and emotional connections are not only generated between the family and the frm but also extended to the rest of the local community, creating a 'local atmosphere' (cf. original Marshallian ideas, Marshall, 1890, 198). Additionally, the kinship and friendship relationships within and beyond frm boundaries contribute to establishing a particular trust-based society (see also Paci, 1980). In this sense, the organisational nature of family frms stimulates organisational proximity by developing communication channels among frms (within and outside the region) and by establishing cooperation and competition. Finally, to a certain extent, family frms are responsible for developing institutional proximity—that is, the implicit and explicit values, cultural norms, ethical principles, and formal rules that frame local and regional economic activities.

 In sum, the regional familiness model is a box of theoretical tools for analysing and interpreting the role family frms can play in regional development. It goes beyond the simple statement that the importance of family frms stems from the fact that the family business is the most common form of organisation and unpacks the meaning of their presence in economic space. Moreover, the model links *spatial factors*, *spatial contexts*, and *spatial processes* (consequences) and enables researchers to operate across various *spatial scales* (from meso- to macro-foundations) (see Figure 1.2).

*Figure 1.2* Regional familiness model (Basco, 2015).

# **Concluding remarks for fertile ground between family business studies and regional studies**

 The spatial family management and regional familiness models can be combined to create a big picture that links all *spatial scales* (micro-, meso-, and macro-levels of analysis) and connects family business studies and regional studies. While the spatial family management model incorporates spatialities into family management decision making (family spatialities), the regional familiness model unpacks the role family frms play in regional economic and social development. What a combined model might look like is subject to further research in the near future (see also Basco & Suwala, 2020). Table 1.1 summarises the main outcomes of the stock-taking studies in both disciplines and their relationships with the analysed spatial entities *factors*, *structures*, *processes*, *contexts*, *scales*, *settings*, *policies*, and *concepts*.


*Table 1.1* Key fndings linking the felds of family business studies and regional studies

(*Continued*)

#### *Table 1.1* Continued


Concerning the research on *spatial factors*, we see a great deal of coherence in the results between both disciplines. The main idea is that family frm owners tend to locate near their families and friends and favour soft locational factors in comparison to non-family frm owners, who tend to make locational decisions based on rational cost-beneft considerations. In addition, family business research reveals the complicated relationship between the family's perspective and the business's perspective in locational choices. With regard to research on *spatial structures and distribution*, there is also an overlap between both streams. Family frms and their accompanying business functions are generally located inside founding regions, which indicates the spatial persistence of family frms. Moreover, family frms often provide both stability and breeding grounds for prosperous economic and social development. All of this happens not only in urban centres but also in rural areas in proximity to metropoles and outside of high-tech agglomerations (e.g., Germany's so-called 'hidden champion' frms). Furthermore, *spatial scales* are perceived as active surroundings in both disciplines rather than passive frames. Although unlimited in variety depending on the research interest at hand, home region (i.e., local region) bias still dominates the research, albeit with declining intensity. *Spatial contexts* still act as a lens to investigate family frms' wellbeing, functional logic, and success factors. This selection mechanism makes spaces and family frm practices unique. There is no 'one location fts all' approach, as both disciplines agree, because family frms, their locational imprints, their people, their practices, and their images are embedded in particular spaces. Here, research interests in both disciplines differ. Whereas regional studies scholars consider family frms and family ties the 'glue' necessary for socio-spatial proximity in territorial innovation models, there is mixed evidence within family business studies concerning the 'family effect' and the 'spatial settings effect' in networks, districts, and clusters.

There are also pertinent research approaches with regard to *spatial processes*. Whereas family business scholars mostly deal with the capabilities and pathways necessary for family frms to internationalise, regional studies scholars focus on regional learning and innovation processes. Research on *spatial policies* and *planning measures* is predominantly in the early stages in both disciplines, with various approaches treating family frms as both targets and initiators of policy interventions. In terms of *spatial concepts*, research has focused on both internal and external aspects: 'spaces of familiness' and 'family spatialities'.

Finally, the two presented models bridge multiple spatial entities: the regional familiness model, which outlines the spatial factors, contexts, and processes (consequences) necessary to enhance regional development by family frms across different spatial scales, and the spatial family management model, which suggests multi-spatial coordination tasks incorporating spatial factors, structures, contexts, settings, and processes into family frm managers' decision processes. Both models complement each other quite well since the latter describes the micro-foundations of managerial decision making that can be scaled upward through aggregated views by the former. Both research streams have overlaps but also idiosyncrasies originating from their felds. Therefore, we call for more interdisciplinary work to address research gaps and exchange insights on theoretical, empirical, and practical grounds to better investigate the phenomena of 'spatial familiness' and 'family spatialities'.

### **Note**

1 Sometimes also called 'managerial geography' (Laulajainen, 1998) or 'economic geographies of management' (Jones, 2016).

# **References**






# **2 A regional perspective of family frms**

Evidence from Europe

*Rodrigo Basco and Fernanda Ricotta* 

### **Introduction**

One important characteristic of capitalism, as an economic and political system, is the fact that its productive structure is formed by a large number of private owners who embrace economic activities through small, medium-sized, and large frms. However, there are two questions that require further investigation: (1) who are these owners and (2) how are private owners distributed across geographical space? To a certain extent, capitalism has been recognised as family capitalism because of the high participation of families in businesses, with family businesses representing the most common form of organisation in developed (Adjei, Eriksson, Lindgren, & Holm, 2019) and emerging economies (Basco, 2018). Family frms dominate national productive structures (Astrachan & Shanker, 2003; Bjuggren, Johansson, & Sjögren, 2011) in today's societies. Indeed, they exist in all size categories from micro to large frms (Basco & Bartkeviciute, 2016), and families own a substantial portion of listed frms around the world (Faccio & Lang, 2002; La Porta, Lopez-De-Silanes, & Shleifer, 1999; Martinez-Garcia, Boubakri, Gomez-Anson, & Basco, 2020). These stylised facts regarding the prevalence of family frms across national contexts have been subjected to an intensive economic history debate (Berghoff, 2006; Burkart, Panunzi, & Shleifer, 2003; Chandler, 1990) about the benefts and drawbacks of having them embedded in national and regional productive structures.

However, the debate about the prevalence and importance of family frms at the regional level is still in its infancy. The empirical evidence about the regional importance of family frms is contradicting (Stough et al., 2015). Whereas family frms are located in less developed regions in the United States (Chang, Chrisman, Chua, & Kellermanns, 2008) and in countries with high level of economic entrenchment (Morck & Yeung, 2004), they are important players in Germany and Italy, usually considered the most industrial countries in Europe (Arrighetti & Ninni, 2012). For instance, the Italian industrial sector is characterised by a large presence of family frms and, in the German context, small and medium-sized frms (i.e., Mittelstand), most of which are family owned and managed, are the backbone of the regional economy, employing 60% of all employees subject to social security contributions (BMWi, 2013).

### 34 *Rodrigo Basco and Fernanda Ricotta*

To continue with the abovementioned debate, this chapter focuses on exploring the prevalence of family frms across European regions. Our fndings present a general descriptive picture of their prevalence in industrial productive structures across seven European countries (Austria, France, Germany, Hungary, Italy, Spain, and the United Kingdom). Additionally, we analyse the two-way relationship between family frms and regional context, and while we do not fnd full support for the argument that family frms are overrepresented in regions with low-quality institutions, we fnd that the aggregate relationship between family frms and regional competitiveness (i.e., productivity, exports, and innovation) varies across regions within each analysed country.

Our chapter contributes to the endeavour to link the felds of family business and regional studies. In doing so, we attempt to analyse family frms across regions by mapping the prevalence of family frms in European regions. We fnd that even though family frms are highly prevalent in European regions, they are unevenly distributed. This evidence reveals the importance of further exploring the connection between family frms and the territories in which they dwell. While the family business feld has traditionally focused on the relationship between the family and the frm and has failed to contextualise the family business phenomenon (i.e., historical, geographical, and institutional contexts) (Gomez-Mejia, Basco, Müller, & Gonzalez, 2020), regional studies research has focused on the relationship between the frm and the region and has omitted frms' heterogeneity by considering the specifcities of family frms. Therefore, our chapter attempts to recognise family frms as economic actors and to link them to their regional context by presenting empirical evidence that could further motivate new research in the felds of family business and regional studies.

### **Family frms and European regions**

In this section, we describe the prevalence of family frms in the manufacturing sector across seven European countries: Austria, France, Germany, Hungary, Italy, Spain, and the United Kingdom. We use the EFIGE (European Firms in a Global Economy) dataset, which is a by-product of the European Union project called 'European Firms in a Global Economy: Internal Policies for External Competitiveness'. This dataset contains data from a survey carried out in 2010 that provides comparable cross-country data on approximately 15,000 manufacturing frms in the seven European countries mentioned above. The information in the survey mostly refers to the three-year period of 2007 to 2009 or, in some cases, to 2008 (for a detailed description, see Altomonte and Aquilante [2012]). Survey samples were constructed based on the Bureau van Dijk AMADEUS database. The sampling design was structured with three strata: industry (11 NACE-CLIO industry codes),1 region (at the NUTS 1 level of aggregation), and size class (10–19, 20–49, 50–250, more than 250 employees).<sup>2</sup> Therefore, the EFIGE database is a frm-level dataset of representative samples of manufacturing frms (with a lower threshold of 10 employees) across the aforementioned seven European countries (for more information, see Altomonte and Aquilante [2012]).3

Our exploratory research focuses on family frm prevalence across 55 NUTS 1 European regions.4 The criterion to distinguish family frms from non-family frms is based on respondents' self-perceptions of whether their businesses are family frms (this specifc self-perception option has been used in other studies, such as Pongelli, Calabrò, and Basco [2018]). To dig deeper into the analysis, we frst provide a descriptive picture of the data by considering the prevalence of family frms in terms of demographic frm characteristics (e.g., frm size and frm age) and family characteristics (e.g., family involvement in ownership and management). Second, we attempt to relate the distribution of family frms across regions by considering the quality of regional institutions. Finally, we close the section by exploring and analysing the prevalence of family frms and its relationship with regional competitiveness (i.e., regional productivity, regional export, and regional innovation).

### *Descriptive analysis of family frms across European regions*

### *Prevalence of family frm in the national and regional context*

We begin our exploratory analysis by describing the importance of manufacturing family frms in Austria, France, Germany, Hungary, Italy, Spain, and the United Kingdom. Table 2.1 shows their prevalence in the manufacturing productive structures of these seven European countries. This evidence indicates that the prevalence of family frms is dispersed across national contexts. For instance, the lowest shares of manufacturing family frms are in Hungary (56%) and France (58%), while the highest shares are in Germany (84%) and Austria (82%). Our frst conclusion is that even though family frms are the most important economic actors in European manufacturing productive structures, they are unevenly distributed across European countries. The national disparity in this prevalence could be related to the context in which family frms are born, compete, and die. Future studies


*Table 2.1* Percentage of family frms at the national level—manufacturing industry

should further investigate the connection between them and contextual dimensions (for more information, see Basco & Suwala, 2020; Gomez-Mejia et al., 2020; James et al., 2020; Krueger, Bogers, Labaki, & Basco, 2020). Beyond the multiple dimensionality of context and level of analysis, we wonder if this uneven distribution persists within each country across regions.

To show the regional distribution of family frms within each of the seven European countries, Figure 2.1 illustrates the regional share of family frms across regions, evidencing that they are not only unevenly distributed across countries but are also unevenly distributed across regions within national

*Figure 2.1* Regional share of family frms (under/upper median).



geographical contexts. For a better interpretation, Table 2.2 contains information about the maximum and minimum values of the regional share of family frms. For instance, the United Kingdom's share of family frms is 64%, but while the lowest regional share is a little more than 50% in some regions, the highest is almost 80%.

#### *Prevalence of family frms across industrial sectors*

Taking a closer look at the distribution of family frms across industrial sectors, Table 2.3 shows the share of family frms in terms of the analysed European countries and 11 sub-industrial sectors. When taking into account the whole sample, they are highly represented in the sub-industrial sector of 'wood and wood products' but are represented less in the 'coke, refned petroleum products, and nuclear fuel' sector. To explain the differences, we speculate that each country has different competitive advantages in specifc industries that attract/ favour family frms. These differences could also be a consequence of specifc regional path-dependent processes that cause family frms to have a higher presence in some sectors than others. Finally, the presence of family frms in specifc industrial sectors could be a consequence of public policies favouring local business families (political connections). For instance, the share of family frms in the 'chemical and chemical products' sector in Germany—one of the largest in the country, with a long-lasting tradition of families in business—is higher than in France or Hungary. This empirical evidence challenges further research to better understand how national competitive advantages, regional path dependence, and active public policies are linked to the regional prevalence of family frms. Even more, future research should explore how the regional distribution of family frms within each country is linked to particular geographical/territorial characteristics and industries.

Looking at the industrial sectors from a different angle, Table 2.4 shows the share of family frms in terms of the Pavitt sectors by country. When taking into consideration the whole sample, family frms are highly prevalent in traditional industries, as expected, because of the competitive advantages that families may bring to their frms, such as patient capital, family human capital, and social capital, among others. Additionally, the presence of family frms in specialised sectors is also high. On the other hand, the prevalence drops in economies-of-scale sectors, in which family frms seem to have fewer competitive advantages because of the way they fnance investments (i.e., based on re-investing profts and avoiding external investors who may jeopardise their control) and manage risk (i.e., more conservative strategies). Additionally, the lower presence of family frms in high-tech sectors seems to confrm the notion that family frms are less prepared to navigate highly dynamic sectors because frms in these sectors may require competitive advantages other than those generated by family involvement. The presence of family frms in high-tech sectors follows a common pattern across the seven European countries. Future research should explore the



*Table 2.4* Share of family frms by Pavitt sector and country

*Figure 2.2* Share of family frms by frm size, country, and region.

aforementioned patterns to explain why family frms are highly represented in traditional and specialised sectors, whether they have competitive advantages to operate in these sectors, what these are, and whether these patterns are the consequence of a natural selection process.

### *Prevalence of family frms based on their demographic characteristics*

Beyond the aforementioned analysis presenting the share of family frms across countries, regions, and industries, an interesting image emerges when analysing their size and age distribution across countries and regions. Family frms are more common in the small and medium-sized frm categories than in the large frm category (see Figure 2.2). However, the share of family

*Figure 2.3* Share of family frms by frm age, country, and region.

frms is high in the large frm category in some regions (e.g., in Germany, Austria, and Spain). These fndings open the door for further research investigating whether a particular relationship exists between regions and the size of family frms. Regarding frm age, Figure 2.3 shows three frm age classifcations: young frms (younger than 6 years old), mature frms (between 6 and 20 years old), and old frm (older than 20 years). Most family frms belong to the mature and old frm categories. However, there is a high dispersion of the share of family frms in terms of age across regions within each country. This descriptive information leads us to call for more research investigating the possible relationships between the age of family frms, the characteristics of regions, and the industries in which family frms dwell.

Finally, we analyse family involvement in the frm, measured as the average number of family managers, across regions in each European country. The aim of this analysis is to visualise whether there is an uneven distribution of family involvement in managerial positions across regions. Table 2.5 shows the extent to which family involvement in managerial positions varies across regions within countries. These differences imply that the level of 'familiness' (i.e., family involvement) of family frms is not the same across regions. For instance, while the dispersion of the average number of family members in German regions is high, ranging from less than 1 (.89) to almost 4 (3.84) family members, the dispersion of the


*Table 2.5* Average number of family managers in family frms

average number of family members in Austrian regions is low, ranging from 1.35 to 1.66. This evidence opens the door for future research to explain the possible connection between the degree of familiarity of family frms and regional characteristics.

In sum, from the aforementioned analyses, our main conclusion is that the phenomenon of family frms is unevenly distributed across European regions. This also seems to be related to industrial sector and, to certain extent, to frm characteristics, such as frm size, frm age, and degree of family involvement. The empirical evidence raises two general research questions that may deserve further investigation among family business and regional science scholars: '*Why is the prevalence of family frms unevenly distributed across regions?'* and *'Why do family frm characteristics (business related and family related) vary across regions?'* In an attempt to address these questions and unveil some preliminary evidence, in the next sub-sections, we explore the relationship between the prevalence of family frms and the quality of regional institutional contexts.

#### *Prevalence of family frms and regional characteristics*

#### *Prevalence of family frms and region size*

Following our intention to visualise patterns underlying the relationship between family frms and regions, Figure 2.4 plots the relationship between the share of family frms and region size, measured by the number of frms in the EFIGE dataset.<sup>5</sup> The majority of UK and German regions are concentrated in the quadrants representing a below-average share of family frms and a below-average number of regional frms. For Germany, only three regions have an above-average share of family frms and an above-average number of regional frms. Four of the six Italian regions have the same characteristics. French regions have a low share of family frms, but the majority of them are larger than average. This preliminary evidence opens the door for further investigations of the relationship between family frm prevalence and regional characteristics, specifcally in terms of external agglomeration.

 *Figure 2.4* Relationship between share of family frms and region size (proxied by total number of frms using the EFIGE dataset). Note: *y* = 0.70 + 0.0003*x* (*p*-value = 0.63).

### *Quality of regional institutions and prevalence of family frms*

The current debate in the academic sphere is whether there is a relationship between the presence of family frms and the quality of institutional contexts. The most well-known hypothesis, at least in explaining the prevalence of family frms in developing countries (Khanna & Palepu, 2000), is that the presence of family frms is higher in low-quality institutional contexts because families and their economic and social networks are able to fll institutional voids. To test whether there is an association between the quality of institutions and the presence of family frms in European regions, we use Charron, Dijkstra, and Lapuente's (2014) European 'quality of government' indicator (EQI) calculated at the regional level. This indicator is based on responses to a survey aimed at capturing average citizens' perceptions of and experiences with corruption and the extent to which they rate public services provided by local authorities as impartial and of good quality. In Figure 2.5, we use the scores of the three pillars that comprise this indicator—namely, corruption, impartiality, and quality (for more information, see Charron et al., 2014).

When we correlate the regional share of family frms and the quality of regional institutions in the European context, we observe contradicting results that both support and reject the argument linking family frms and

*Figure 2.5* Relationship between share of family frms and EQI. Note: Panel a: *y* = 0.71 − 0.009*x* (*p*-value = 0.55); Panel b: *y* = 0.71 − 0.029*x* (*p*-value = 0.06); Panel c: *y* = 0.69 + 0.061*x* (*p*-value = 0.00); Panel d: *y* = 0.70 − 0.019*x* (*p*-value = 0.33).

low-quality institutional contexts. While Panel a in Figure 2.5 shows that there is no association between the quality of government index and the prevalence of family frms, the analysis considering each dimension comprising the quality of government index shows differences. There is negative relationship between the quality dimension and the regional share of family frms (Panel b in Figure 2.5). This means that the lower the quality of services, the higher the percentage of family frms in the region, supporting the argument that family frms may replace formal institutions. However, the dimension of impartiality (i.e., the impartiality of institutions that exercise government authority) has a positive association with the regional share of family frms (Panel c in Figure 2.5). This means that the higher the impartiality, the higher the percentage of family frms in the region, which goes against the previous argument. For the last dimension, corruption, we do not fnd any association (Panel d in Figure 2.5). This preliminary evidence indicates that the prevalence of family frms in regions is complex and requires further attention. If formal institutions create conditions for individuals to exploit economic opportunities and incentives to use specifc business forms (i.e., family frms), we wonder—and future studies should further investigate–what combinations of formal institutional dimensions favour family frms and other types of organisational forms.

### *Family frms and regional competitiveness across European regions*

In this sub-section, following recommendations from Stough et al (2015), we focus on the aggregate effect of family frms across European regions on regional competitiveness in terms of productivity, exports, and innovation. While we implicitly argued in the previous sections that the family frm phenomenon is a consequence of regional context, in this section, we unveil an alternative contribution of family frms for regional development—namely, the economic and social impacts of having high- or low-level family frm prevalence. In other words, we wonder if the aggregate presence of family frms affects regional competitiveness.

### *Regional productivity*

Taking the whole sample, it looks like there is no relationship between regional total factor productivity (TFP) and the share of family frms.6 As we can see in Figure 2.6, German regions are mainly positioned in the top-right quadrant, characterised by a high share of family frms and high regional TFP. On the contrary, French and UK regions are positioned in the lower-left quadrant characterised by a low share of family frms and low regional TFP. Finally, Italian and Spanish regions share a similar pattern: most of their regions are positioned in the lower-right quadrant characterised by a high

*Figure 2.6* Relationship between share of family frms and TFP by region. Note: *y* = 0.66 + 0.041*x* (*p*-value = 0.396).

share of family frms and low regional TFP. Based on this preliminary fndings, future studies should deeply explore whether, when, and how family frms contribute to regional productivity.

### *Regional exports*

Our analysis shows that there is no relationship between the share of regional exporters and the prevalence of family frms across European regions.7 Only in the UK context does the relationship seem to be positive—that is, the higher the presence of family frms, the higher the exporters in the region. Even though it is widely recognised that exports and productivity are generally correlated at the frm level (i.e., the most productive frms are more likely to become exporters), we wonder if this effect can be extrapolated to the regional level. As shown in Figure 2.7, the majority of German regions are in the quadrant characterised by a high share of family frms and a below-average share of exporters, while three out of the six Italian regions are in the quadrant characterised by a high share of family frms and a high share of exporters. In this sense, further studies should investigate the possible connection between regional exports and productivity and the regional prevalence of family frms.

### 46 *Rodrigo Basco and Fernanda Ricotta*

### *Regional innovation*

There seems to be a slightly positive relationship between the share of family frms and regional innovation.8 The majority of German regions are located on the right side of Figure 2.8, indicating a high share of family frms, but these areas have near- or below-average regional innovators (calculated for all European regions). In the case of Italian regions, four out of six are characterised by both a high share of family frms and above-average regional innovators. This preliminary evidence opens the door for future research exploring the relationship between innovation and tradition in family business and its contribution to regional innovation systems.

# *Combining family frm exporters and innovative behaviour across regions*

Focusing on family frms, we classify regions in terms of family frms' exporting and innovation activities. Indeed, Figure 2.9 combines the measures of regional family frm exporters and innovators,9 forming four quadrants: less dynamic regions, exporter regions, dynamic regions, and innovator regions.10 Most Italian and Austrian regions fall into the dynamic quadrant in which the share of family frm exporters and innovators is high. In contrast,

Note: *y* = 0.524 + 0.154*x* (*p*-value = 0.081).

*Figure 2.9* Regional share of family frms combining innovators and exporters. Note: *y* = 0.167 + 0.675*x* (*p*-value = 0).

most German regions fall into or are close to the quadrant characterised by less dynamic regions in which the share of family frm exporters and innovators is low. The preliminary fnding for Germany may require further research because of the importance of family frms as the backbone of innovation and internationalisation in German regions. However, this general perception may be based on only a few frms, so the same pattern may not necessarily hold for other family frms that form the regional productive structure. French regions follow a similar pattern to that of German ones, and Spanish regions are distributed between dynamic and innovator regions.

# **The case of Italy**

To further explore the prevalence and importance of family frms across regions, in this section, we focus our analysis on the Italian context. Italy is an interesting case study for two main reasons. First, it has high geographical heterogeneity in economic conditions that has created large and persistent disparity between the south and the rest of the country (Iuzzolino, Pellegrini, & Viesti, 2011). Second, the backbone of its productive system is characterised by a large number of small and medium enterprises (SMEs) that are often organised within industrial districts and by the progressive disappearance of large frms in strategic sectors (Amatori, Bugamelli, & Colli, 2011). A large portion of these SMEs are run as family businesses or have family shareholders present who are able to infuence company decisions (Cucculelli & Micucci, 2008).

Data for our analysis come from the Xth UniCredit-Capitalia survey (2008), which covers the period 2004–2006 and was compiled using information collected through a questionnaire sent to a representative sample of Italian manufacturing frms. While the survey covers the universe of manufacturing frms in Italy with more than 500 employees, it uses a representative sample of manufacturing companies with 10–500 employees considering three strata: four geographical areas, four Pavitt sectors, and fve frm size classes.11

Following the structure we used in the previous section, we start the analysis with a descriptive view of the prevalence of family frms across Italian geographical areas and regions.12 For the sake of completeness, in the tables, we leave regions with few frms, such as Calabria, Basilicata, and Molise.13 For these regions, the results are only representative of the frms reported in the sample. Additionally, we focus on family frm prevalence by considering business and family characteristics, such as industry, frm size, frm age, and family involvement at the frm level. Second, we analyse the relationship between the prevalence of family frms and the quality of regional institutions. Finally, we reveal the importance of family frms in terms of regional productivity, regional exports, and regional innovation.

### *Descriptive analysis of family frms across Italian regions*

In this descriptive analysis, a frm is considered to be a family frm if the respondent declared in the UniCredit-Capitalia questionnaire that the company is controlled or owned by an individual or a family. The average share of family frms in Italy is 66%. The four geographical areas (northeast, northwest, centre, and south) have a similar average share of family frms, ranging from 65% for the northeast to 68% for the south. All Italian regions have a high presence of family frms, and the share of family frms across regions varies from approximately 50% in the region of Sardegna to more than 70% in Sicilia (see Table 2.6). These fndings confrm our argument in the previous section that family frms are unevenly distributed across regions.

### *Prevalence of family frm across industries*

The prevalence of family frms across sectors is high in almost all subsectors except for 'coke, refned petroleum products, and nuclear fuel', for


*Table 2.6* Share of family frms across Italian regions

(\*) Statistics calculated considering regional values for the share of family frms.

which the percentage is less than 50% (see Table 2.7). The distribution of the share of family frms oscillates between 57% for 'medical, precision and optical instruments, watches and clocks' and 76% for 'offce machinery and computers'. In terms of Pavitt sectors, the prevalence of family frms is evenly distributed across traditional, high-tech, specialised, and economies-of-scale sectors (Table 2.8). However, when looking at the share of family frms across sectors and regions, we observe that in the northwest, they are highly represented in traditional sector, whereas in the south, they are highly represented in the high-tech sector. However, it is worth noting that only 13 frms in the south are in the high-tech sector. These results for Italy challenge what we found in the previous section when analysing all seven European regions. It seems that the relationship between family frms and industries requires further research.


*Table 2.7* Share of family f rms by sector and geographical area


*Table 2.8* Share of family frms by Pavitt sector and geographical area



### *Prevalence of family frms based on their demographic characteristics*

Regarding the share of family frms considering the size of frms across Italian regions, the empirical evidence shows mixed results (Table 2.9). Looking at the different geographical areas, in the northeast, the presence of family frms is similar across all three frm size categories; however, in the remaining geographical areas, their prevalence in the large frm category is lower than in the other size categories. Focusing on regions, there are regions in which the prevalence of family frms is lower in the medium and large frm categories than in the small frm category, such as in Campania and Umbria, and there are other regions in which the prevalence remains stable across all three frm size categories, such as in Emilia Romagna. Even though these results may partially support the general belief that family frms are more prevalent in the small and medium frm categories, this is not the case in all regions. It is important to highlight that Emilia Romagna is well known for hosting clusters of frms related to mechanical engineering, automotive manufacturing, and agrifood. Thus, one could assume that there might be a relationship between family frm prevalence and external localisation economies that could be explored further. Future studies should thus investigate what regional mechanisms can explain the uneven/even regional distribution of family frms across different size categories.

When considering family frm prevalence across age categories (Table 2.10) in the Italian regional context, we also observe that family frms tend to belong to the mature (between six to 20 years old) and old (more than 20 years old) frm age categories. This result may support the argument that with time, frms become family frms by incorporating family members and by developing founders' intention to transfer ownership and/or management to upcoming generations. In general, the percentage of family frms in the new frm category (less than 6 years old) is lower than in the other categories


*Table 2.10* Share of family frms across frm age categories and regions


*Table 2.11* Family involvement across Italian regions (average number of family managers)

across almost all Italian regions (with some exceptions, such as Campania and Puglia). It is important to highlight the high presence of young family frms in regions located in the south. This geographical area has the highest share of young family frms. One reason for this high prevalence of young family frms could be the incentives provided in this part of the country for new start-up frms. Further research should investigate the phenomenon of entrepreneurial families that seems to have emerged in the south of Italy in contrast with the presence of mature family frms in the north of Italy and the consequences for future regional economic and social development.

Finally, Table 2.11 shows the familiarity of family frms (measured as the average number of family managers) across Italian regions. There are regions with an average of two family managers or higher, such as in Friuli Venezia Giulia and Liguria, while regions in the south all have an average number of family managers lower than the national mean (1.78). As expected due to the higher presence of mature and old family frms, in the north of Italy we have more family members involved than family frms in the south.

#### *Prevalence of family frms and quality of Italian regional institutions*

Following the aim to relate the prevalence of family frms and the quality of institutions for the Italian sample, we use Golden and Picci's (2005) corruption indicator as a proxy for the quality of regional institutions. This

*Figure 2.10* Quality of Italian regional institutions and share of family frms. Note: *y* = 0.7191 − 0.00058*x* (*p*-value = 0.169).

index was created in 1997 by focusing on the difference between the total amount of fnancial resources allocated to build infrastructure in regions and the physical inventory of public capital that has effectively been built after controlling for regional differences in the cost of public construction. The intuition underlying this indicator is that, all else being equal, if the difference between these measures is large, that means the focal government is not getting the infrastructure it has paid for due to the presence of corruption (Golden & Picci, 2005). The indicator is based on the ratio between the aforementioned two measures and is expressed as ratio to the national average. A higher value implies lower corruption and therefore higher-quality regional institutions.

Even though it is not signifcant, Figure 2.10 shows a preliminary pattern indicating that the prevalence of family frms decreases with the effciency of public administration, supporting the arguments that family frms are better equipped to survive in less developed formal institutional contexts. Even though this theory has been used to explain the presence of family frms in emerging economies, it looks like it can also be applied to developed economies by considering the different stages of regional development.

#### *Family frms and regional competitiveness across Italian regions*

### *Regional productivity, regional exports, and regional innovation*

To explore the possible relation between the prevalence of family frms and regional competitiveness, we use three aggregate regional measures: labour productivity, exports, and innovation.14 First, there is no evidence that the

*Figure 2.11* Share of family frms and labour productivity across Italian regions. Note: *y* = 65,473.45 − 7,685.174*x* (*p*-value = 0.809).

*Figure 2.12* Share of family frms and regional exports across Italian regions. Note: *y* = 1.246 − 1.13*x* (*p*-value = 0.013).

prevalence of family frms is positively or negatively related to regional labour productivity (Figure 2.11). When analysing regional exports, there is preliminary evidence that the relationship is negative. However, there is a group of regions (Friuli Venezia Giulia, Lombardia, Marche, and Umbria) characterised by a high presence of family frms and a high share of exporters

 *Figure 2.13* Share of family frms and regional innovation across Italian regions. Note: *y* = 0.45 + 0.15*x* (*p*-value = 0.552).

 *Figure 2.14* Share of family frm innovators and family frm exporters across Italian regions.

(Figure 2.12). Finally, the share of regional innovators does not seem to have a relationship with the regional prevalence of family frms (Figure 2.13). In line with fndings from Stough et al. (2015), our results show the need to further investigate the aggregate effect of family frms on regional competitiveness to uncover their contributions to regional development.

Finally, to explore the possible combination of family frm exporters and family frm innovators across regions, Figure 2.14 shows the position of Italian regions in four quadrants combining the share of regional family frm exporters and the share of regional family frm innovators.15 It seems that in some Italian regions, there is an interesting combination of a high level of family frm exporters and a high level of family frm innovators. These regions are positioned in the 'dynamic' quadrant.

# **Conclusion**

The aim of this chapter was to present a descriptive picture of the phenomenon of family frms across European regions and to analyse its prevalence, its relationship with context, and its consequences for regional competitiveness. Our main conclusion is that the phenomenon of family frms is unevenly distributed across regions. Additionally, our intention was to dig into the relationship between family frms and regional context to provide some empirical evidence of the two-way relationship between family frms and regions—that is, the extent to which the territory determines the existence of family frms and the extent to which family frms contribute to regional


*Table 2.12* Future research questions

competitiveness. Instead of providing defnitive answers, this book chapter has raised several research questions that future research should investigate further (see Table 2.12 for a summary of the most important research questions).


### **Appendix A**

### **Notes**


3 Given the representativeness of the EFIGE database, we are confdent that the distribution of family frms in our dataset refects the distribution in each country.

 


# **References**


# **Part II Micro-foundation channels**

# **3 Urbanization economies, proximity dimensions and productivity**

A family frm perspective

*Rodrigo Basco, Stefano Amato, Silvia Gómez-Ansón and Andrea Calabrò* 

### **Introduction**

After more than three decades of research on family businesses, one of the main unresolved research questions revolves around why family and non-family frms differ in terms of behaviour and performance. The most common answer is that family involvement in business affects the way an organisation is owned, governed, and managed, thereby causing family and non-family frm behaviour, performance, and survival to differ. Even though extensive research has attempted to explain, justify, and predict frm performance differences between family and non-family frms, the results are still contradictory (Basco, 2013; Mazzi, 2011).

Family business researchers have built a new body of knowledge to explain differences between family and non-family frms by investigating the family's effect on the frm. More specifcally, these scholars have argued that the dominant coalitions in family and non-family frms have different goals (Williams, Pieper, Kellermanns, & Astrachan, 2019). In the former, owner families imprint their unique goals on their frms, such as their intentions to transfer the ownership from one generation to another, to embrace and support family members around economic projects, and to preserve and sustain their family image and reputation across time (Aparicio, Basco, Iturralde, & Maseda, 2017). Even though some of the differences between family and non-family frms originate from the interrelationship between the family and business domains, family business research is still *contextless* (Gomez-Mejia, Basco, Müller, & Gonzalez, 2020; Krueger, Bogers, Labaki, & Basco, 2020), with context being defned as 'circumstances, conditions, situations, or environments that are external to the respective phenomenon and enable or constrain it' (Welter, 2011, 167). Specifcally, scholars have given insuffcient attention to how and to what extent family and non-family frms are affected by and able to affect context. Consequently, a future challenge for the feld is to explore and interpret how contextual dimensions (e.g., spatial, social, institutional, and temporal), insofar as they are accounted for, shed new light on family frms' heterogeneity.

### 66 *Rodrigo Basco et al.*

Focusing on differences in productivity between family and non-family frms, we explore the contingent effect of the size of the municipalities in which frms dwell. We focus on urban settings for two main reasons. First, municipalities confate spatial and social relationships in administratively bounded areas, hence creating well-defned territories (Bathelt & Gluckler, 2003). Second, urban areas are a source of externalities (i.e., urbanisation economies) for the frms located within them, as refected in these frms' static (e.g., higher proftability, productivity) and dynamic (e.g., higher innovative capacity) advantages compared to frms located elsewhere (Capello, 2002). Our main conjecture is that due to different family and non-family frm specifcities, neither is a superior form of organisation in terms of productivity; however, for both types, productivity depends on frms' ability to exploit economies of proximity and urban agglomerations. Following the so-called 'regional familiness approach' (Basco, 2015), we argue that family frms, which are more locally embedded in their home territories than non-family frms, are better positioned to exploit proximity dimensions as vectors of territorial competitiveness because they can create particular economic, social, and emotional connections within their local socioeconomic *milieus* (Boschma, 2005). Therefore, the specifcities of family frms give them locational advantages over their non-family counterparts when operating in small municipalities. In small municipalities, the embeddedness relationships developed business families both internally (i.e., among family members) and outside their organisational domains (i.e., in their local settings) substitute for the lack of spatial agglomerations.

We test our conjecture on a large panel dataset of Spanish manufacturing frms covering the 2002–2015 period. Our empirical evidence reveals a negative association between municipality size and the productivity of familymanaged frms. In particular, being located in a large urban setting is a source of diseconomies of agglomerations for family-managed frms. As such, they are better suited to exploit highly embedded contexts, such as small municipalities. Accordingly, we address the call made by Stough et al*.*  (2015) to link the research felds of regional development and family business by exploring the relationship between spatial-temporal context and the nature of the frm to better understand frm growth in terms of productivity. In particular, by emphasising the spatial dimension of embeddedness (Hess, 2004)—that is, by anchoring economic action in territorially bounded networks of social relationships—our book chapter offers new evidence on how family frms are distinctively affected by their immediate surroundings.

# **Theoretical background**

Family business research has mainly focused on the family-business relationship to explain differences in frm behaviour and frm performance between family and non-family frms. The primary argument to explain the specifcities comes from the family frm goal approach (Aparicio et al., 2017; Basco, 2017). According to this approach, family frms are able to pursue multiple goals that combine economic and non-economic orientations as well as business and family orientations. Since organisation goals alter the reference point for decision making, we expect that the reference point used by family frms is different than that used by non-family frms (Gomez-Mejia, Cruz, Berrone, & De Castro, 2011).

Beyond extensive research taking an internal view (i.e., how the interplay between the family and business domains shapes frms' behaviour) to explain family and non-family performance differences, existing empirical evidence is still contradictory, and there is no clear understanding of whether the family frm is a superior form of organisation (Mazzi, 2011). The main issue is that family business research is *contextless* (James et al., 2020) that is, the spatial, social, institutional, and temporal dimensions of context have usually been overlooked in favour of the traditional internal view. To address this research gap, Amato, Basco, and Lattanzi (2020) recommend accounting for context to better interpret differences between family and non-family frms.

Context is a dimension that can both constrain and bolster frm competitive advantages and performance while simultaneously explaining frm heterogeneity (Amato, Basco, & Lattanzi, 2020). This line of research is gaining importance in family business studies, and recent research has shown that frm differences in terms of performance and behaviour vary across socio-spatial contexts, such as rural and urban areas (Backman & Palmberg, 2015; Baù et al., 2019), municipalities (Amato, Basco, Gómez-Ansón, & Lattanzi, 2020), and regional settings (Adjei, Eriksson, & Lindgren, 2016). Consequently, context matters.

### *Regional familiness: an urban perspective*

Context matters because family and non-family frms differ in the extent to which they are 'anchored' to the territories in which they dwell (Backman & Palmberg, 2015). While the economic link with territory defnes how frms behave to exploit economic opportunities, family frms also usually develop specifc social and emotional connections with their home territories (Smith, 2016). However, family frms' social and emotional attachments to their territories are not necessarily what make them perform differently from non-family frms; rather, these performance differences stem from family frms' ability to either exploit the advantages offered by or overcome the constraints inherent to a given location (Capello, 2002). The regional familiness approach, which encompasses 'the embeddedness of family businesses in social, economic, and productive structures within a spatial context' (Basco, 2015, 260), provides a theoretical explanation to link the study of family frms and regions.

The interpretation of space as 'diversifed-relational' has restored the concept of external (or agglomeration) economies as sources of territorial competitiveness (Capello, 2009). The concentration of economic activity in spatially bounded areas gives rise to advantages for frms in the form of reduced production and transaction costs, enhanced effciency of production factors, and increased innovative capacity. The term 'external' means that agglomeration economies are beyond frms' control and typically result from the presence of collective action among other frms and institutions, thus making them external to a focal frm but internal to either the industry or urban concentration (Parr, 2002).

Urban spaces are generally regarded as localities in which agglomeration economies occur. In particular, the local diversity of cities facilitates access to a qualifed and diversifed workforce, various infrastructure, a variety of facilities, and—even more importantly—to complementary knowledge, with industry diversifcation fostering frms' productivity and innovation performance (Galliano, Magrini, & Triboulet, 2015; Jofre-Monseny, Marín-López, & Viladecans-Marsal, 2014). Hence, a diversifed spatial setting typically leads to increasing returns by giving rise to so-called 'urbanisation economies', or 'Jacobs externalities'. A city arises as a spatial cluster of productive and residential activities (Parr, 2007). The concentration of a mix of sectors and diversifed activities, the density of contacts that develop within them, and the easy access to advanced information and knowledge are clear advantages arising from being located in an urban setting that affect the productivity of the frms situated therein (Capello, 2002). At the same time, cities are able to generate 'dynamic' advantages (Jofre-Monseny et al., 2014). In particular, urban settings are characterised by shared values, common codes of behaviour, a sense of belonging, and mutual trust—all of which play a signifcant role in reducing uncertainty and in the socialisation process of knowledge development and collective learning that affects the innovativeness of co-located frms (Capello & Faggian, 2005).

Urbanisation economies are generally regarded as being dependent upon the overall scale of the respective city (Fu & Hong, 2011), with urban size being closely related to industrial diversity (Gao, 2004). That said, it is reasonable to ask whether frms beneft equally from urbanisation economies or, conversely, whether being located in an urban setting is a source of differential advantages or disadvantages for some frms and not for others. For instance, while city size has been found to affect small frms' productivity due to their higher reliance on a more diversifed external environment than larger enterprises (Fu & Hong, 2011; Henderson, Kuncoro, & Turner, 1995), in small spatial settings, where access to agglomerations is more restricted, frm performance is strongly dependent upon proximity dimensions as conduits of interactive learning, cooperation, and knowledge exchange (Gordon & McCann, 2000). However, particular types of frms (e.g., family-managed frms) may have a greater ability than others to leverage proximity dimensions in certain spatial contexts, such as cities.

Geographical proximity, which refers to the physical distance between economic actors (e.g., customers and suppliers) and regional factors (e.g., raw materials for production processes), seems to be important for any frm, including family and non-family ones. Being close to raw materials, customers, suppliers, and sources of knowledge could be considered an advantage that any frm can exploit. In particular, research has argued that geographical proximity is a key element for the promotion of externalities (Martin & Simmie, 2008). However, not all frms are able to capitalise on the condition of being geographically close to each other, which is a necessary but insuffcient condition to improve communication and trust among economic actors and which ultimately affects the effciency of economic activity. Geographical proximity has to be lubricated with additional dimensions of proximity, such as cognitive and social proximity, which mirror feelings of similarity with and belonging to a given location, respectively (Lähdesmäki, Siltaoja, & Spence, 2019).

While cognitive proximity refers to 'the similarity of the subjective mental framework of actors and the tacit and codifed knowledge owned by actors' (Westlund & Adam, 2010, 112), social proximity refers to socially embedded relationships among agents based on trust and reciprocity derived from friendship, kinship, and experience (Boschma, 2005). For family frms, cognitive proximity is manifested through the emotional connection that business families are able to develop within their home territories. It is through this attachment that knowledge related to business fows and consolidates economic activity, thus enabling family frms to exploit competitive advantages from being family frms. Alongside cognitive proximity, social proximity seems to be a family-inherited condition stemming from families' long-standing presence in their territories (Cucculelli & Storai, 2015). Family members bring social and kinship relationships to their frms, which help them exploit economic opportunities, ease the exchange of tacit knowledge, and reduce opportunistic behaviours (Baù et al., 2019). Hence, we expect that cognitive and social proximity reinforce each other to foster the economic activity in a given location.

Even though family frms seem to be positioned to uniquely exploit the benefts of social and cognitive proximity, their outcomes depend on the size of the spatial contexts in which they dwell (Martin & Simmie, 2008). In particular, municipality size could constrain or broaden family frms' ability to transform their social and emotional embeddedness into superior performance. In small municipalities, where access to urbanisation economies is more restricted (Fu & Hong, 2011; Gordon & McCann, 2000), frms have to develop high-trust, cooperative, and reciprocal relationships both among their members and with external networks of local actors (i.e., local communities) to facilitate the economic activity. In this context, because of their emotional and social connections, family frms are in a superior position to beneft from interactive learning and knowledge exchange both within and outside their organisational boundaries (Adjei et al., 2016). On the other hand, in large urban settings, frms can beneft from Jacobs externalities such that family frms' competitive advantages appear to vanish, and any frm can exploit the externalities stemming from knowledge spillovers, a large pool of skilled labour, and increased effciency in labour market matching (Galliano, Magrini, & Triboulet, 2015).

### 70 *Rodrigo Basco et al.*

Hence, while in small urban settings, family frms' unique physical, social, and emotional connections substitute for the shortage of external agglomerations, their comparative locational advantages disappear as municipality size increases. That said, our conjecture is that family frms located in small municipalities have higher performance than their non-family counterparts. Therefore, we propose the following:

Hypothesis 1: *Municipality size affects the performance of familymanaged frms in such a way that the larger the municipality in which a family frm dwells, the lower its productivity.* 

# **Method**

### *Data and variable*

To test our hypothesis, we rely on micro-data obtained through a survey of a representative sample of Spanish manufacturing frms. The survey, known as Encuesta sobre Estrategias Empresariales (ESEE, or Survey on Business Strategies), is conducted yearly by the SEPI Foundation in collaboration with the Ministry of Industry, Trade and Tourism of Spain. ESEE is oriented toward capturing information about frms' strategies, technological activities, manufacturing processes, markets served, and employment. The sample's representativeness is ensured by combining exhaustiveness and random sample criteria. In particular, while all companies with more than 200 workers are surveyed, frms employing 10–200 workers are selected based on stratifed, proportional, and systematic sampling.1 The fnal sample includes 3,331 frms distributed across 20 different manufacturing industries (NACE Rev. two-digit level)<sup>2</sup> and 17 Spanish autonomous communities (NUTS 2).<sup>3</sup> It consists of 21,573 frmyear observations for the 2002–2015 period.

The dependent variable used in this study is labour productivity, defned as per capita value added (Adjei et al., 2016). We take the log of the values to reduce the skewness of the distribution.

The main exploratory variable is represented by the family status of the frm. As the defnition of 'family frm' is a matter of longstanding debate among researchers (Mazzi, 2011), we adopt the so-called 'demographic approach' to identify family frms. This approach considers the involvement of a family in a frm—in its ownership, control, and management—as a suffcient condition to capture families' infuence on businesses (Basco, 2013a). Since ESEE reports the number of owners and relatives holding management positions, we defne a family frm as any frm in which two or more members of the controlling family hold managerial positions in the company. Therefore, we employ a binary variable coded 1 when the frm is family managed and 0 otherwise. To capture urbanisation economies, our moderator variable, we consider the size of the municipality in which the frm *i* is located (Fu & Hong, 2011). In particular, ESEE reports a categorical variable based on fve different municipality sizes: fewer than 2,000; 2,001– 10,000; 10,001–50,000; 50,001–500,000; and more than 500,000 inhabitants.4

Additionally, we control for a set of frm-level characteristics that potentially affect the level of productivity. To account for time-invariant heterogeneity across industries and regions, we include a series of categorical variables corresponding to the NACE two-digit code level and to the Spanish autonomous communities in which frms have their headquarters, respectively. Finally, we use a series of categorical variables to control for the years associated with each observation. Table 3.1 summarises the variables employed in the study.


LExpressed in natural logarithm.

# 72 *Rodrigo Basco et al.*

# *Empirical model*

To test the proposed hypothesis, we use panel data analysis with randomeffects specifcation. The preference for random-effect models is due to the low 'within' variance of the family status of the frm, which changes very little across time.

With the purpose of investigating the effect of municipality size on the level of productivity of family versus non-family frms, we estimate the following model:

 ln *Yi*, *t* = *α*0 + *β*1*Fi*, *t* + *β*2*Mi*, *t* + *β*3(*Fi*, *t* × *Mi*, *<sup>t</sup>* ) + *β*΄*Ci*, *t* + *β*΄*Ti* + *β*΄*Si* + *β*΄Ri + *εit*,

where


# **Results**

The descriptive statistics and pairwise Pearson correlation results are reported in Tables 3.2 and 3.3, respectively. Panel 2A provides the summary statistics for the whole sample. Family-managed frms account for 23% of the total sample. For a more straightforward depiction of the difference between family and non-family frms, Panel 3B shows the means of the variables grouped by the nature of the frm (family versus non-family), along with the results of a test for differences in the means and the results of the Wilcoxon rank-sum test. Non-family frms are older, bigger, more innovative, more internationalised, and more open to foreign investors than family frms. Additionally, in relation to our frm performance variable—labour productivity—non-family frms are more productive than family frms.

An analysis of the variance infation factors (VIFs) suggests that multicollinearity is not a concern because all the VIF coeffcients are below the generally accepted threshold of 10, as shown in Table 3.3.


# *Table 3.2* Descriptive statistics



AExpressed in absolute terms. LExpressed in natural logarithm.

Level of statistical signifcance: +*p* < 0.10, \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001. a The Wilcoxon rank-sum test analyses whether the two samples are from different distributions (Sample 1: non-family frms; Sample 2: family-managed frms).


Number of observations = *21,573.* Mean VIF = 1.27. Level of statistical signif cance: **\****p* < 0.05.

*Table 3.3* Pairwise Pearson correlation coeff cients


*Table 3.4* Family-managed frms, labour productivity, and municipality size

The table presents random-effect models based on a panel dataset of frm with at least 10 employees over the 2002–2015 period. Robust standard errors are reported in parentheses. Level of statistical signifcance: +*p* < 0.10, \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001.

The results of the econometric analysis are reported in Table 3.4. In Model 1, we introduce all control variables and our main exploratory variable—family-managed frm. We observe that process innovation, export, age, size, foreign share, being part of a group, and being listed positively

*Figure 3.1* Labour productivity of family versus non-family frms across municipalities.

affect frms' labour productivity. Conversely, R&D intensity and fnancial constraints frm performance. Additionally, even though the family frm coeffcient is negative, it is not signifcant.

 In Model 2, we introduce our moderator variable—municipality size which has a positive and signifcant relationship with labour productivity (*β* = 0.011, *p* < 0.10). This result suggests that municipality size is source of increasing returns to city scale in general, hence providing support for urbanisation economies (Fu & Hong, 2011). Finally, in Model 3, we test our hypothesis by introducing the moderating effect of municipality size. The interaction term is negative and statistically signifcant (*β* = −0.023, *p* < 0.05), which suggests that municipality size has an adverse effect on family frms' productivity.

For a more straightforward interpretation of this result, we plot the two-way interaction in Figure 3.1. We observe that municipality size has antithetical effect on the productivity of the two types of frms—family and non-family frms. In particular, family-managed frms appear to be more productive in small municipalities than non-family frms, with their performance decreasing with an increase in city scale. Conversely, non-family frms beneft of the size of the municipalities in which they are located.

We perform a sensitivity analysis to corroborate our results. Specifcally, in lieu of the dichotomous variable, we use a continuous measure of family involvement (*Family management*) to capture the number of family members


*Table 3.5* Robustness check: Family management, labour productivity, and municipality size

The table presents random-effect models based on a panel dataset of frms with at least 10 employees over the 2002–2015 period. *Family management* is a continuous variable that counts the number of family members in managerial positions. Robust standard errors are reported in parentheses. Level of statistical signifcance: +*p* < 0.10, \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001.

in managerial position. Table 3.5 confrms our previous results—that is, family frm performance is negatively affected by municipality size, with the differential of productivity between the two types of frms being particularly large in small municipality.

# **Conclusion**

One of the main research questions that family business scholars have focused on is whether the family frm is a better form of business organisation than the non-family frm. Because current fndings comparing family and non-family frm performance are contradictory (Mazzi, 2011), in this book chapter, we addressed the call made by Amato et al. (2020) to include some contextual dimensions to tease out when and where family and non-family frms have superior performance.

Our main theoretical reasoning is that family and non-family frms are able to exploit their competitive advantages in different external environments. That is, each type of frm can materialise their strengths and reduce their weaknesses in different ways depending on the socio-spatial contexts in which they are located (Cucculelli & Storai, 2015), the set of formal and informal institutions defning the 'rules of the game' in their societies (Soleimanof, Rutherford, & Webb, 2018), and particular events or historical periods (Smith, 2016). More specifcally, in this chapter, we explored the family frm-territory nexus stemming from being located in cities, as a particular type of socio-spatial context (Capello, 2002). By drawing on the regional familiness approach (Basco, 2015), we posited that family frms are better positioned to beneft from proximity dimensions (i.e., geographical, social, and cognitive) arising from being located in small urban areas than from the resulting agglomeration economies. Because of their territorial embeddedness and socio-emotional characteristics, family frms are in a superior position to exploit economic opportunities, facilitate the exchange of tacit knowledge, and acquire tangible and intangible resources in peripheral areas, such as in small municipalities, where access to agglomerations is normally more diffcult.

Our fndings reveal that family frms' labour productivity is strongly affected by the size of the municipalities in which they dwell—a proxy for the continuum between proximity economies and urbanisation economies. In particular, while family frms' productivity is negatively associated with municipality size, non-family frms beneft from increasing returns to city scale. Our results are in line with previous studies (Backman & Palmberg, 2015; Baù et al., 2019) that show that family frms located in sparsely populated areas, such as rural ones, have better performance than non-family frms. Taken together, this evidence shows how reciprocal trust-based relationships established both among family members and between family frms and their immediate surroundings equip them with unique locational advantages as compared to non-family frms.

# *Contributions*

Our book chapter makes important contributions to theory and practice. First, it contributes to the family business feld by shedding new light on the micro-territorial foundations of family frms. In particular, our empirical evidence supports the view that family frms are inherently spatial. That is, they are distinctly responsive to geographic fragmentation and spatial variation in terms of resource availability and business opportunities, both of which are strongly related to the size of the municipalities in which frms are located (Backman & Palmberg, 2015). At the same time, family frms are inherently local, meaning they are deeply anchored in territorially bounded areas. Family frms' embeddedness in localised networks of relationships results in a superior position to leverage the advantages of external proximity dimensions and, hence, leads to them having competitive advantages in small municipalities.

Second, our research contributes to regional studies specifcally by introducing family frms in the debate on urbanisation economies and local economies (economies of proximity). While frm size and industry have been extensively explored in this respect (Fu & Hong, 2011; Henderson et al., 1995), our chapter highlights the opportunity to take the frms' family status into account to better comprehend Jacobs externalities. Additionally, given the prevalent view that regional development ultimately stems from the balanced growth of individual cities (Capello, 2009), regional scholars' acknowledgment of family frms could provide new insights into the role these organisations play in the competitiveness and sustainability of urban settings.

Finally, our chapter has important contributions for policymakers. Based on our evidence, policymakers should be aware of different types of economic actors (e.g., family and non-family frms) when tailoring polices to develop prosperous, diversifed, and sustainable regions because the demographics of economic structures may determine the effectiveness of policies. For instance, since diversifed urban environments have an asymmetrical effect on frm performance—the benefts of urbanisation are better exploited by non-family frms than family frms. On the other hand, family frms are better positioned to leverage the external advantages of proximity in small municipalities. Therefore, development policies should be designed carefully, taking into consideration both the territory and the economic actors who form the economic structure of the territory itself.

### *Limitations and future lines of research*

Our research is not exempt from having limitations, which in turn pave the way for future lines of research. First, our contextual variable (i.e., municipality size) represents only a partial measure of urban agglomerations. Even though municipality size is strongly related to the degree of diversifcation of urban settings, future research should employ a fner-grained measure of urban diversity and simultaneously test the effect of urban size and industrial diversity on family frm performance (Fu & Hong, 2011). Second, because cities are sources of dynamic economies as refected in the higher innovativeness of co-located frms (Capello, 2002), future studies should explore whether and to what extent family-managed frms' innovation outputs (e.g., patents) are affected by urban agglomerations. Finally, future research could improve the classifcation of frms. In this chapter, our main classifcation comprised family and non-family frms, but we know this is only a rough measure to capture the heterogeneity among economic actors. Within family and non-family frms, there are sub-classifcations capable of producing richer information. Regarding family frms, not all are alike, and different types may react differently to and be affected differently by the socio-spatial contexts in which they are located (Smith, 2016). For instance, generational involvement and level of 'familiness' are two traditional dimensions used to capture different types of family frms that account for family frms' heterogeneity in terms of both demographics and competitive advantages (Habbershon & Williams, 1999).

# **Notes**


ally expressed as 1 minus the Herfndahl index in terms of employment in one-digit *<sup>M</sup>*<sup>2</sup> ˝ *<sup>M</sup>*<sup>ˇ</sup> industries in a given city: Urban diversity = −<sup>1</sup>˜ <sup>ˆ</sup>*E mk* / ˜ *E mk*  , *<sup>m</sup>*=1˙ *<sup>m</sup>*=1 ˘ where *Emk* is the number of employees in one-digit industry *m* in city *k*, and *<sup>M</sup>* is the total number of one-digit industries in city *k*. The value of *urban diversity*  ranges from 0 to 1, with a value closer to 1 suggesting more diversifcation of the urban setting.

# **References**

Adjei, E. K., Eriksson, R. H., & Lindgren, U. (2016). Social proximity and frm performance: The importance of family member ties in workplaces. *Regional Studies, Regional Science*, 3(1), 303–319.


### 82 *Rodrigo Basco et al.*


# **4 Family co-occurrence and frm productivity**

*Evans Korang Adjei and Rikard Eriksson* 

### **Introduction**

Does family co-occurrence in businesses affect frm productivity? While this general question is at the core of research in family business, previous investigations are plagued by inconclusive answers. Several scholarly works have found variegated impacts of family co-occurrence on frm productivity (e.g., Gedajlovic, Carney, Chrisman, & Kellermanns, 2012). While some studies have argued that family co-occurrence in businesses, especially in family frms, helps promote localized learning and alleviate conficts of interest between owners and managers (Chrisman, Chua, & Litz, 2004; Jensen & Meckling, 1976), others have found family involvement to be detrimental for frm productivity due to moral hazards and adverse selection (Schulze, Lubatkin, Dino, & Buchholtz, 2001).

Despite the ambiguity regarding the role of family involvement in business, kinship ties still constitute an important part of the recruitment process. Although a systematic cross-country mapping of family involvement in business is missing, family co-occurrence constitutes about 14% of all employment in Sweden (Holm, Westin, & Haugen, 2017). This fnding supports the notion that meritocracy in the labour market co-exists with other types of hiring practices (Adjei, 2018), such as nepotism. Despite the obvious representation of family ties in the workplace and research indicating that family typologies play a major role in regional development in terms of gross domestic product (GDP) per capita (Duranton, Rodríguez-Pose, & Sandall, 2009), the relationship between family co-occurrence and frm productivity is relatively under-researched. Basco (2015), for instance, argues that regional development studies have neglected to investigate the family's role in frm behaviour and the subsequent consequences for regional economic and social development. While some regional development studies have recognized the role of social capital in shaping competitive advantages (Saxenian, 1994), the potential role of familial relationships has often been studied only through case studies (Gurrieri, 2008; Johannisson et al., 2007).

Consequently, because there is hardly an aspect of society that is not affected by the family (Alesina & Giuliano, 2014), we focus on the link between family co-occurrence and frm productivity. Specifcally, in this chapter, we investigate the combined effect of the type of family tie, the skills involved, and the regional context on frm productivity. To address this aim, we analyze a Swedish dataset containing different relational longitudinal population registers from 1995 to 2012. We found a signifcant and positive relationship between family co-occurrence and frm productivity. However, this relationship varies across geographical space and types of familial relationships.

# **Literature review: family co-occurrence and frm productivity**

The presence of family co-occurrence in frms (i.e., the presence of familial relationships inside a frm, including among co-workers and between employees and owners) is a function of the effects of agglomeration economies and hiring practices. Manifold studies have argued that because competition is high in larger regions, only the most productive frms and workers select each other (Bjerke & Mellander, 2017; Combes et al., 2012; Florida, Mellander, Stolarick, & Ross, 2012). Therefore, because the most productive frms and skilled workers tend to be located in larger regions, smaller regions might have relatively fewer productive frms and workers (Combes et al., 2012; Glaeser & Maré, 2001). Due to the spatial sorting of skills into larger regions, frms in smaller regions with thinner labour markets are likely to have alternative hiring strategies. For instance, these frms may resort to hiring through referrals and family networks (Montgomery, 1991), resulting in more and stronger family ties at the frm level. Hence, we expect that the phenomenon of family co-occurrence is more prevalent in smaller and more peripheral regions compared to larger urban regions.

Moreover, the effects of globalization have increased migration patterns, leading to high movement into larger regions because of the availability of job opportunities. This trend has torn families apart and weakened familial relationships. However, because smaller regions facilitate frequent face-toface interactions, social network density among economic actors tends to be relatively higher in smaller regions compared to in larger ones (Lengyel & Eriksson, 2016). Therefore, it is reasonable to expect that in smaller regions, stronger and more trustful familial relationships are likely to be found in frms. Supporting this argument, Holm et al. (2017) found low levels of kinship density in workplaces in Sweden's metropolitan regions, somewhat high levels in intermediate regions (urban regions), and higher levels in remote and sparsely populated areas (rural or small regions). They further showed that kinship density decreases with rising education levels, which means that workers with low education are over-represented in workplaces with high kinship density, a phenomenon that is likely highly associated with smaller and/or rural regions.

Considering the prevalence and importance of family co-occurrence across regions, it is important to further analyze the infuence of family co-occurrence on frm productivity. This potential relationship can be studied through the lenses of learning and agency costs. First, family co-occurrence can enhance localized learning through shared identity and transgenerational knowledge transfer (Wenger, 2000), as well as via joint social arrangements and mutual trust (Boschma, 2005). Since organizational learning involves frm members' ability to create, retain, and share both general and complex knowledge (March, 1991), some level of trust is needed to enhance this process. Promoting kinship or familial relationships among top managers (De Massis, Kotlar, Campopiano, & Cassia, 2015) increases trust-boosting information symmetry between family managers and encourages learning from others' experience. On the other hand, family co-occurrence can impede or weaken frms' learning capacity, which can in turn negatively affect their productivity. Specifcally, a strong family culture can counteract learning by locking family members into a particular way of doing things, hence making them infexible, resistant to change, and inclined to stick to path-dependent traditions (Chirico & Nordqvist, 2010) at the expense of their own innovative and learning capacities (Boschma, 2005). Family co-occurrence can also result in the accumulation of similar and suboptimal knowledge, which can also affect frms' learning capacity and, thus, slow down growth at the frm level (Boschma, Eriksson, & Lindgren, 2009). In other words, the advantages of family co-occurrence are likely to be offset by the low availability of diverse perspectives and knowledge in decision-making processes when family involvement is excessive (De Massis et al., 2015).

 Second, family co-occurrence can also infuence frm productivity by reducing agency costs. An agency cost is incurred when a principal (owner) has to establish appropriate structural mechanisms to monitor agents (managers) or incentives to promote interest alignment. Moreover, since frms are viewed as sets of contracts among different factors of production, we can expect agency costs to differ between a principal and different agents (Jensen & Meckling, 1976). Therefore, frm performance in the form of cost minimization and greater effciency is the outcome of principal-agent relationships involving family agency contracts (Corbetta & Salvato, 2004). For example, when analyzing Italian small and medium size enterprises (SMEs), De Massis et al. (2015) found that family ownership, family involvement, and the share of family members on top management team have an inverted U-shaped effect on frm productivity. The implication is that while family co-occurrence is likely to infuence frm productivity, the effect is non-linear. Additionally, because family agency contracts are based on bonds and sentiments, some argue that they are prone to depart from economic rationality and thus hamper frm performance (Gómez-Mejía, Núñez-Nickell, & Gutierrez, 2001; Schulze et al., 2001).

Apart from the abovementioned general explanations related to the roles of learning and agency costs in the relationship between family cooccurrence and frm productivity, there are two other important aspects to further consider: the types of familial relationships and geographical context. First, the family co-occurrence and frm productivity may also vary depending on the types of familial relationships in frms since different ones represent different types of resources and capabilities in the form of social and human capital (Brannon, Wiklund, & Haynie, 2013). The family as a social group has complex familial relationships with varying levels of trust and solidarity between family members (Hasenzagl, Hatak, & Frank, 2018), which can affect their economic behaviours (Wiklund, Nordqvist, Hellerstedt, & Bird, 2013). The question is whether different types of family cooccurrence affect frm productivity.

Second, while the argument on hiring practices and agglomeration economies seems to suggest a skewed spatial sorting of family co-occurrence in smaller and rural regions (Bjerke & Mellander, 2017; Combes et al., 2012; Florida et al., 2012), there is limited empirical evidence on the relationship between family co-occurrence and frm productivity across different spatial levels. In other words, the effects of family co-occurrence on frm productivity contains a geographical problem—that is, family co-occurrence may lead to varying outcomes depending on the spatial context. The question is whether family co-occurrence is more likely to positively infuence frm productivity in smaller regions than in larger regions since the former are characterized by labour market–matching defciencies and a lack of variety, coupled with the over-representation of family co-occurrence in frms.

# **Study design**

## *Description of data*

To address whether family co-occurrence infuences frm productivity, we explored a matched employer-employee dataset containing different relational longitudinal population registers from Statistics Sweden (SCB). The database offers several advantages. First, it is comprehensive, as it records every family in Sweden. Families are identifed with a unique family identifcation code, which further indicates an individual's position in his or her family. Second, it is longitudinal with annual observations of people and frms, thus allowing us to follow both people and frms over time. Third, it contains a wide range of socioeconomic attributes, such as education. Fourth, the geo-referenced attributes of frms/plants and people make it possible to conduct relational investigations at several spatial levels. The basic unit of analysis in the study is the frm. In the database, the term 'plant' represents a separate economic unit (workplace) of a frm. Thus, we use the terms 'plant' and 'frm' interchangeably because our analysis draws on only single-plant frms. With single-plant frms, it is relatively easy to trace the owner. The data included in our analysis are from 1995 to 2012. The industrial classifcation of the frms was defned by the Swedish Standard Industrial Classifcation 2002 (SNI02). The geographical reference point for the

 *Map 4.1* FA regions (A), the share of family co-occurrence (B), and the share of family frms by region (C).

Source: Adjei, Eriksson, & Lindgren, 2016; Adjei, Eriksson, Lindgren & Holm, 2019.

analysis is the local labour market region. Sweden is divided into 290 municipalities, which are aggregated into 72 local labour market regions called FA regions (see Map 4.1A). The FA regions are based on labour-commuting patterns between municipalities, representing regions where people can live and work without long commuting distances.

# *Variables*

The dependent variable used in this study is *frm labour productivity*, measured at the frm level and defned as per capita value-added. 'Value-added' is a straightforward measure of economic or industrial output since it refects the magnitude of a frm's contribution to the entire economy (Rigby & Essletzbichler, 2002). Other indicators like patents, citations, and innovation indices cannot necessarily provide this information. Specifcally, per capita value-added is an indicator of how a frm utilizes the strengths and skills of its employees; hence, it is an important indicator of business effciency. We calculated frm productivity by frst compensating for the effects of infation and then dividing defated frm value-added by the total number of employees in the frm. Logs of the values were used to reduce the effect of skewness in the data.

### 88 *Evans Korang Adjei and Rikard Eriksson*

Due to the multigenerational nature of the family register, the database provides information on spousal couples (i.e., partners) as well as biological family members, including parents, children, siblings, etc. (the data do not distinguish between biological and adopted children). More importantly, the family register contains unique identifcation codes indicating the different kinds of relationships in each family. The family and workplace identifcation codes enabled us to identify family co-occurrence in the same workplace or frm. We adopted a simple defnition of family based on the SCB records: consanguineous familial relationships (i.e., blood related) and conjugal familial relationships (i.e., marriage) (Brannon et al., 2013).

We defned two groups of family co-occurrence in a workplace. First, we defned *family co-occurrence without a relationship with the frm owner*. This was done by randomly selecting an employee from the employee dataset connected to a frm and subsequently checking whether any of his or her family members were present in the same frm. If none of his or her family members were present in the frm, another random employee was selected, until we encounter a family member in the same frm. We then summed the total number of family members in every frm. This randomized approach of selecting family members eliminated systemic bias by giving all families present in a frm an equal chance of being selected. However, since families are heterogeneous and since we identifed and selected just one family group for every frm, there is a high probability of randomly selecting a family with specifc characteristics (e.g., dysfunctional communication), which could have affected the results.

Second, we defned *family co-occurrence with a relationship with the frm owner or entrepreneur*. We did so by using the frm owner as the hub connecting family members in the frm. With the family identifcation and employment identifcation codes, we linked all family members related to the entrepreneur/owner in the frm. The family members were further grouped based on the type of relationship they had with the entrepreneur (e.g., being his or her spouse, child, or sibling). This approach provided an opportunity to assess the impacts of the family and different familial relationships on frm productivity in family frms.

We controlled for a number of frm-level factors (i.e., frm size, capital intensity, share of higher education) and regional-level factors (i.e., regional size and specialization) known to co-determine productivity (Eriksson & Lindgren, 2009). Since we are also interested in how familial relationships interact with the skills present in frms, we also controlled for the effects of skill variety. We did so using entropy measurement to defne skill variety based on employees' educational background, as frst proposed by Boschma et al. (2009). We calculated the similarity in formal skills (SIM) for each plant as the inverted entropy at the three-digit education level.

In Equation (1), *Pi*  3 is the share of three-digit education categories *i* and *N*<sup>3</sup> is the number of three-digit education categories. A high score means that the frm's in-house **formal** skills or competencies are more similar, which does not promote novelty and productivity. The scores were log-transformed to reduce the effect of the variable's distributional skewness.

$$SIM = \frac{1}{\sum\_{l=1}^{N} P\_l^3 \quad \log\_2 \left[ \frac{1}{P\_l^3} \right]} \tag{1}$$

We calculated relatedness in formal skills (REL) as the weighted sum of <sup>2</sup> entropy at the three-digit level within each two-digit education category. *Pj*  in Equation (3) is the share of two-digit education categories, found by summing the shares of all three-digit education categories belonging to <sup>2</sup> *Si* . Hj in Equation (4) is a weight that controls the degree of similarity within the two-/three-digit education categories. A high score indicates higher inhouse formal skill relatedness, which promotes localized learning processes and productivity.

$$\text{REL} = \sum\_{j=1}^{N^2} P\_j^2 H\_j,\tag{2}$$

where

$$P\_j^2 = \sum\_{l \in S\_l^2} P\_l^3,\tag{3}$$

and

$$H\_j = \sum\_{l=S\_l^2} \frac{P\_l^3}{P\_j^2} \log\_2 \left[ \frac{1}{P\_l^3} \right]. \tag{4}$$

The last entropy variable is unrelatedness in formal skills (UNREL), which we measured at the one-digit education level. *Pi*  1 in Equation (5) is the share of one-digit education categories. A high score indicates higher differences in formal skills (hence, higher unrelatedness), which hinders localized learning and productivity. Due to the de-compositional structure of the variables, research has shown that they do not capture identical features of skill composition (Frenken, Van Oort, & Verburg, 2007).

$$UNREE = \sum\_{l=1}^{N^1} P\_l^{l} \log\_2 \left[ \frac{1}{P\_l^{l}} \right]. \tag{5}$$

#### 90 *Evans Korang Adjei and Rikard Eriksson*

With these sets of skill variety, we empirically tested how familial cooccurrence moderates the effects of similar and unrelated sets of skills based on the argument that social ties can reduce cognitive distance between economic actors over time (Boschma, 2005).

 Additionally, we controlled for the effects of agglomeration economies with two measures. Given the general claim that frms in co-located industries enjoy externalities not available to more isolated frms (Jacobs, 1969; Marshall, 1920), we controlled for the effects of specialization and urbanization, respectively. We resorted to using a location quotient (LQ), which has been used in the regional science literature to capture the effects of specialization. In Equation (6), *Sir* is the degree of specialization in industry *i* in region *r*; *Empir* is the number of employees in two-digit industry *i* in region *r*; *Empr* is the total number of employees in region *r*; *Empi* is the number of employees in two-digit industry *I*; and *Emp* is the total number of employees in Sweden. LQ is a relative measure of the regional share of workers relative to the national average in a specifc industry. If LQ > 1, it implies that the region is more specialized in that industry than the average region. For the effects of general urbanization, we used the log of total regional employment (a representation of employment stock and human capital). We believe that size is a more appropriate measure to capture the effects of urbanization economies than density because of the generally sparsely populated structure of the Swedish economy (Andersson & Klaesson, 2009). Since the regions considered in this analysis are defned on the basis of commuting distances between dwelling places and municipalities, and not administrative borders like municipalities, we argue that size captures regional potential for interactions. We expect frms located in larger and specialized regions to beneft more from knowledge spillovers and therefore also be more productive than those in small or very diverse regions.

$$\text{S}\_{\text{lr}} = \frac{\text{Emp}\_{\text{lr}}}{\text{Emp}\_{\text{l}}} \frac{\text{Emp}\_{\text{r}}}{\text{Emp}}.\tag{6}$$

### **Empirical model**

Considering the panel nature of the data, two models come to mind: a fxed-effects (FE) model and a random-effects (RE) model – see Equation (7) for the general panel data model. Thus, µ*it* is the product of the individual specifc time-invariant effects (*ui*) and the time-varying random component (*vit*) – see Equation (8). The FE and RE models both apply the panel structure differently. That is, the FE model applies the within transformation and controls for unobserved heterogeneity by allowing correlation between the individual specifc error term and the independent variables. The RE model, on the other hand, assumes that variation across observations is random and that there is no correlation between the unique errors and the independent variables. In this case, the Hausman test rejected the hypothesis that there is no correlation between the unique individual specifc error term and the independent variables, suggesting that the FE model is more likely to provide consistent estimates than the RE one. Although the Breusch-Pagan Lagrange Multiplier test suggested that the panel effect is minimal, as shown in the considerably lower within variation values for most of the variables, the FE model provides more consistent estimates and was therefore chosen. In Equation (7), ln ˜*it* is the log average labour productivity (or per capita productivity) in frm *i* at time *t.* 

$$
\ln \gamma\_{it} = a + \beta^i X\_{it} + \mu\_{it}. \tag{7}
$$

$$
\mu\_{ll} = \iota \eta + \mathbf{v}\_{ll}.\tag{8}
$$

The models include time, region, and industry fxed effects to control for unobserved factors not captured in the models (e.g., year-specifc effects, technological differences across industries or sectors, and institutional differences across regions) and cluster-robust standard errors to remedy heteroscedasticity.

# **Results**

### *The geography of family co-occurrence*

Map 4.1 shows the different types of FA regions in Sweden (A); the geography of family co-occurrence in the Swedish economy (B); and the share of family frms, defned as frms in which family co-occurrence is directly linked to the owner (C). On average, family co-occurrence constitutes about 18% of workplace employment and family frms make up 26% of the sampled frms. From Map 4.1B, we observe that the three metropolitan regions (Stockholm in the east, Gothenburg in the west, and Malmö in the south) are all characterized by a relatively low representation of family co-occurrence. As seen in Table 4.1, the dominant dyadic familial relationship (involving the frm owner and other family members) is the parent-child relationship followed by the spousal/partner and the sibling relationships. The dominant gender in the dyadic familial relationships is male, which is particularly evident in all the relationships. Whereas there are no major differences in age distribution across the different familial relationships, some differences in income and education are observed.

Map 4.2 shows the spatial distribution of the different type of familial relationships (familial relationships with frm owners) in family frms across Sweden. The dominance of parent-child and spousal familial relationships

### 92 *Evans Korang Adjei and Rikard Eriksson*


*Table 4.1* Description of the dyadic relationships (family relationships/ co-occurrence)

Source: Adjei et. al., 2019.

*Map 4.2* Regional composition of different familial relationships/co-occurrence. Source: Adjei et al., 2019.

in the workplace corroborate the argument that small family businesses are indeed nuclear-family based. In summary, although scholars have argued that the family has lost its economic signifcance (Hollinger & Haller, 1990; Inglehart & Welzel, 2010), our fndings show that informal family networks still play an important role in frm recruitment even in advanced capitalist societies. This is especially the case in regions with smaller labour market since family hiring may be a cheaper and quicker means of recruitment characterized by higher levels of trust.

### *Family co-occurrence and frm productivity*

In this section, we present the estimations of the relationship between family co-occurrence and frm productivity. Although we controlled for a number of frm and regional factors, we focus on the relationship between family co-occurrence and frm productivity and how the former moderates the effect of regions and different skillsets on the latter. To simplify the presentation of our fndings, we show them in a bar graph for better virtualization of the estimates.1 Due to differences in scale across the variables, we present standardized coeffcients. This, for example, implies that for every increase of 1 standard deviation in family co-occurrence, productivity increases by 0.0117 standard deviation, assuming the other variables are held constant (see Figure 4.1a).

All the variables in Figure 4.1a are signifcant and show the expected effects on productivity. The model shows that there is a small albeit positive association between family co-occurrence and frm productivity. This fnding corroborates regional familiness arguments that social ties and specifc familial relationships can indeed enhance information diffusion and, consequently, frm performance (Basco, 2015). Figure 4.1b further indicates that familial relationships involving children and/or spouses are more likely to positively and signifcantly affect frm productivity than any others. While research has argued that the family is an important strategic resource, this fnding suggests that it is a constellation of different forms of relationships and resources with varying effects on economic frm outcomes. The general positive relationship between family co-occurrence and frm productivity is actually mainly a product of familial relationships involving children and/ or spouses and less so of sibling relationships.

Figure 4.2 focuses on the moderating effects of family co-occurrence. While the effects of family employment on productivity have received signifcant attention, especially in family business studies, spatial differences of the effects have not received that much attention, even in regional science (Adjei, 2018). The model indicates that family co-occurrence is more likely to positively infuence frm productivity in specialized regions than in more diverse ones (Figure 4.2A). Moreover, only familial relationships involving spouses or partners are more likely to affect frm productivity in specialized regions (Figure 4.2B).

Source: Adjei et al., 2016, 2019.

Figure 4.3 shows how the different familial relationships moderate the effects of skill variety on labour productivity. All the interactions except those involving spousal/partner relationships and parent-child relationships are signifcant. These results indicate that parent-child familial relationships abate the negative effects of similarity in formal education on productivity. To some extent, the fndings show that family frms in which the children

 *Figure 4.2* Fixed-effects estimates showing the moderating effects of family cooccurrence (a: without a relationship to the frm owner; b: with a relationship with the frm owner) on the relationship between regional size and specialization and average labour productivity.

As expected, the control variables remain stable with the introduction of the interaction terms. Thus, capital intensity is positive and signifcant, as is human capital. However, the effect of frm size on performance is negative and signifcant, which runs counter to previous fndings. The effect is only valid for smaller frms. Moreover, the regional variables indicate that specialization and regional size enhance frm performance.

Source: Adjei et al., 2016, 2019.

have similar educational levels as their parents are more likely to perform better. The fndings also show that familial relationships involving spouses/ partners are positively correlated with related competencies, and also mitigate the negative impacts of similar and unrelated competencies on labour productivity.

 *Figure 4.3* Fixed-effects estimates showing the moderating effects of family cooccurrence on relationship between skill variety (formal skills) and average labour productivity.

As expected, capital intensity remains positive and signifcant, as does human capital. The effect of frm size on performance is, however, negative and signifcant, which runs counter to previous fndings. The effect is only valid for smaller frms. Concerning in-house formal skill variety, our results are in line with previous studies, showing that similarity in formal skills (SIM) (signifcant and negative) and unrelatedness in formal skills (UNREL) (signifcant and negative) have a negative impact on productivity, while relatedness in formal skills (REL) enhances performance due to potential complementarities that facilitate interactive learning. The regional variables indicate that specialization and regional size enhance frm performance. Source: Adjei et al., 2019.

# **Concluding remarks**

The aim of this chapter was to analyze the effects of family co-occurrence on frm productivity. We looked at two groups of family co-occurrences: (1) general family co-occurrence (i.e., the presence of family members within workplaces) and (2) family frm co-occurrence (presence of family members with relationships with frm owners). This selection of two groups is motivated by the assumption that these varied types of family co-occurrence may provide different environments for learning and alternative incentives to reduce agency costs with corresponding effects on frm productivity.

Based on data on a sample of Swedish SMEs, our results suggest that family co-occurrence is over-represented in workplaces in smaller regions. In other words, we found family co-occurrence to be less common in larger and more diverse regions. This fnding is likely driven by the spatial sorting of skilled individuals and frms (Bjerke & Mellander, 2017; Combes et al., 2012; Florida et al., 2012), such that familial relationships are more prevalent in smaller and relatively thinner labour markets due to labour market imperfections. This fnding shows that family contacts and networks may be an important mechanism for fnding jobs in the Swedish labour market (Korpi, 2001) for two reasons. First, family contacts can compensate for the relative shortage of regional agglomeration economies and variety of skills (Puga, 2010). Second, family contacts and networks can facilitate the transgenerational intention of human capital transmission (Riggio & Saggi, 2015).

Dominant dyadic familial relationships are characterized by the presence of the owner with other family members, such as with children or a partner/ spouse. Other familial relationships, such as relationships with siblings or other family members, are less common. The dominance of parent-child and spousal familial relationships in the workplace corroborates the argument that small family frms are indeed nuclear-family based. For family members co-occurring in the same workplace, we found that female family members have relatively higher education levels than their male counterparts. Common among the discussions on job following and family co-occurrence (i.e., children in the same workplace as their parents) has been job followers' low educational levels. However, our fnding shows that relatively few children in the same workplace as their parents have low education levels. This fnding dispels the argument that children who follow in their parents' career footsteps have lower education levels (Kramarz & Nordström-Skans, 2014), but rather suggest that this job following may be a mechanism for the transmission of specifc forms of human capital (Riggio & Saggi, 2015).

Regarding our main research question, we fnd that family co-occurrence is indeed related to frm productivity. Namely, the mere co-occurrence of family members has a general positive effect on frm productivity. However, the main contribution of this book chapter is the fact that we can show the varying benefts of family co-occurrence across space and types of familial relationships.

First, family co-occurrence is more likely to positively infuence frm productivity in specialized regions compared to more diverse regions. This fnding builds on Gordon and McCann's (2000) argument that the presence of localization economies in larger regions reduces the need for social proximity (e.g., family co-occurrence) by offering more local opportunities to access related knowledge and skills. While social proximity, and family cooccurrence for that matter, may be less relevant in larger specialized regions, our results indicate that social proximity measured as family co-occurrence is indeed associated with higher productivity in specialized regions. This is not surprising because most specialized regions in Sweden are also smaller ones characterized by relatively high shares of family co-occurrence.

### 98 *Evans Korang Adjei and Rikard Eriksson*

Secondly, different types of familial relationships are benefcial in frms that we defned as family frms (i.e., the owner has family members employed the frm). In this case, relationships involving entrepreneurs and their children and/or spouses positively infuence frm productivity. While family cooccurrence in general could be argued to be an important strategic resource, our fndings suggest that the family is a constellation of various forms of relationships that affect economic frm outcomes differently (Melin & Nordqvist, 2007; Wiklund et al., 2013). Our fndings also suggest that in family frms where ownership and management are vested in a family, information symmetry between members is enhanced, which in turn has positive effects on productivity (Corbetta & Salvato, 2004; Cruz, Justo, & De Castro, 2012; De Massis et al., 2015). Our analyses further indicate that family co-occurrence involving spousal relationships can actually abate the negative effects of having an internal skill mix that is too diverse or too similar (cf. Boschma et al., 2009). This fnding is important because it allows us to examine the complementary and substitutional relationships of different types of familial co-occurrence and other factors of production.

In summary, although some have argued that the family has lost its economic signifcance in modern society, our fndings show that not only does the use of informal family network still play an important role in frm recruitment, even in advanced capitalist societies (Ioannides & Datcher, 2004), it also has a corresponding positive effect on frm productivity. In other words, while family co-occurrence in smaller regions may be a cheaper and quicker means of recruitment, especially for family businesses, it also infuences productivity.

### *Contributions*

Our fndings have a number of implications. First, the over-representation of family co-occurrence in workplaces in smaller regions is an indication of the importance of matching frm and regional needs with specifc recruitment practices. It is important to reiterate that research on the spatial sorting of skilled workers and frms has shown that the process leaves smaller regions with less productive workers and frms. Therefore, to overcome this problem, managers of SMEs and/or family frms in smaller regions may rely on familial connections to compensate for the lack of diversity, while simultaneously knowing that the process is relatively cost effective. Though this has previously been described as an act of nepotism, it can actually be seen as a process of compensating for a lack of agglomeration economies. We believe this evidence could further inform academic research on the intentions and recruitment practices of SMEs and/or family frms in smaller regions.

Second, our results show that family co-occurrence positively infuences frm productivity, especially, familial relationships involving entrepreneurs or frm owners and their children and/or spouses. We expect this fnding to inform family business managers the possible trade-offs involving family members. It is evident from our fndings (in line with other research such as Bird, 2014 and Brannon et al., 2013) that certain familial relationships are more important for productivity, sales, and proftability than others. This should further inform family frm managers the need to harness familial idiosyncratic resources to effciently leverage familial labour or resources.

Finally, our fnding that family co-occurrence, particularly relationships involving entrepreneurs and their spouses, abates the negative effects of having too similar or too different skills should be a concern for scholars in both family business and regional science as well as for family business practitioners. For academic researchers, this fnding partially informs previously reported divergence in the effects of skill variety on productivity (Boschma et al., 2009; Östbring & Lindgren, 2013; Timmermans & Boschma, 2014). For family business practitioners or owners, it is imperative for them to know that some familial relationships are important contingencies in shaping the economic effects of skill variety.

### *Future lines of research*

Although we were able to exploit a very rich dataset to further our understanding of the varying productivity effects of different family ties, we encourage further studies to delve more into the potential mechanisms that infuence the varying effects of different familial ties (e.g., different trustladen relationships). For example, the results presented here could at least be partially a consequence of the characteristics of the sample and measurements. The data we used were from a population of SMEs, which may not necessarily be representative of family and non-family frms in general even though the former often tend to be rather small. Future studies could therefore examine the relationship between family co-occurrence and frm productivity using larger frms because the family co-occurrence effect may not be the same across all frm sizes.

Further qualitative studies on hiring practices across regions could also be done to better understand the motives behind different types of family recruitment. Another potential future research avenue could be the extent of recruiting couples. Some of our results could be driven by relationships initiated between co-workers, so the presence and potential effects of recruiting couples could be interesting. This is especially relevant in relation to the increasing diffculties of labour sourcing in small regions in combination with over-heated housing markets in many urban regions, which can make it diffcult for couples to fnd affordable housing when forming a family. Finally, the most frequent dyadic relationship we identifed is between two men (father and son). Understanding the mechanisms behind the gender differences in family co-occurrence is of critical importance for addressing the issue of gender inequality in family frms. Future studies could therefore address the selection of sons and daughters, respectively, in family frms.

# **Acknowledgements**

Parts of this book chapter builds on previously published Open Access (OA) articles in *Regional Studies, Regional Science* (Adjei et al., 2016) and *Entrepreneurship and Regional Development* (Adjei et al., 2019). We acknowledge the insightful comments from the editors and the anonymous referees in the process of preparing and revising this chapter.

# **Note**

1 Full result tables with all variables are available upon request.

# **References**


### 102 *Evans Korang Adjei and Rikard Eriksson*


# **5 Developing digital innovation in family frms**

Evidence from Italian industrial districts

*Stefano Amato, Alessia Patuelli and Nicola Lattanzi* 

# **Introduction**

Since the 1970s, the extraordinary growth of spatial agglomerations of manufacturing small and medium-sized enterprises (SMEs) in Italy has attracted the interest of the scientifc community in an effort to understand the competitiveness of local networks of fexible and specialized frms, as opposed to the vertically integrated Fordist model. With his pioneering studies, Giacomo Becattini restored the concept of the Marshallian Industrial District (MID), defned as "a social and territorial entity characterized by the active presence of a community of people and a group of frms in a natural and historically delimited area or zone" (Becattini, 1992, 62). In MIDs, competitive advantages are embedded in localization externalities related to a pool of qualifed workers, specialized suppliers, localized knowledge, and information (i.e., knowledge spillovers) available in the economic actor networks (Boix & Trullén, 2010). Both market and community logics govern economic activities in MIDs (Chiarvesio, Di Maria, & Micelli, 2010), where competitive dynamics stimulating innovativeness are balanced with trust-based cooperation (Boschma & Lambooy, 2002; Dal Maso & Lattanzi, 2014). District frms are embedded in a homogenous setting, stemming from belonging to the same community and sharing values, practices, and behaviors. This setting enables the creation of a "Marshallian atmosphere" (Belussi & Caldari, 2008), bringing various advantages for frms, including a higher level of productiveness/ effciency (Cainelli & De Liso, 2005; Signorini, 1994), export performance (Belzo-Martìnez, 2006), and innovation capabilities (Muscio, 2006).

While MIDs represent an ideal environment for the faster diffusion of innovations among frms (Cainelli, 2008), there is an intra-district heterogeneity related to the different abilities of local frms within the district (Hervás-Oliver, Albors-Garrigos, Estelles-Miguel, & Boronat-Moll, 2018). Indeed, local frms show different patterns in accessing the available knowledge and information fowing in the *milieu*, which affects the intensity of innovation adoption (Giuliani, 2007). In this perspective, some studies affrm the family status of the frm as a source of intra-district heterogeneity (Pucci, Brumana, Minola, & Zanni, 2017). This is related to the ability of family frms to exploit the territorialized network of knowledge and relationships developed in the district (Cucculelli & Storai, 2015). However, intra-district heterogeneity related to family frms' innovation appears to be understudied in MIDs.

To address the aforementioned research gap, our chapter, drawing upon a sample of 152 manufacturing frms located in fve Italian MIDs, investigates the development of digital innovation among family and non-family frms. In particular, our focus is whether they differ in their approach to digital innovation, focusing on the adoption of technologies related to the fourth industrial revolution, also known as Industry 4.0 (Schwab, 2017). Our results suggest that family frms are more oriented towards Industry 4.0-driven digital innovation than their non-family counterparts. Therefore, our preliminary conclusion, which requires further research in the future, is that while territorialized networks of interconnected frms, such as MIDs, represent a source of dynamic advantages for co-located frms, the family status of the frms seems to matter in explaining the existence of "differential" advantages in the digitalization among them.

The contribution of this book chapter is three-fold. First, to the literature on industrial districts, our study provides new evidence on district frms' innovation performance. However, while a dichotomic approach aimed at ascertaining the performance of frms located in MIDs versus non district frms has traditionally prevailed, our book chapter, focusing on intra-district frm heterogeneity, sheds new light on the existence of specifc characteristics that explain why some particular types of frms beneft more from localization externalities than others. Second, it contributes to family frm studies, in line with the latest research efforts intended to include the physical and social-spatial contexts (e.g., Basco, 2015; Baù et al., 2018), our study offers new insights on family frms and MIDs recognizing that family frm specifcities matter in industrial districts contexts for adopting digital innovation. Finally, we contribute to digital innovation literature, with a specifc interest in Industry 4.0, which is emerging as a promising research topic (Arnold, Kiel, & Voigt, 2016; Schmidt et al., 2015). This chapter also makes practical contributions by uncovering the importance of frms' family nature when public policies (e.g., the Italian Industry 4.0 plan) aimed at improving the competitiveness of enterprises, local production systems, and regions are tailored.

This chapter is structured as follows. First, we review the literature on industrial districts, family frms, and innovation. Subsequently, we introduce our methodology, illustrating the study design. Finally, we present and discuss our results, concluding with fnal remarks, contributions, and future research lines.

# **Theoretical background**

# *Digital innovation in MIDs*

Innovation, generally regarded as the introduction of a technical or organizational novelty within a frm, a new idea, or behavior (Schumpeter, 1934), is a "genetic ability" of MIDs resulting from the combined effects of local rivalry, interaction, and cooperation, which together favor the dissemination of non-codifed knowledge and information among district frms (Cainelli, 2008). The socialization processes occurring in the local *milieu*, backed by the intra-regional mobility of human resources, dense informal local networks, and a common cultural and institutional setting, ease the spontaneous exchange of innovation-relevant knowledge and information (Boschma & Lambooy, 2002). In this perspective, Boix and Galletto (2009) coined the term "I-district effect", which indicates the existence of a higher innovative intensity in MIDs compared to non-district areas. Current evidence shows a positive association between frms' localization in MIDs and their innovation performance in terms of product innovation (Muscio, 2006) and number of patents (Boix & Trullén, 2010; Santarelli, 2004), with an above-average rate of innovation both in times of economic stability (Boix & Galletto, 2009) and during adverse conditions (Boix, Galletto, & Sforzi, 2019).

MIDs are currently experiencing deep structural transformations, driven by the consequences of the international fnancial crisis (De Marchi, Lee, & Gereff, 2014), the gradual integration of district frms in the global value chain (Chiarvesio, Di Maria, & Micelli, 2010), the emergence of leading frms with asymmetric market and fnancial power (Randelli & Boschma, 2012), and the rise of new digital industrial technologies related to the fourth industrial revolution, generally known as Industry 4.0 (Bellandi, De Propris, & Santini, 2019a).

Today, Industry 4.0 is emerging as a new type of digital innovation (Kang et al., 2016) that is establishing the premises for a manufacturing "renaissance" and socioeconomic development (Bellandi, Lombardi, & Santini, 2019b). It is aimed at transforming frms' value creation and business models with digitalization, automation, and robotics (Götz & Jankowska, 2017), ultimately changing the traditional source of competitiveness of both frms and regions (Schwab, 2017). Technologies related to Industry 4.0 include simulation, augmented reality, robots, Internet of Things (IoT), cloud services, cybersecurity, additive manufacturing, horizontal and vertical system integration, Big Data, and analytics (Wang & Wang, 2016). MIDs, traditionally characterized by a strong manufacturing specialization (Becchetti, De Panizza, & Oropallo, 2007), perceive digital technologies related to Industry 4.0 as both threats and opportunities (Bellandi et al., 2019b). However, how and to what extent district frms adjust their structures to the new technological paradigm will depend on the combined efforts of local institutions and public policy initiatives,1 as well as frms' specifc characteristics.

 In terms of frms' specifc characteristics, as a source of intra-district heterogeneity (Cucculelli & Storai, 2015; Pucci et al., 2017), the family status of the frm might be one reason that explains different levels of digital innovation adoption in MIDs. The lack of research on family frms in MIDs is paradoxical since MIDs represent a long-lasting localization (Belussi & Caldari, 2008) and a natural socio-spatial basin of entrepreneurial families. In these contexts, the phenomenon of family in business renews itself across generations, becoming a pervasive institution for creating social structures that drive MIDs' evolution (Randelli & Boschma, 2012). In MIDs, family frms generally show a superior position in inter-organizational and inter-personal networks (Cucculelli & Storai, 2015), which may result in a frm-specifc capability in intercepting and exploiting a knowledge-rich local environment (Lattanzi, 2017). These specifcities, in turn, can affect the extent to which family frms belonging to industrial districts engage in new digital and technological innovations as compared to their non-family counterparts.

# *Family frms and digital innovation in MIDs*

Family frms are a widespread phenomenon across Europe. While they are the most pervasive form of organization among all OECD nations and European nations in general, accounting respectively for 85% and 70–80% of all companies (Alberti & Pizzurno, 2013; Mandl, 2008); these percentages are even higher in Italy. Family frms are the backbone of both Italian MIDs (Intesa San Paolo, 2018) and the national economy – more than 85% of enterprises are family frms, and they generate 70% of total employment (AIDAF, 2018). Family frms show specifc behaviors that differ from non-family frms in many ways (Alberti & Pizzurno, 2013). The core reason of their peculiar behaviors resides in the so-called "familiness", a unique characteristic that can lead to certain synergies, advantages, and disadvantages (Pearson, Carr, & Shaw, 2008). Familiness can infuence a number of aspects, including goals (Tagiuri & Davis, 1992), corporate governance (Golinelli, 2000; Randoy, Jenssen, & Goel, 2003), fnancial structure (López-Gracia & Sánchez-Andújar, 2007), and entrepreneurship orientation (Zahra & Sharma, 2004). It can also infuence innovation in different ways (Calabrò et al., 2019). On one side, family frms might be more conservative, less entrepreneurial, rigid, risk-averse, more willing to keep control, resistant to change, and more reluctant to pioneering new products, processes, and markets (Kraus, Pohjola, & Koponen, 2012). They also usually have more limited access to capital markets and may be less eager to grow (Craig, Pohjola, Kraus, & Jensen, 2014). On the other hand, family frms typically have a long-term orientation and the capacity to involve multiple generations, which may affect innovation dynamics (Nieto, Santamaria, & Fernandez, 2015). Overall, family frms' equilibrium of internal forces can lead to fostering or limiting innovation (Ingram, Lewis, Barton, & Gartner, 2016).

Having said that, it is not the family condition of the frm per se that determines its innovativeness orientation but the unique connections that frms are able to establish with their territory or place (Smith, 2016). In particular, family frms are regarded as having tight links with their territory, with a strong interdependence between their economic activity and their place. There is an interactive relationship with the *milieu* and a feeling of identity and attachment of family members to the place (Kim, Haider, Wu, & Dou, 2019).

Family frms are generally committed to renewing and reshaping their social interactions, both within and outside the family domain, as a way to access valuable resources (Salvato & Melin, 2008), hence differing in the way they interact with their surroundings (Basco, 2015). Even though the literature on family frms and innovation is growing, only recently have scholars begun to explore how these peculiar organizations behave when located in MIDs. For instance, by drawing on a large sample of Italian manufacturing district frms, Cucculelli and Storai (2015) show how the advantage of being located in MIDs (i.e., district effect) is dependent upon the size of family frms with only the medium-sized ones being able to leverage the localization benefts of MIDs, refected in a level of higher proftability when compared with non-district frms. With regard to the innovation performance in geographical clusters of frms, Pucci et al. (2017) show how family frms are better able to leverage the localized network of relationships, positively affecting their innovation capability.

The specifc traits of family frms appear to be well-suited for MIDs (Cucculelli & Storai, 2015). Social capital arises as a critical asset in explaining how local knowledge and innovation-relevant information is gathered, regenerated, and shared among district frms (Malecki, 2012). Conceived as a relational asset based on trust, norms, and reciprocity (Coleman, 1988), social capital alludes to the development and exploitation of social ties among actors in a local network (Lambooy, 2010). In MIDs, it is both a "lubricant" for frms' organizational decisions and a "glue" for the local production system as a whole (Bertolini & Giovannetti, 2006).

Family frms are generally endowed with frm-specifc social capital (Zahra, 2010) based on interdependence and trustworthiness among family members (i.e., "internal social capital"), which tends to be replicated outside the organization (Arregle, Hitt, Sirmon, & Very, 2007) (i.e., "external social capital"). Family members show a certain degree of closure and centrality in local entrepreneurial networks, resulting in social and professional ties strengthened by trust (Anderson, Jack, & Dodd, 2005). This is particularly true in MIDs, where there are blurred boundaries between informal and formal networks (Chetty & Agndal, 2008). Therefore, family frms located in MIDs are more likely to show a differential advantage than non-family frms (Hess, 2004).

Indeed, while geographical proximity favors contacts and facilitates the exchange of tacit knowledge (Boschma, 2005), the inherent willingness of family frms to establish socially proximate relationships with local actors reduces the risk of opportunistic behavior, enhancing interactive learning and innovation capabilities (Baù et al., 2018). Additionally, while shared values, norms, and agreements mediate interactions within MIDs (Dei Ottati, 2002), family frms play a crucial role in the sedimentation of local institutional mechanisms (Raco, 1999). Hence, a higher level of institutional proximity of family frms facilitates the access to and the transfer of tacit knowledge in district networks (Soleimanof, Rutherford, & Webb, 2018).

### 108 *Stefano Amato et al.*

Finally, the high level of specialization of MIDs results in a common knowledge base that enables mutual understanding among local actors. The local embeddedness of family frms results in a shared language and domain of a district-specifc knowledge, continuously transmitted to generations, which facilitates effective communication and interactive learning needed to successfully engage in innovative activities (Anselmi & Lattanzi, 2016).

Based on the above-mentioned arguments, we infer the existence of intra-district heterogeneity based on frms' family status<sup>2</sup> with an impact on frms' ability to innovate. Firm-specifc social capital, as refected in the centrality and closure in the local networks, and strong territorial embeddedness enable family frms to alter the social and economic relationships underlying the dissemination of knowledge and information in MIDs, which is relevant for digital innovation adoption and exploitation. Thus, we propose the following hypothesis:

Hypothesis*: The pace of digital innovation in MIDs is higher for family frms than for their non-family counterparts* 

# **Empirical design**

To test our hypothesis, we relied on a dataset from a survey carried out by the research unit of Intesa Sanpaolo (Direzione Studi e Ricerche Intesa Sanpaolo).3 The survey was aimed at capturing information on the strategies, innovation patterns, international trade, ownership, and management structures of a sample of manufacturing frms located in industrial districts across three Italian regions: Piedmont, Tuscany, and Veneto.4 In particular, the leather and jewelry districts of Arzignano, Santa Croce sull'Arno, Vicenza, Arezzo, and Valenza were chosen.5 Firms were identifed using the ATECO code<sup>6</sup> and the province of operations. The initial sample included 584 frms. Those in liquidation or non-active ones were not included. The survey was sent to each frm between October 2018 and February 2019. Reminders were sent in seven instances, approximately 14 days apart from each other. Eventually, 158 frms completed the questionnaire with a response rate of 27%. It is worth noting that selection bias might have occurred as data collection was part of a wider project. However, pure random samples are diffcult to fnd in family frm research, as a national family frm database does not exist (Beck, Janssens, Debruyne, & Lommelen, 2011).

Additionally, survey-based information was complemented with accounting data retrieved from Aida-Bureau Van Dijk, a database containing the historical fnancial and commercial data from approximately 540,000 companies operating in Italy. After removing companies not included in Aida, we obtained a fnal dataset consisting of 152 companies, the width of which is comparable to others used in family business research (Beck et al., 2011; Cucculelli, Le Breton-Miller, & Miller, 2016).


#### *Table 5.1* Description of variables

*Province* 

Table 5.1 shows the variables employed in our study. The dependent variable (*Industry 4.0*) measures the degree to which frms are committed to digital innovation, with a specifc reference to investments in new digital technologies related to the fourth industrial revolution. As constructs to measure the development of digital innovations related to Industry 4.0 are still scarce, we adapted one item from the Kellermanns, Eddleston, Sarathy, and Murphy (2012) innovation scale,7 based on a fve-point Likert scale, where 1 is a strongly negative and 5 a strongly positive commitment to frm's digitalization.

Tuscany and Veneto

Province in which frm *i* is headquartered: Piedmont,

Our variable of interest is the family nature of the frm (*Family frm*). As the defnition of family frm is a debated issue, we followed the so-called "demographic approach" to identifying them, which posits the mere involvement of the family members in the frm – ownership, control, or management – as a suffcient condition for capturing the infuence of the family on the frm (Basco, 2013). Accordingly, we defned it as a family frm if two not mutually exclusive conditions occur: (i) the majority of decision-making rights are in the possession, both directly and indirectly, of either the entrepreneur who established the frm or persons linked to the founder by family relationships; and, (ii) two or more family members sit on the board of directors of the frm.8 Finally, we controlled for other frm-level characteristics that may affect frms' commitment to digitalization.

With the purpose of investigating the association between the family nature of the frm and digital innovation in the context of industrial districts, we estimate the following model:

*Industry* 4.0*i* = *α*0 + *β1Family frmi*+ *β*2*Sizet* + *β*3*Agei* + *β*4*Human Capitali* + *β*5*Export intensityi* + *β*6*Absorptive capacityi* + *β*7*Product innovationi* + *β*8*Process innovationi* + *β*9*Organizational innovationi* + *β*10*Foreing sharei* + *γIi* + *δPi* + *ε<sup>i</sup>*

where *α*0 is the constant, *β*1 is the coeffcient of our variable of interest, *Ii*  and *Pi* are respectively the industry-specifc and province-specifc dummy effects, *γ* and *δ* are the vectors corresponding to the coeffcients, and *εi* is the error term. We address heteroscedasticity concerns in our estimations by computing heteroscedasticity-robust standard errors.

# **Findings**

The descriptive statistics and pairwise Pearson correlation are reported in Tables 5.2 and 5.3, respectively. In particular, Panel 2A shows the descriptive statistics for the whole sample. Family frms account for 42% of the total sample, corresponding to 64 of the 152 frms. On average, sampled frms are 30 years old, export more than 40% of their sales, and are overwhelmingly


*Table 5.2* Descriptive statistics

*Panel 5.2A* Descriptive statistics for the whole sample

LExpressed in natural logarithm.


*Panel 5.2B* Difference of means and Wilcoxon rank-sum test

 LExpressed in natural logarithm. WWinsor at 1% and 99% tail. Level of statistical signifcance + *p* < 0.10, \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001. <sup>a</sup> The Wilcoxon rank-sum test analyses whether the two samples are from different distributions (Sample 1: Non-family frms; Sample 2: Family-frms).

held by domestic owners. For a more straightforward depiction of the difference between family and non-family frms, Panel 2B reports the mean of the variables grouped by the nature of the frms along with the results of a test for differences in the means and the results of Wilcoxon rank-sum tests. The results show that family frms are more committed to digital innovation than their non-family counterparts (2.781 *versus* 2.227, *p* < 0.05). Additionally, family frms are smaller, older, and more internationalized, and these differences are statistically signifcant. With regard to the innovation inputs, the descriptive statistics show that family frms devote more efforts in R&D activities, considering the respective sales levels, than their non-family counterparts (7.796 *versus* 5.886, *p* < 0.10). An analysis of the variance infation factors (VIFs), shown in Table 5.3, rules out multicollinearity concerns in our data; all the VIF coeffcients are below the generally accepted threshold of 10 (Bird & Wennberg, 2014).

 The results of the econometric analysis are reported in Table 5.4. We estimated our coeffcients by using the ordinary least squares (OLS) method. Overall, 21% of the variance of the response variable is explained by the covariates included in the model. Among the control variables, the coeffcient of *Size* is positive and statistically signifcant (*β* = 0.222, *p* < 0.001), indicating that larger frms are in a better position to sustain investments in Industry 4.0. Additionally, the frm's digitalization appears to be contingent on the innovative results achieved by the district frms. In fact, Table 5.4 shows a positive association between *Product innovation* and *Process innovation* and the adoption of Industry 4.0 technologies. Surprisingly, the ratio of graduated human



 *Table 5.4* Regression results

 Note: Robust standard errors are reported in parentheses. Level of statistical signifcance +*p* < 0.10, \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001.

resources has an adverse effect on a frm's digitalization. In fact, the coeffcient of *Human capital* is negative and statistically signifcant (*β* = −0.023, *p* < 0.001). The remaining controlling variables reveal that neither in-house R&D activities (i.e., *R&D intensity*) nor the degree of internationalization (i.e., *Export intensity*) affect the adoption of digital innovations.

With regard to the explanatory variable, our fndings show that within industrial districts, all things being equal, family frms pursue digital innovation more intensively than non-family frms, as refected in the adoption of technologies related to Industry 4.0. Indeed, the coeffcient of our variable of interest (*Family frm*) is positive and statistically signifcant (*β* = 0.456, *p* < 0.05), hence providing support for our hypothesis.

# **Final remarks**

The contemporary industrial revolution, which encompasses the digitalization of manufacturing through the adoption of Industry 4.0 technologies, is initiating an unparalleled transition to new ways of production, business models, modes of value creation, and distribution systems. Local production systems such as MIDs, characterized by a strong manufacturing specialization, are highly exposed to the new technological paradigm, which is a source of threats and opportunities for industrial district frms. While localization in MIDs favors increasing returns and superior technological performance because of the potential access to localization externalities, some frms are in a better position than others to leverage the industrial district's advantages. In particular, the concept of "intra-district heterogeneity" has emerged as a new research area aimed at understanding which frm-specifc characteristics explain some frms' unique abilities to exploit district knowledge and information. Among the sources of heterogeneity, the family status of the frm has been historically overlooked by the industrial district and cluster literature.

To address the aforementioned research gap, we draw on survey microdata of Italian frms located in fve MIDs to explore whether family frms develop more digital innovations related to the fourth industrial revolution (Industry 4.0) than their non-family counterparts. Our fndings reveal that family frms are more prone to adopt digital innovations related to Industry 4.0. Hence, compared to their non-family counterparts, family frms are more actively preparing to compete in the new technological wave, playing a crucial role in the transformation of MIDs. This research shows that the distinctive characteristics of family frms adapt well to the peculiarities of industrial districts, where the social dimension of economic activities (a set of informal, trustbased, and reciprocal relationships regulated by a shared system of cultural values) distinguishes MIDs from other types of local production systems (e.g., business cluster). Due to the rich social capital and high degree of embeddedness in the local network, family frms are in a unique position to intercept and exploit the relevant knowledge and information that fows freely in the industrial district. Hence, while the localization in MIDs is a source of both static and dynamic advantage for co-located frms, family frms appear to beneft more from externalities and the resulting localization economies, as refected in a higher pace of digital innovation than their non-family counterparts.

This study makes several contributions. First, for the industrial district and cluster literature, we shed new light on the digital innovation taking place inside MIDs. However, different from the prevailing approach aimed at measuring the so-called "I-MID effect" (i.e., the existence of a dynamic effciency in the form of a positive innovative differential compared to non-district frms), we provide new evidence on so-called "intra-district heterogeneity," highlighting whether and to what extent some frm-specifc characteristics account for differential advantages. Testing the family status of the frm as a source of intra-district heterogeneity, we extend the results of Hervás-Oliver, Sempere Ripoll, Estelles-Miguel, and Rojas (2019), who found how frms with higher absorptive capacity are able to exploit a knowledge-abundant milieu*,* such as MIDs.

Second, we contribute to the family business innovation literature. We explore family frms' development of digital innovation with the adoption of Industry 4.0-related technologies. Whereas Industry 4.0 is receiving increasing interest in Europe (Schmidt et al., 2015), research is just starting to explore the topic (Arnold et al., 2016), showing that SMEs are not yet fully exploiting the advantages of Industry 4.0 (Moeuf et al., 2018). However, our chapter shows some evidence that family frms are more inclined towards such transformation. Additionally, we contribute to the growing efforts geared towards the inclusion of the physical-relational space for the comprehension of family frms' behavior and performance (e.g., Basco, 2015; Stough et al., 2015). In this perspective, while the interplay between family and business domains has emerged as the traditional lens to explain the distinctiveness of family frms towards innovation, the incorporation of the spatial dimension – that is, the frm's localization in MIDs – unveils further valuable insights for the comprehension of family business innovation. Even though MIDs are a source of dynamic advantages for co-located frms, the family frm's specifc social capital and its superior position (i.e., embeddedness) in the territorialized network of relationships confer a differential ability to beneft from agglomeration economies.

Our study has relevant implications for policy makers. Industrial districts, as the backbone of Italian manufacturing systems and a source of regional competitiveness, are experiencing deep changes. Driven by the consequences of the international fnancial crisis, these include the growing reliance on outsourcing, the integration in the global value chain, the emergence of vertically-integrated frms, and new digital innovations, which, together, are reducing industrial districts' internal cohesion. Therefore, the fourth industrial revolution may represent an opportunity for the upgrading and "revitalization" of district frms and areas as a whole. In this light, national governments across Europe, including Italy, have been promoting measures for all enterprises to support investments in digital transformation. Since our chapter unveils the distinct reaction of family frms located in industrial districts to this new way of digital-driven disruption, the design and evaluation of the effectiveness of these policy initiatives should consider family frms as peculiar actors for the development strategies of local and regional economies. To this end, it is worth noting the peculiar criticalities of family frms that undermine their competitiveness and survivability – such as ownership and leadership transition, the need of professionalization, and cultural rigidity – which thus necessitate tailored-made policies (Basco & Bartkevičiūtė, 2016).

This study has some limitations that pave the way for future research. As our data pertains to only fve Italian MIDs specialized in two industries (i.e., leather and jewelry), the possibility of generalizing the fndings is rather limited. Hence, future research should explore the digital transformation of family and non-family frms across multiple heterogeneous MIDs and across a wider time span (i.e., longitudinal studies). Additionally, comparative studies of MIDs with different levels of technological sophistication may unveil specifc patterns of innovation among family and non-family frms. In this regard, the nexus of family-frms and MIDs is a promising research area worth exploring by means of qualitative methods as well, which may complement quantitative ones to investigate the organizational and socio-spatial dynamics taking place in the district areas. Social capital and spatial embeddedness theories could contribute to efforts to interpret such phenomena. Since the district effect is also related to static advantages, as refected in higher levels of productivity/effciency compared to frms located in non-district areas, future research should consider frms' family status to explore the existence of differential fnancial advantages in addition to those related to innovation. Finally, future research could also investigate whether and to what extent family frms, as compared to their non-family counterparts, have contributed to the resilience of MIDs in the face of the global fnancial crisis of 2007–2008 and to their structural transformations such as the growing integration of district frms in the global value chain.

# **Acknowledgments**

We would like to thank Giovanni Foresti and Sara Giusti of Intesa Sanpaolo Research Unit (Direzione Studi e Ricerche Intesa Sanpaolo) as part of the joint research program with IMT School for Advanced Studies Lucca.

# **Notes**


National Institute of Statistics) carried out in 2011, Italy counts 141 industrial districts located mostly in the northeast of the country (45) and specialized in the sectors related to the so-called "Made in Italy" initiative: mechanical industry (27.0%), clothing/textiles (22.7%), furniture (17.0%), and leather and shoes (12.1%). For further information, please refer to ISTAT (2011).


### **References**

 

AIDAF. (2018). *Dieci anni di capitalismo familiare*. *X osservatorio AUB.* 


# **Part III Meso-Foundation channels**

# **6 Family frms and their regional ties**  A bond made for the future?

*Regina Lenz* 

### **Introduction**

The overlap between family and business represents a common characteristic for a large number of organisations across the world. Family frms are those organisations in which family is involved in the ownership and management of the frm, with the intention to keep the business within the family across generations (Mandl, 2008; Miller, Le Breton-Miller, Lester, & Cannella, 2007). The family component distinguishes these frms from any other type, with regard to the way families manage and govern, make decisions, allocate resources, and interact with their regional contexts (Basco, 2017; Soleimanof, Rutherford, & Webb, 2018; Stough et al., 2015).

The unique interaction between family frms and their environment is infuenced by the family's focus on frm longevity, and their decision-making is led not solely by economic principles, but also by the strong personal commitment of owner-managers and their feelings of responsibility toward their employees and home region (Zellweger, Eddleston, & Kellermanns, 2010). Research agrees that it is this "regional familiness" (Basco, 2015) or "community citizenship" (Baù et al., 2019) of family frms that leads ownermanagers to build long-term relationships with local stakeholders, which forges a particularly strong and mutually benefcial bond between family frms and the regions in which they are embedded (Baù et al., 2019; Bird & Wennberg, 2014). Additionally, since family frms are generally considered to be more conservative and risk-averse than non-family ones, and thus more focused on their home market than international ones (Gomez-Mejia, Makri, & Larraza Kintana, 2010; Hamelin, 2013), their regional commitment seems to be durable and uncontested.

However, the commitment of family frms to their regional context may vary across generations. For instance, in their study of Spanish manufacturing frms, Fernández and Nieto (2005) showed that second or subsequent generation family frms are more involved in international markets than frst-generation ones. Considering the potential differences in attitudes, interests, and qualifcations between incumbent owner-managers and their successors, the research question arises as to whether the strong regional

### 126 *Regina Lenz*

bond of family frms persists across generations. Given that every family frm ultimately has to face the challenge of transferring ownership and management to the next generation and that the number of family frms looking for successors has been rising continuously across Europe in the last few years (Duh, 2012; KPMG, 2019), I wonder how succession affects the regional anchoring of family frms.

This chapter adopts a dynamic perspective and analyses the connectedness between successor generations and their regional context. By using a qualitative approach and through the lens of an institutional perspective, I explore case studies of successors in Basque family frms and analyse differences in the attitudes and behaviours between predecessors and successors. I observe that these differences lead to successors having weaker regional embedding than their predecessors. This weak regional embedding shows that transgenerational family frms are immersed in a regional disembedding process that could have positive or negative consequences for regional economic development.

# **Family frms and their regional and institutional ties**

When analysing the linkages and mutual effects between context and family frms, research usually looks at the embeddedness of family frms in terms of the geographically constrained area where they are situated. However, beyond the spatial characteristics of context, relational aspects play a key role in infuencing frm behaviour over time.

# *Family frms and their regional context*

The literature usually highlights the positive effects of the links between family frms and their regional contexts (Gomez-Mejia, Basco, Müller, & Gonzalez, 2020). The assumption is that local embeddedness – "the involvement of economic actors in a geographically bound social structure" (Baù et al., 2019, 360) – favours contacts with customers and suppliers in a geographical area and thus facilitates access to both tangible and intangible assets such as a skilled labour force and localised knowledge, respectively (Baù et al., 2019). Existing research can be divided into two groups, depending on whether they analyse the effect of the region on family frms or, vice versa.

In terms of family frm contribution to regional economies, existing research has shown that family frms are conducive to economic and social development in their regional contexts in several ways. For instance, by contributing to the production and export of goods and services; providing jobs, professional training, and qualifcations; and engaging in regional philanthropy (Basco, 2015; Bjuggren, Johansson, & Sjögren, 2011; Glückler & Suarsana, 2014; Memili, Fang, Chrisman, & De Massis, 2015). Across Europe, family frms constitute the economic backbone of regional economies: 85% of companies are family-owned, generating 70% of Europe's GDP and employing 60% of its workforce (Ernst&Young, 2014). Since family frms are often situated in peripheral areas, they also provide job opportunities in rural regions and thus act as a counterweight to the ongoing trend of urbanisation (Gottschalk et al., 2017). Taking these benefts into account, family frms' transgenerational continuity and regional embeddedness are essential for long-term regional stability in terms of employment, innovation, and the regional use of profts for investments and societal good (Daspit, Holt, Chrisman, & Long, 2016).

Studies looking at the effect of the regional context on family frms are usually based on the assumption that "family businesses are more embedded within the regional community than their non-family counterparts," and that this affects their strategic choices (Bird & Wennberg, 2014, 424). Research shows that by using their local connections, family frms enhance their socio-emotional wealth, gain localised competitive advantages, and beneft more from local embeddedness than non-family frms, especially in rural areas (Backman & Palmberg, 2015; Baù et al., 2019). Baù et al. (2019) even claim that being embedded in a rural context with tight social networks can counterbalance the potential negative effects that family involvement can have on frm growth. Along the same lines, Soleimanof et al. (2018) argue that it is precisely the embeddedness of family frms that makes them less dependent on the formal features and procedures of their regional contexts because they can access family resources, such as labour and capital, and use their personal connections to draw on social capital. Accordingly, family frms are able to thrive in resource-scarce and objectively less favourable regions as well (Bird & Wennberg, 2014). This evidences the importance of social relationships for the regional anchoring of family frms.

### *The importance of relationships, institutions, and time*

To understand what makes a frm behave the way it does, not only must we consider a region's available assets, but also the structures of social relations in which agents are embedded and that infuence their decisions and actions. The theory of embeddedness posits that the economic activities of individuals and organisations are determined by a system of relationship ties that facilitate or inhibit certain economic actions (Granovetter, 1985). Thus, an actor's behaviour is not always rational and proft-maximising oriented, but also informed by past experiences and institutionalised behavioural expectations (Glückler & Bathelt, 2017). Similarly, institutional theory postulates that organisations not only react to economic, market, and technical pressures but also to the social and institutional pressures of their surrounding environment, comprised of other organisations (Coraiola & Suddaby, 2018; DiMaggio & Powell, 1983; Suddaby & Greenwood, 2009). In this respect, the concept of the organisational feld highlights that these pressures and normative expectations may not only originate from those organisations in the actual geographical proximity, but also from "those organizations that, in the aggregate, constitute a recognized area of institutional life" (DiMaggio &

### 128 *Regina Lenz*

Powell, 1983, 148). Thus, by interacting more frequently with each other than with actors outside the feld, organisations within a feld share common cognitions, meaning systems, and institutions that act as guidelines for their actions.

Defned as relatively stable interaction patterns that are based on mutual normative expectations on how to behave legitimately (Glückler & Bathelt, 2017), institutions are hard to identify in real-life interactions. However, the concept of relational infrastructure enables making institutions visible, based on the idea that "because institutions are means of providing regularity to behaviour, they are always underpinned by networks of persons" (Storper, 2018, 213). Although institutions are not reducible to their networks, it is these inter-personal and inter-organisational linkages that enable the communication and alignment of social norms and legitimate beliefs, and help visualise the establishment of corresponding interaction patterns as observable manifestations of institutions (Glückler, Punstein, Wuttke, & Kirchner, 2020; Storper, 2018). For example, Storper (2018) found that the more successful development of the San Francisco Bay Area compared to Greater Los Angeles was due to the former's relational infrastructure, comprised of informal networks, elite leadership networks, cross-network connections, and organisational sites that facilitate and sustain such crossover links. In applying the framework of relational infrastructure, Glückler et al. (2020) were also able to explain the otherwise inexplicable economic success of the German region of Heilbronn-Franconia, based on strong frm-internal bonds, disconnection in terms of work collaboration between frms, but cohesive non-business networks within the home region.

Thus, research so far has focused on identifying the specifc relational infrastructures of regions with the purpose of understanding their regional economic performance at a given point in time. Despite these valuable insights, the role of time has received less academic attention (James et al., 2020; Sharma, Salvato, & Reay, 2014). This chapter, therefore, further elaborates on the notion of relational infrastructure by adding a dynamic perspective and by exploring how established relational infrastructures may change over time. The regional connections of family frms and their potential changes in the course of successions are well-suited units of analysis in this respect. In this sense, the research questions are as follows: *How does succession affect the regional connectedness of family frms? Do successors have different relational infrastructures than their predecessors due to different characteristics and experiences?* 

# **Methodology**

In order to analyse the effect of succession on the regional embeddedness of family frms, this research is based on qualitative case studies of second or subsequent-generation family frms in the Spanish Basque Country, a region that champions regional distinctiveness and cohesion.

### *The study region: many family frms and strong regional distinctiveness*

The Basque Country1 is characterised by successful and resilient economic development, a strong industrial base, and an extensive number of family frms with an increasing need to fnd successors (El Foro de EmpresaFamiliar, 2008). Family frms make up 70% of all enterprises in the region and account for two-thirds of its GDP and 80% of its workplaces (Christensen Zaracho et al., 2008). But not only does the Basque Country boast a high number of family frms, it also enjoys a pronounced regional autonomy and community, expressed through its own regional government since 1980, fnancial autonomy since 1981, and its own regional language, Euskera (Moso & Olazaran, 2002). Additionally, the Basque Country features a number of community-building institutions such as the practice of regularly meeting strongly connected childhood friends (*cuadrilla*) in the streets before dinner (*txikiteo*) or cooking together in a gastronomic society (*txoko*) (Hess, 2007; Ramírez Goicoechea, 1984).

Previous research has showed that Basque family entrepreneurs actively participate in these cultural activities together with their local community (Lenz and Glückler, 2020). Thus, the relational infrastructure of traditional Basque family frms can be described similarly to what Glückler et al. (2020) found for the region of Heilbronn-Franconia: little connection between frms in terms of collaboration, but a cohesive non-business network within the local community, and strong intra-frm relations due to pronounced family logics as guiding principles.

### *Data and methods: qualitative case study*

In order to see if these characteristics also describe the relational infrastructures of Basque family frm successors, between February 2015 and May 2016, a total of 28 interviews were conducted with representatives of 16 family frms of the automotive and machine-producing sector. In each of the 16 cases, interviews were carried out with the successor and new ownermanager of the family frm, and, as far as possible, with his or her predecessor and/or a longstanding employee of the frm, in order to further validate successors' statements (see Table 6.1).


*Table 6.1* Overview of interviews conducted

### 130 *Regina Lenz*

To learn from their experiences, I only selected those frms that had already completed succession within the last few years or were just in the process of handing the frm over to the next generation (see Table 6.2). The majority of cases analyse intra-family succession from the frst to the second generation, and the predecessors were still either involved in business decisions or available as consultants, which made it easier for the successors to directly compare their expectations, attitudes, and ways of doing business with those of their predecessors. Three of the family frms in the sample turned into cooperatives, with the employees of the frm becoming shareholders, and only one frm was sold to external investors (non-family members).

The semi-structured interview guide scanned for organisational and personal facts such as education and work experience, and focused on the changes the successors had planned or already implemented within the frm and in their relationships with other frms, as well as on their ways and attitudes toward running the family frm. The interviews were done in Spanish and lasted between 60 and 120 minutes. They were recorded, transcribed, and translated. Afterwards, I coded and analysed the data using the MAXQDA software in a bottom-up, inductive approach, looking for patterns emerging from the data and across cases (Reay & Jones, 2016).


*Table 6.2* Detailed overview of family frm interviews

# **Findings**

 The results of this study show that successors differ from predecessors due to their different educational background and previous work experiences. These different characteristics seemed to affect the internal structure of their frms in terms of professionalisation, decentralisation, and inter-frm relationships. Successors tended to be more open and customer-oriented, and showed a more proactive approach. Ultimately, successors seemed to have more, and be more interested in supra-regional linkages. These linkages are one of the reasons for their lower commitment to their regional contexts compared to their predecessors.

# *Characteristics of successors vs. predecessors*

First of all, our analysis shows that successors and predecessors can be distinguished by particular characteristics. Predecessors usually did not have professional training and were often the founders of the businesses, either by necessity (lack of other alternatives in the labour market) or opportunity (after having worked in a similar frm in the same region, they found a business opportunity). In contrast to that, all successors in our sample had gone to university (except in one instance) and in most cases they studied engineering, economics, or business administration (see Table 6.2). Due to this fact, they can be described as white-collar workers who focus on managing the frm, but who lack the technical or product competencies of their predecessors. Even though some family successors had grown up with the frm and had worked in the factory on weekends and during holiday periods, many of them could not operate the machines anymore. They were only rarely seen on the shop foor – unlike their predecessors, who liked to be at the production site and work at the machines alongside their employees:

That's another one of my defcits. For example, my father had come from the workshop. He had no college education. My father was a machinist. So everybody respected him because: "This one knows what he's talking about." And he knew. And I don't know.

(Successor, Itsasondo, May 2015)

Additionally, as Table 6.2 also indicates, the majority of successors had worked in other companies before taking over the family frm, both in other Spanish regions and abroad, which was made possible by their higher education levels and knowledge of the English language.

# *Effects on the family frm*

The aforementioned differences in successor characteristics have consequences for the internal organisation of family frms in terms of professionalisation and decentralisation. In accordance with common management

# 132 *Regina Lenz*

concepts, successors introduced business processes and management tools that, to their surprise, were not there before:

There was no strategy plan, no budget, no structure or training plan, no purchase manager, not even a board of directors. So I asked them: "How do you manage? How do you do [it]? How can you control anything?" – "Oh, we work, we do invoicing, we check it every month, and the more invoicing we do, the better for the company." – "And what about the costs, and if you lose some money?" – "Ahh, at the end of the year we know if we lost or won." Oh my God. […] But I don't want to make them look bad, I am not criticising. That system has worked, and it got us here. But we cannot continue like this.

(Successor, Zumaia, February 2015)

Along the same line, successors planned to reduce the number of family members working in the frm and to hire employees based on their professional skills instead. Because although the inclusion of family members provides frms with easy access to resources and networks, it comes at the cost of satisfying the social expectations of providing jobs for extended family members (Soleimanof et al., 2018), as one successor illustrates:

From 80 people in the company right now, there were 25 family members, so this is the policy. If a family member is there, you will always contract a family member rather than somebody from the outside. I think it was more so in the past, but we are still doing that. But now, it is reduced to 20, so we are reducing family members and hope for a more professional management.

(Successor, Asteasu, March 2015)

 Similar to this, the leadership styles between the previous and the new ownermanager differ in that the latter seem to be more participative and willing to decentralise, asking for the advice of their employees and newly-founded boards of directors. Contrary to that, the predecessors' leadership style was described as monolithic, taking care of every decision themselves:

My father's way of working was very much centred on him. A very vertical organigram where "I decide everything." And with everything, I really mean everything. "We're going to paint the back door there, and I'm going to decide the colour." I'm not exaggerating, it was absurd. The frst [few] years we worked together, I struggled to fnd a path, an overall management principle, if there was one. For my father there was no path. It was work.

(Successor, Elgoibar, May 2015)

In some cases, successors even shared their leadership with a second manager – either to outsource all technical questions to him or her, or to separate daily business from more general decisions concerning the strategic orientation of the frm.

# *Effects on inter-frm relationships*

These different attitudes about how to manage family frms also affect their interactions with other frms. In general, successors in our sample were found to be more outgoing and proactive in establishing new business contacts and more customer-oriented in contrast to the more reactive and product-oriented leadership styles of their predecessors:

We started to be closer with the customers. This was a big change, because in the past they didn't speak English so much, they were more waiting for the orders to come in. So they stayed here, and the orders came, and they manufactured. So suddenly we started travelling more and being more in contact with the customers, at business fairs, for example. We started to look at things more deeply, at the needs of the customers. To understand them a bit better – why are they buying from us, what are the reasons, these kinds of things.

(Successor, Beasain, February 2015)

Furthermore, successors were more open to seeking professional advice, knowledge, and other services from external partners, leading to more open business relations with consulting frms and intermediate organisations:

Maybe another characteristic of Basque family businesses, they are not so collaborative. They are very closed, very black-box style. They are very non-transparent companies, they keep to themselves. When I talk to my uncles, they say, "Ahh, why are you talking to this guy?" They don't like it very much if you are talking to competitors, or research centres, or whoever. They think they will copy our ideas, and we know exactly what we need, so why would we need these research centres. "Are they going to teach us how to do our product? We are the best in that." That's the state of mind. They are not very open to collaborate with research centres or these kinds of things.

(Successor, Zumaia, February 2015)

# *Effects on the regional ties of family frms*

Regarding the geographical proximity of their business partners, all successors in our study reported a growing internationalisation in their relationships with clients, leading to a reduction in personal contact:

Our customers are further and further away. In the past they were very close, and now [they are] further and further away. You're giving all customers the same advice, the same product, but it's not as easy any more to just go next door. The importance of personal relationships is still the same, but the opportunity is less. In the past, when my uncles started, all the customers were within a distance of 20 km,

### 134 *Regina Lenz*

they would even deliver the product themselves with their own car, but I cannot do this anymore.

(Successor, Lazkao, February 2016)

Some successors still consider personal meetings important, as shown in the quote above, and they used the few opportunities for personal encounters to cook for their business partners in a gastronomic society. However, others prefer to separate their business and private lives and, accordingly, limit their business meetings to relevant matters only, reserving visits to gastronomic societies for their friends and family.

In contrast to their client relationships, family frms under the leadership of successors continued using mostly local suppliers. However, successors also signalled their readiness to substitute their local supplier relationships for more advantageous international ones:

For example, yesterday one of our customers was talking about a company from Asgoitia that sends the drawings, the plans, to China and in three days the pieces are here. And he was telling this as some kind of joke, but it was clear that it is not very well seen to send the work that might be done here to China or Morocco, for example. […] At the moment, our suppliers are all here, in the Basque Country. But maybe we will have to change that, I am not close to that. We have to adapt to our customers and their prices.

### (Successor, Elgoibar, March 2015)

This quote not only shows the successor's rather weak regional bond, but also the social pressure exercised by regional stakeholders to keep a close connection with the region. This social pressure that arises when family entrepreneurs live in the same area as their employees is also described by Block and Wagner (2014), who found that family ownership affects the social responsibility of family frms in the way that owners care about their reputation and about receiving recognition for the social activities they undertake in their local community. However, this kind of pressure seemed to be less effective with successors, who prefer to live in Bilbao or San Sebastián and commute to work in their rurally located family frms:

His [the predecessor's] mentality was: "I want to live quietly when I leave the house. And half of the employees are from Tolosa, and half of their families are from Tolosa. If I go to a bar, I meet half of them. I don't want to go to the bar and have stones thrown at me. Now, the successor does not live here anymore."

(Employee, Belauntza, April 2016)

The above quote is a good illustration of the need to appear legitimate and fulfl the social expectations of interaction partners. However, successors and their predecessors seemed to belong to two different organisational felds when it comes to determining legitimate behaviour. While predecessors were concerned about the reactions of regional stakeholders, successors seem to be more aware of the normative expectations of an international community that is bound together by their professions as engineers or economists, for instance. Due to these different connections, the reference group sanctioning non-compliant behaviour is not the local community any more, but international clients and former colleagues working in other companies, most of them located abroad. Successors, as family entrepreneurs, still want to be legitimate, but because their interaction partners have changed, predecessors and successors have different sources of legitimacy.

# **Discussion: on the road to regionally dis-embedded family frms?**

This study highlights the importance of interpersonal and inter-organisational linkages for understanding the regional embeddedness of family frms over time. These linkages constitute the relational infrastructure of family frms, through which normative expectations and beliefs are exchanged and maintained. Comparing the personal characteristics of successors and predecessors of family frms and their linkages with other individuals and organisations have revealed differences that suggest a generational shift in their relational infrastructure and, accordingly, in their degree of regional embeddedness.

I observed that while successors still share close frm-internal relationships, they are more focused on professionalism than family, are more oriented toward international interaction partners, and have stronger supra-regional networks than regional ones. Thus, regional embeddedness seems to be less important to them than to their predecessors. Whereas these fndings only represent a frst indication of an increasing regional dis-embeddedness of family frms, this tendency might affect the future patterns of regional development in Europe by getting amplifed in three ways.

First, the number of family frms approaching the time for succession is rising in the Basque Country, as well as throughout Europe. This suggests that the generational changes in terms of dis-embeddedness in individual family frms can turn into change on a wider scale in Europe. Taking an institutional perspective, successors can act as institutional entrepreneurs and consciously pursue changes by actively trying to modify business behaviours in their regional contexts. Alternatively, individual successors might engage in unconnected institutional work by simply following – to some degree unknowingly – their underlying, taken-forgranted beliefs of how to do business legitimately. Future research should investigate the consequences of dis-embeddedness of family frms for regional development.

### 136 *Regina Lenz*

Second, like Europe as a whole, the Basque Country faces challenges in terms of an ageing population and low fertility rates, making it more and more diffcult to pass on the business within the family (Duh, 2012; Isusi, 2008). This means that family frms increasingly have to look for successors from outside the family, or even from outside the frm or region. However, since such external successors are likely to have fewer regional connections and might thus respond less to local social obligations and expectations, this trend could aggravate the regional disconnectedness of successors even more. Future studies should explore the consequences of family frm dis-embeddedness for regional development when comparing family and non-family successors.

 Third, due to increasing global competition and price pressures, family frms will have to seek competitive advantages and growth prospects beyond their regional or national boundaries (De Massis, Frattini, Majocchi, & Piscitello, 2018), again increasing their international linkages while at the same time weakening their regional and local ties. While this pressure to internationalise affects all companies, successors – of their own accord – seem particularly willing to participate in a globalised economy. It is important to further investigate the globalisation process in family frms and how it affects their competitive advantages across generations.

 Finally, from a policymaker's perspective, I wonder what it would mean for regions to have a weaker bond with their family frms. On one hand, this outlook threatens long-established connections and well-functioning ways of doing business, adversely affecting other local frms operating in supplier relationships. Regions would lose job opportunities and the social and philanthropic contributions of family frms, making them overall less attractive for people to settle in. On the other hand, "a successful succession can give a new push to the frm" (Fernández & Nieto, 2005, 79) by providing it with new resources via successors' acquired capabilities, which might be needed to adapt the frm to new standards and to secure its survival in a globalised world. In this respect, successors' different leadership and management styles might be needed tools to prevent regional lock-ins and to ensure that they are not lagging behind. This might ultimately also beneft frms operating in the region or within the value chains of family frms. In a survey of German young professionals, Mohnen, Mückenhausen, and Toporova (2016) found that what renders family frms less attractive than other potential employers was their location and perceived lack of internationality. Thus, in order to enable regions to beneft from globalisation while retaining family frms locally, policymakers need to create ways to make regions and family frms attractive to potential successors and employees, and to ensure successful takeovers while embracing the increasing processes of globalisation.

For instance, policymakers may facilitate the succession process itself by providing succession consulting or innovative fnancing options in the case of costly family-external takeovers. In the case of the Basque Country, the regional government has introduced tax benefts to keep family frms local, by incentivising transfers to their employees instead of selling the frms to unknown shareholders from outside the region. To generally increase the attractiveness of family frms in rural regions and compete with the attractive working conditions of big corporations, policymakers need to provide educational, social, and cultural opportunities within the region and help address other problems of family frms such as high wage costs and skilled worker shortages (KPMG, 2019; Mohnen et al., 2016).

The results of this study imply that regional ties cannot be taken for granted; what is even more important is to consider the temporal dynamics (such as succession) that may trigger family frm dis-embeddedness. This chapter attempts to further research on the networks and regional linkages of family frms before and after succession and their consequent effects on economic and social development.

### **Note**

1 This term refers to the autonomous community of the Basque country in Spain (*País Vasco*). It should not be confused with the greater region of the Basque Country that, in addition, includes the Spanish autonomous community of Navarre, as well as three provinces (Labourd, Lower Navarre, and Soule) in the French Department of Pyrénées-Atlantiques.

### **References**


### 138 *Regina Lenz*


# **7 Are family frms more locally embedded than non-family frms?**

Findings from the fnnish context

*Stefano Amato, Mikaela Backman and Juhana Peltonen* 

### **Introduction**

The important contributions of family frms to the employment growth, gross domestic product, and innovativeness of both nations and regions is widely recognized (Astrachan & Shanker, 2003; Block & Spiegel, 2013). Hence, the growth of research concerning family businesses is not surprising and, consequently, neither is the progressive legitimacy that these organizations have gained among academicians, practitioners, and policymakers (Rodríguez & Basco, 2011). However, while existing research has focused on explaining the paradox of two highly intertwined domains—namely the family and the business—being able to shape frm behavior and its consequence on frm outcomes, the family business debate has overlooked the role of the geographical space in which the economic activity takes place (Basco, 2015). While, until the 1970s, space was mainly conceived in physical terms (i.e., as existing irrespective of any objects or relationships, space as a "container"), the concept of "relational space" (i.e., space as created by and consisting of a set of relationships among economic actors) has become the most infuential paradigm in both economic geography and regional studies (Capello, 2016). According to this view, space arises as an independent productive factor and can be a source of both static (e.g., higher productivity) and dynamic (e.g., enhanced innovation capabilities) advantages (Capello & Faggian, 2005; Parr, 2002) and also the basin of social infuence for corporate decisions (Lähdesmäki, Siltaoja, & Spence, 2019).

The relational view interprets space in terms of "social space" or networks of relationships (Dicken & Malmberg, 2008) in which economic actors arise not as separate agents but rather as participants in a fow of actions and interactions with the economic activity that is "embedded in concrete, ongoing systems of social relations" (Granovetter, 1985, 487). As opposed to traditional and sociological interpretations (Granovetter, 1985, 2005), the concept of embeddedness has attracted growing interest in both economic geography (Hess, 2004) and regional studies (Cooke, Clifton, & Oleaga, 2005), wherein it has been fundamentally conceived in spatial terms. The emphasis is now placed on the social nature of economic processes and their manifestation in territorial bounded areas (e.g., regions). According to this view, to be embedded is the frm ("who") in the community networks and institutional settings ("what") with local networks and localized social relationships that arise as the spatial logic of embeddedness (Hess, 2004; Kalantaridis & Bika, 2006). In light of this, "local embeddedness" can be defned as the nature, depth, and the extent of a frm's ties to the local social and economic environment.

However, it is reasonable to assume that not all frms distinguish themselves by having the same level of embeddedness in their socio-economic *milieu*. In other words, there may exist different types of frms with varying anchoring in territories—i.e., different kinds of frms may refect varying intensities with which they absorb and engage in, and in some cases become constrained by, the economic activities and social dynamics of these places (Oinas, 1997). From this perspective, the local embeddedness of family and non-family frms may differ because the former are generally regarded as being inextricably tied, physically, socially, and emotionally, to the territories in which they are located (Basco, 2015). In addition, the quality and depth of social connections they establish with the local community are deemed to be instrumental in keeping a family's control over a frm's operation and ownership, for perpetuating the family dynasty, and, last but not least, for sustaining family image and reputation across time (Naldi, Cennamo, Corbetta, & Gomez-Mejia, 2013).

While local embeddedness has traditionally been the main paradigm used to examine the unique way family frms and geographical space interact (Backman & Palmberg, 2015; Bird & Wennberg, 2014), existing research has not empirically proven *whether* and *to what extent* family frms are more locally embedded than their non-family counterparts. In order to address this research gap, we relied on micro-level data of Finnish frms. The Finnish case provides important insights as it allows us to make a detailed analysis of local embeddedness of family frms. It further provides us an opportunity to assess the relationship in a country characterized by large regional differences and where family frms make a large contribution to the overall economy (Statistics Finland, 2017). In particular, we employed two different and complementary measures of local embeddedness—the place tenure and the geographical distance between multiple and infuential stakeholders at ownership, managerial, and organizational levels (i.e., CEOs, owners, managers, non-managers/employees, chairpersons, and other board members) of the local frm. Our results show a positive association between the family nature of the frm and the degree of local embeddedness across different types of stakeholders. Thus, the paper shows, through both descriptive statistics and regression analysis, that stakeholders in family frms tend to be more locally embedded, through both place tenure and shorter distance to their local frms, compared to stakeholders in non-family frms. All stakeholders, except the owners of the frm, have a greater level of embeddedness in terms of place tenure, if they are part of a family frm. The same is true with regard to the shorter distances of family frms' stakeholders such as employees of the frms and the chairperson. Of all the stakeholders, the chairperson stands out and tends to be more locally embedded in family frms.

This chapter contributes to the existing literature in several ways. First, for family business studies, we provide further evidence about the microterritorial foundation of family frms. While the study of the interrelationship between family and business systems has traditionally represented the main area of interest in the investigation of the family frm phenomenon, it is only recently that an outward perspective, that is, one looking at the "familyfrm territory nexus," has gained growing interest (Basco, 2015). Second, to regional studies; by emphasizing the spatial dimension of embeddedness (Hess, 2004), our chapter offers new evidence about the association between the family nature of the frm and the degree of anchorage of the frm in its home territory. Family frms arise as peculiar regional actors that are distinctly affected by and able to infuence the geographical space in which both the economic activity of the frm and the social life of the family take place.

Finally, our contribution has policy-making implications. Recognizing the existence of long-lasting, pervasive, and socially infuential local actors such as family frms is deemed instrumental for enhancing the effectiveness of regionally-based strategies and implementation structures in support of the competitiveness of territorial innovation models (TIM) as well as local production systems (e.g., industrial districts, business clusters), the promotion of local entrepreneurship, and the correction of regional disparities. In particular, including family business dimensions in public policy would mean tailoring family-oriented policy interventions aimed at fostering their survivability and longevity, frm growth, innovation, internationalization, and entrepreneurship among others, which the economic development of a given region or locality ultimately depends on.

# **Theoretical background**

In the analysis of frms across space, the concept of frms' heterogeneity that is, the acknowledgment of a frm's internal characteristics affecting its outcomes in conjunction with the regional context in which the frm itself is embedded—has attracted growing interest (López-Bazo & Motellón, 2018). The increasing availability of fne-grained data at the micro and macro levels has allowed the estimation of, among other things, the impact of spatial economic and socio-institutional differences on a frm's innovation capability (Beugelsdijk, 2009) and effciency (Aiello, Pupo, & Ricotta, 2014; Fazio & Piacentino, 2010). Among the sources of frm heterogeneity, the family nature of the frm stands out as a pivotal attribute in explaining the distinctive behavior of some organizations across regional contexts (Baù et al., 2018) and the extent to which particular types of frms are affected by (Dekker & Hasso, 2016) and able to infuence the wider socio-spatial environment (Block & Spiegel, 2013) in which they dwell.

Family frms stand out as enduring economic actors that are characterized by the strong historical, cultural, and social connections with the community in which they are located (Bird & Wennberg, 2014). The social relationships among family members—which are built on the basis of trust, reciprocity, identifcation, and obligations—usually spread out from a family frm's boundaries (Adjei, Eriksson, & Lindgren, 2016), permeating the way these businesses develop and nurture social ties that, because of geographic boundaries, become "territorialized" (Rutten, Westlund, & Boekema, 2010). In exploring the micro-territorial foundations of family frms, the local embeddedness has been widely advocated as the foundation of the unique way these organizations and the territories in which they are located interact (Stough et al., 2015). That being said, understanding family frms' local embeddedness requires unveiling what makes these organizations so strongly anchored to their home territory and which are mainly attributable to their strong territorial identity (Capello, 2018; Pallares-Barbera, Tulla, & Vera, 2004) and economic integration at the local level (Courtney, Lépicier, & Schmitt, 2008; Kalantaridis & Bika, 2006).

The territorial identity of family frms has its frst source in what is generally known as "similarity," that is, common and shared values, traditions, culture, and language prevailing in a particular local milieu (Capello, 2018). Similarity is generated by ongoing historical and social actions in which the family, as an enduring, pervasive, and infuential institution, performs a crucial role in their formation (Soleimanof, Rutherford, & Webb, 2018). Because of their historical roots, long-term orientation, and social ties with their communities, family frms are able to infuence both political processes (i.e., formal institutions) (Craig & Moores, 2010) and the establishment of norms, values, and codes of conduct (i.e., informal institutions) (de Blasio & Nuzzo, 2010).1 These institutions, that prescribe the appropriateness of certain societal behaviors (North, 1991) and that regulate the interaction of economic actors become "crystallized" in the local and regional setting. They are deemed to shape the internal characteristics of a given territory (Pallares-Barbera et al., 2004) and to guarantee the reproduction of a "collective regional consciousness and a shared sense of belonging to a place" (Capello, 2018, 143). In addition to a symbolic meaning, as refected in the institutionalized patterns infuencing the way and the extent to which actions are coordinated, the territorial identity can be attributed a patrimonial value (Pollice, 2003). In particular, territory becomes a *locus* of highly recognizable and local-specifc products and, hence, an important site of meaning and identity construction for current and future family members.

The recognition of the product as being family business heritage—whose knowledge, traditions, and production methods are handed down each generation—and the perception of the family—whose status and reputation stem historically from the success of the business—result in these factors being inextricably linked with the territory, forming a background wherein the entrepreneurial dynasty is regenerated over time (Jallinoja, 2017; Smith, 2016). Moreover, similarity results in a sense of "solidarity" toward the territory where both the owning family and the frm are located. Solidarity, which refers to a sense of togetherness and anchorage to the territory, manifests in concrete social actions in support of the local community (Capello, 2018). The strong desire to develop and preserve the positive reputation and legitimacy of the family business from external and non-family stakeholders leads these organizations to engage in philanthropic activities (Campopiano, De Massis, & Chirico, 2014) aimed at supporting scientifc, educational, religious, and healthcare institutions, among others (Feliu & Botero, 2016). From this perspective, it is quite common for a foundation to be created, generally bearing the name of the founder or the owning family, as a vehicle tasked with the coordination of local community-oriented initiatives (Lähdesmäki & Takala, 2012). Additionally, the feeling of attachment to a given local area strongly infuences the investment strategies of family frms, whose evaluation incorporates the issues, concerns, and expectations of physically and socially proximate stakeholders (Mitchell, Agle, Chrisman, & Spence, 2011).

In this regard, it has been shown that family businesses have better environmental performance than non-family counterparts, and these difference are even more remarkable when family frms are more locally embedded (Berrone, Cruz, Gomez-Mejia, & Larraza-Kintana, 2010; Dekker & Hasso, 2016).<sup>2</sup> From a socio-emotional perspective, because at the local level the distinction among family and frms becomes more blurred, "social monitoring of family owners is strengthened, and the likelihood of enforcing social sanctions increases" (Berrone et al., 2010, p. 91). Hence, local roots deeply infuence family frms' behavior in terms of the preservation of the family's status and the quality of valuable and locationspecifc social ties, which would be irreparably damaged in the case of socially irresponsible actions (Amato, 2019).

Along with identifcation with the territory, a strong level of local economic integration stands out as a complementary element of the foundation of local embeddedness of family frms. This refers to the extent to which a given frm relies on the local context as a source of labor, materials, and a market for fnal output and information (Courtney et al., 2008). Family frms depend heavily on the local setting as a source of generic and strategic inputs with local contextual factors and circumstances that, therefore, may strongly infuence their behavior, survival, and competitiveness (Backman & Palmberg, 2015). The attributes and historical background of the family appears to be crucial for the intensity—expressed in terms of duration, frequency, and monetary value3—of the frm carrying out its economic transactions locally (Kalantaridis & Bika, 2006). A frm's headquarters corresponds typically with the place the founder was born, where the business venture started, and where the entrepreneurial dynasty is continuously nourished (Pallares-Barbera et al., 2004). This link, which is physical, emotional, and socioeconomic, is rarely severed; rather, it becomes the socio-spatial platform from which new growth paths, mainly through expansion in extra-local/regional markets, begin (Kontinen & Ojala, 2012; Amato, Basco, Backman & Lattanzi, 2020).

By virtue of their historical connections with a given place, family frms are in a position to adhere to and infuence the establishment of a set of common habits, routines, established practices, and value of conducts prevailing in the socioeconomic *milieu* (Pallares-Barbera et al., 2004); to develop a common knowledge base and expertise characterizing the local area (De Massis et al., 2016); and to build trust-based and reciprocal relationships with physically proximate local actors (Stough et al., 2015). This, in turn, "facilitates contacts, improves bargaining capacity, and attracts and maintains favorable conditions for the frm" (Pallares-Barbera et al., 2004, 648). This is deemed to explain performance divergences among family and non-family frms in more remote regions, such as rural areas (Backman & Palmberg, 2015; Baù et al., 2018; Greenberg, Farja, & Gimmon, 2018), and local production systems, such as industrial districts (Cucculelli & Storai, 2015) and business clusters (Pucci, Brumana, Minola, & Zanni, 2017), whereby family frms tend typically to exhibit stronger economic integration in the locality (Amato, Basco, Gomez-Anson, Lattanzi, 2020; Courtney et al., 2008). What's more, proximity dimensions become crucial for acquiring tangible and intangible resources and, hence, for competing successfully (Boschma, 2005). Finally, the strong local roots of family frms are found to shape the competitiveness of the region itself. By aggregating the collective actions of family frms, current empirical evidence shows how the regional innovation output is related to the density of family frms present in that region (Berlemann & Jahn, 2016; Block & Spiegel, 2013).

Therefore, given the aforementioned arguments, we posit the following hypothesis:

Hypothesis: *Being a family frm is positively associated with the degree of local embeddedness.* 

# **Empirical design**

### *Data and variables*

In order to empirically analyze how the ownership of a frm—where we distinguish between family and non-family frms—relates to its local embeddedness, we used register data from Statistics Finland covering the years from 2006 to 2014. The data has restricted public access as it allows for detailed analysis by providing information about the frm, its employees, CEO, managers, and also the connection to other important stakeholders such as the chairperson and other board members. It is further possible to assess information at the individual level regarding where the different stakeholders live and work. The data are longitudinal, which means that we can track individuals and frms over time. By combining information about the individual and the frm, we built several measures that capture the local embeddedness of the different stakeholders.

First, local embeddedness is measured as *place tenure,* which refers to the number of years an individual has lived in the same municipality as the local frm. The second measure entails the geographical distance between the stakeholder and the local frm—that is, *distance to frm*—measured in kilometers. To capture local embeddedness from a wider perspective, both place tenure and the distance to the frm were calculated for each frm stakeholder. We used several stakeholders engaged in a frm through management, ownership, and at the organizational level, who are important for the frm's development: (i) the CEO, (ii) the owner, (iii) other managers, (iv) non-managers, that is, employees, (v) the chairperson of the board, and (vi) other board members. Thus, we get a composite measure of local embeddedness with 12 indicators in total, each corresponding to different categories of stakeholders for both place tenure and distance to the local frm.

In order to differentiate family frms from non-family frms, we used the Finnish Longitudinal Employer-Employee Data (FLEED) combined with data from the Population Register Centre. For the purpose of differentiating frms' family ties, information on each individual's spouse and matches between children to their parents was utilized. By using this, we identifed the parents of the lion's share of the individuals in the working population and also many of the grandparents. The family ties were used to identify family frms. Data accessed through the FLOWN (Finnish Longitudinal Owner-Employer-Employee Data) database further allowed us to assess the ownership structure of each frm where the individual data could be matched to the ownership data, thus enabling us to identify who owns different shares of a frm. Family frms are, in this chapter, defned as frms where the majority of the owners belong to the same family, following (Litz, 1995; Villalonga & Amit, 2006). Families are defned as individuals, who (i) own shares of the frm either directly or indirectly, are (ii) connected by family ties, and (iii) jointly own the highest share of the frm, compared to other individuals. For detailed information about the construction of the variable, please see Peltonen (2018). To distinguish the different frms, a binary variable was used, denoted "1" for family-frms and "0" for non-family frms.

In the empirical estimations, we control for the basic frm and municipal features that infuence the local embeddedness of individuals: frm age, frm size in terms of number of employees, fnancial development in terms of turnover and total assets, whether the frm is part of a multi-establishment frm, and the size of the municipality. The following table presents the descriptive statistics for the variable of interest (local embeddedness), divided into family and non-family frms (Table 7.1).

The descriptive statistics for local embeddedness separated into family and non-family frms show a coherent picture where the former are more locally embedded compared to the latter. Family frms have a higher level of place tenure—i.e., all the different stakeholders have lived in the same municipality as the local frm for a longer time—compared to their non-family counterparts. In some cases, there are signifcant differences. For example, the average years a chairperson in a family frm has lived in the same municipality is 12.6 compared to 10.5 for the chairperson in non-family frms. In relation to the distance to the local frms, we observe the same pattern where stakeholders in family frms have on average a shorter distance to their frm compared to stakeholders in non-family frms. Based on the embedded measures used in this chapter, we fnd that family frms are more locally embedded than non-family frms in terms of their stakeholders. The relationship between local embeddedness and frm performance is something that has been discussed in previous studies (Backman & Palmberg, 2015; Baù et al. 2018; Bird & Wennberg, 2014). It should be noted that there is a large variation in both the place tenure and distance to local frms among stakeholder, regardless of ownership structure, as indicated by the large standard deviations.

In relation to frm-specifc characteristics, family frms are on average older and larger compared to non-family counterparts. That the frms are older can naturally lead to stakeholders having greater embeddedness in terms of place tenure. This is reasonable since we measured the number of years the individual has lived in the same location as the frm. Older frms that have not changed location thus enable a person to live longer in the same location as the frm and for the person to be engaged in the frm for a longer period of time compared to a younger frm.


*Table 7.1* Summary statistics of the focus variables of local embeddedness in family and non-family frms

Note: \*, \*\*, and \*\*\* indicates signifcance at 5%, 1% and 0.01%. Errors clustered at the frm level.

### 148 *Stefano Amato et al.*

#### *Empirical model and results*

The descriptive statistics show that stakeholders in family frms are on average more locally embedded. In addition, they show that family frms are on average older, which, in fact, could enable the stakeholders of the frm to be more rooted to the territory, as measured through place tenure. Thus, there seem to be factors associated with family frms that relate to the level of local embeddedness. In the next step, we therefore perform a random effects estimation to capture whether stakeholders in family frms are more embedded, controlling for frm and location specifc effects. The estimations performed are described in Equation 1.

$$\text{Loc.}\\_emb\_{l,1} = \alpha + \beta\_l\\\text{Family }\;firm\_{l,1} + \gamma X' + \text{Year}\_l + \text{Industry}\_{l,1} + u\_{l,1} + \text{c}\_{l,1} \tag{1}$$

Here *Loc*. *embit* is the local embeddedness measure for both place tenure and distance to their respective local frm for all six stakeholders; there are 12 variables in total. *Family firmit* is a dummy variable that is "1" for family frms and "0" otherwise. *X* ˜ is a vector that captures the control variables at the frm and municipal level. ˜1 and ˜ are the parameters to be estimated. *Year* captures the fxed effects for the different years (2006 to 2014) and *Industry* controls for the 2-digit industry codes of the frm. *uit* is the between-entity error and ˜*it* is the within-entity error. Standard errors are clustered at the frm level.

Having access to panel data allows us to explore the data in more detail compared to using a cross-section or a time-series model. Normally, when using panel data, a fxed or a random effects model is used. An advantage of the random effects model in our case is that it allows us to add timeinvariant variables that would otherwise be captured by the individual/ frm fxed effects. In this case, the family structure—that is, being a family frm—is normally something that does not change over time, and by using a fxed effects estimation, only the frms that change ownership structure would be captured in the estimations. Because we are interested in the differences between family and non-family frms, it is more appropriate to use the random-effect model. There are also disadvantages to choosing the random effects model over the fxed effects one; one such issue is omitted variable bias, as you need to control for the aspects that are time invariant as they are not captured by the fxed effects. We added several control variables to mitigate the problem of omitted variable bias.

The following tables show the results of how the ownership structure of frms infuences the local embeddedness of different stakeholders. Table 7.2 gives the results for the place tenure while Table 7.3 presents the fndings in relation to local embeddedness, defned as distance to local frms. The number of observations change across the different models due to missing values when measuring local embeddedness for some of the stakeholders.


*Table 7.2* Empirical fndings: random effects model by using place tenure as dependent variable

*Table 7.3* Empirical fndings: random effects model by using distance to the local frm as the dependent variable


Note: \*, \*\*, and \*\*\* indicates signifcance at 5%, 1% and 0.01%. Errors clustered at the frm level.

#### 150 *Stefano Amato et al.*

We can confrm the previous fndings from the descriptive statistics where most stakeholders in family frms overall had greater local embeddedness compared to stakeholders in non-family frms. The results hold also when controlling for frm-specifc characteristics where the age of the frm might be particularly important as a control variable. All stakeholders have a greater level of local embeddedness if part of a family frm, in terms of place tenure, except the owner of the frm. This is an interesting and also rather surprising fnding as the owner of family frms is often a person that belongs to the family. The results indicate that the owners of family frms are less locally embedded in terms of place tenure compared to those of non-family frms. One reason for this could be generational shift in family frms where the new owner could have a lower local embeddedness if he/she has lived somewhere else before the ownership shift.

The positive association between being part of a family frm and the local embeddedness of the stakeholders can be derived from several different factors. One aspect is the historical roots of family frms that relate to their probability of operating mainly locally and thus being embedded in the local *milieu* (Amato, 2019; Kalantaridis & Bika, 2006). One example is their strategy in competence development and hiring where a strategy to be more local would lead to more employees, managers, and board members being hired from the local community. The ability of members in family frms to build strong and long-lasting relationships, as noted by Stough et al. (2015), is a key component in being locally embedded. The local embeddedness can in turn shape the competitive advantage of the frm (Pallares-Barbera et al., 2004). Another factor that builds the local embeddedness among stakeholders in family frms is if the frm is founded and run in the same location where the frm founder was born (Pallares-Barbera et al., 2004).

Comparing the different stakeholders, the magnitude of the coeffcient for the chairperson stands out, as it is much larger compared to the other values and, in most cases, is triple the size. Thus, it seems that the chairpersons of family frms have a much stronger local connection compared to other types of frms.

It is interesting to observe the negative association between family frms and the owner as this indicates that the owners in family frms have lived a relatively shorter time period in the same municipality as the frm, compared to owners of non-family frms. Based on the previous theoretical section, one would assume a positive association. As the owners in family-frms are, according to our defnition, family members, one would assume that they would be more locally embedded. One possible explanation is that families also capture marital links where spouses are considered as family. Spouses have a higher probability of being born and having lived outside the municipality the local frm is located in and are, in many cases, part of the team that owns the frm.

The results in Table 7.2 also highlight the possibility of reverse causality. It might also be that family frms are less able, compared to non-family counterparts, to attract stakeholders from outside their own municipality. Family frms might actively engage in activities to recruit and attract stakeholders from outside but are less successful in this endeavor as they might not be perceived as attractive employers by individuals outside of the family. If this is the case, local embeddedness might act as a hindrance for family frms to grow and develop and to get new insights from external stakeholders. Thus, local embeddedness and being a family frm will, in this case, act as lock-in effect.

In the next table, the results for local embeddedness defned as distance to the local frm are presented.

The same patterns that were found for local embeddedness in terms of place tenure are not mirrored for the distance measure. In this case, we only fnd a signifcant relationship between being part of a family frm and local embeddedness in three out of the six measures. For employees that are not managers in the frm and for the chairperson of the frm, being part of a family frm increases their local embeddedness. The negative association between the family frm dummy translates into a shorter distance between the stakeholder and the frm, and hence a greater local embeddedness. On the other hand, for the owner of the frm, we once again fnd that being part of a family frm decreases the local embeddedness of the individual.

When analyzing the control variables, we observe that older frms increase the level of local embeddedness for each of the stakeholders. This is an expected fnding as the age of the frm also affect how long a stakeholder can be locally embedded in terms of place tenure. The relationship between the size of the frm, in terms of employees, and local embeddedness is less clear and is only positively and signifcantly related to local embeddedness in a few estimations. The fnancial size of the frm, in terms of turnover and total assets, lowers the local embeddedness of the stakeholders. This can be viewed as supporting evidence for the argument that there is selectivity in how successfully different frms are able to attract stakeholders external to the local community. More successful frms are plausibly more able to attract individuals from different regions as they are seen as an attractive possibility and people will be willing to move to the frm, and hence lower the local embeddedness of the different stakeholders. Being part of a multiestablishment frm increases the local embeddedness of the stakeholders as employees have not only more options to change positions within the frms but also across establishments and through this fnd new career paths. The size of the municipality seems to be weakly related to the local embeddedness of stakeholders.

### **Conclusion**

Examining the relationships between family frms and territories, or "family frm-territory nexus," using an outward perspective, has only recently begun to attract interest in the feld of family business studies. Family frms, like other types of frms, are intrinsically territorial—that is, their operations take place in spatially bounded areas. But, differently from their non-family counterparts, the territory is more intensively local for family-led businesses in such a way that these economic actors, due to physical, emotional, and socioeconomic attachments, are uniquely affected by and able to infuence the local *milieu* in which they dwell. In particular, the frm's embeddedness in spatial structures of social relationships—i.e., local embeddedness—has been advocated as the foundation of the micro-territorial perspective of family businesses. While previous studies have shed light on the distinctive conditioning of local embeddedness on family frms' decisions (Dekker & Hasso, 2016) and sources of differential performance as compared to non-family counterparts (Baù et al., 2018), the current literature is somewhat silent on *whether* and *to what extent* family frms are more locally embedded than non-family counterparts.

By relying on a large dataset of Finnish frms, we explore the association between the family nature of the frm and local embeddedness, as expressed in place tenure and the spatial distance, of six different categories of infuential stakeholders of the frm. Our results reveal that stakeholders tend to reside for a longer period of time in the same place as local family frms, which, hence, rely greatly on more persistent and geographically proximate stakeholders. Thus, family frms are more locally embedded where the chairperson of the family frm tends to be a relevant link with their home-territory. Additionally, our fndings show a negative association between family frms' status and spatial distance, with some categories of stakeholders unveiling the higher responsiveness of family frms to geographic distance in the availability of resources (i.e., labor and directorial skills and capabilities) provided by some stakeholders (i.e., workers and chairperson). These two results, taken together, reveal how family frms, as compared to their non-family counterparts, are more embedded in the local setting.

Our book chapter contributes to the convergence efforts between regional and family business studies. While family business research has traditionally overlooked the regional context in which the economic activity of the frm and the social life of the family take place, the interaction between family frms and territory is steadily emerging as a piece missing in the comprehension of family frms' distinctiveness (Basco, 2015; Stough et al., 2015). For family frms, region arises not only as a socio-spatial platform to which they are functionally and economically bonded, but also as symbolic and emotional constructs inside of which these organizations evolve across generations. Therefore, introducing the "locality" in the study of family frms would account for the existence of sets of physical, social-institutional, and historical attributes that mesh with the attributes of both family and frm. On the other hand, for regional studies, the recognition of family frms enables investigations in the role of space as an independent production factor and generator of distinctive static and dynamic advantages for the frms located inside.

Finally, our book chapter has relevant public policy implications. First, as the study shows a substantial reliance of family frms on particular stakeholders, any "dis-embedding" event de-anchoring the frm from its own territory may potentially undermine the conditions of longevity and growth. This dis-embedding might be a result of the cooling of family ties, due to generational transfers or, even worse, the presence of factors preventing intra-family succession, among others. From this perspective, the introduction of taxation relief in cases of inherited business transferring or tax allowances for the purchase of management consulting services needed for successful succession planning, may be extremely useful for the improvement of the legal and institutional contexts of family frms. Second, since our results reveal that family frms are more susceptible to geographic distance and spatial variation in the availability of essential resources, mainly labor and management skills, any policy-making interventions should be directed at correcting the qualitative and quantitative imbalances in the geographical distribution of the aforementioned resources. In fact, given the importance of family frms as both absolute (that is, in terms of the total number of operating businesses) and relative (that is, in terms of contribution to the GDP and economic well-being), the proper endowment of productive factors at both regional and local levels appears to be crucial for their competitiveness and, ultimately, survival.

While our chapter represents one of the frst attempts to analyze in a systematic and detailed way whether and to what extent family frms differ in their local engagement and how anchored they are, future research should further investigate the role of local embeddedness in the exploitation of spatial advantages as refected in higher productivity (i.e., static advantages) and innovation capabilities (i.e., dynamic advantages) of family frms. The embeddedness of family frms may result in a unique network position at the local level, potentially infuencing the acquisition and exploitation of a spatially bounded fow of knowledge and information (i.e., local knowledge spillovers) that, in turn, affect their performance. Additionally, as our study uses administrative borders (i.e., municipalities) as geographical levels of analysis, future studies should investigate local embeddedness in settings such as local production systems (e.g., industrial districts and business clusters) wherein the strong territorial identity and high level of economic integration are deemed to play substantial roles in the behavior and competitiveness of family frms and in the evolution of the locations themselves. Finally, even though local embeddedness has been tested at the micro-level, an interesting future research avenue would be to aggregate the collective actions of family frms, that is, to adopt a macro-level of analysis, in order to test whether attachment to the region in which they are located results in a source of resilience of the region itself when adverse events (e.g., recessionary shocks, natural disasters) occur.

### **Acknowledgment**

We would like to thank Edvard Johansson, Faculty of Social Sciences, Business and Economics, Åbo Akademi University, Finland for his help with the data processing.

### **Notes**

1 In his seminal contribution, North (1991) defned institutions as "humanly devised constraints that structure political, economic and social interaction"

### 154 *Stefano Amato et al.*

(p. 97). They consist of formal rules (constitutions, law, and property rights) and informal constraints (sanctions, traditions, rules of conduct) that, throughout history, "have been devised by human beings to create order and reduce uncertainty in exchange" (p. 97).


# **References**


### 156 *Stefano Amato et al.*


# **8 Family frms and corporate responsibility in peripheral regions**

*Martin Graffenberger and Franziska Görmar* 

## **Introduction**

Research on family frms has traditionally focused on the analytical distinction between them and non-family frms, implying, amongst others, differences in terms of demographic, structural, and behavioural dimensions (Basco, 2015). While family frms, like any economic agent, pursue pecuniary goals, non-fnancial goals are important aspects of their strategic agendas (Aparicio, Basco, Iturralde, & Maseda, 2017). Specifcally, long term orientation of strategic decisions (Bird & Wennberg, 2014; Zellweger, Chrisman, Chua, & Steier, 2019), responsibility for employees and communities (Campopiano, De Massis, & Chirico, 2014; Gallo, 2004), and sensibility towards aspects of social status and reputation (Chrisman, Chua, Pearson, & Barnett, 2012) appear as distinct attributes of family frms' business practices to mediate economic and non-economic goals.

The aforementioned attributes help explain why family frms exhibit high degrees of local embeddedness (Baù, Block, Cruz, & Naldi, 2017; Stough et al., 2015). Embeddedness creates linkages between frms and their diverse contexts and, thereby, functions as an essential lever to facilitate and channel regional development dynamics regarding both economic and noneconomic aspects (Pike, Langendijk, & Vale, 2000). Given that around 90% of businesses in Germany are family frms, accounting for 58% of private employment and 52% of aggregate turnover, and that many of them are of considerable size (Stiftung Familienunternehmen, 2019), family frms are a crucial determinant of regional development processes (Basco, 2015). However, there is a lack of research on interpreting and understanding the role they play in local economic and social development (Baù et al., 2017; Lengauer & Tödtling, 2010). Because the nexus between family and regions still requires further research (Basco, 2015; Stough et al., 2015), we aim at offering insights on wider community-related and societal implications by considering regional engagement and philanthropic activities through the lens of corporate local and regional responsibility (CLRR). That is, we attempt to shed new light on the following research question: *How and why do frms take responsibility for their local communities?* 

### 158 *Martin Graffenberger and Franziska Görmar*

The empirical material, gathered as part of a wider research project, relates to the so-called "Hidden Champions" located in small towns in peripheral regions in Germany. Hidden Champions, frms that are market leaders in specifc product segments, are regarded a suitable example to study corporate responsibility as they typically possess suffcient resources for engagement, tend to exhibit a high degree of local embeddedness, and are mostly family controlled (Vonnahme, Graffenberger, Görmar, & Lang, 2018). There are three main stylised facts that guide our focus on frms located in small towns in peripheral regions. First, small towns are a central part of the German settlement system but face particular development challenges. Second, research has neglected to focus the attention on them (Academy for Spatial Research and Planning, 2019; Bell & Jayne, 2009). Finally, in the German context, successful and innovative frms are frequently located in small towns.

In this chapter, we examine three companies located in two peripheral small towns in Germany – Bad Berleburg in North Rhine-Westphalia (headquarters of two family-owned Hidden Champions) and Schierling in Bavaria (headquarters of one management-led Hidden Champion). We selected the cases based on company specifcations such as frm size and (changes in) the nature of frm ownership as well as the location and size of the towns where they dwell. Furthermore, the selected case frms refect variance regarding their ownership structures, by recognising both long-lasting family ownership and the transition from family to non-family ownership with subsequent integration into an international family holding.

We fnd that successful and locally embedded companies are generally willing to provide substantial resources for community purposes, in particular regarding local/regional cultural, social, and educational matters. Engagement is coupled with diverse frm interests such as securing human resources by maintaining and increasing the attractiveness of place and region. Hence, these activities can have pronounced effects on small-town development, especially as the spatial focus of engagement is mainly on the immediate frm environment. The actions of locally embedded companies constitute a substantial supplement to the activities and investments of municipal and civil society actors and, thereby, open room to manoeuvre and to shape local and regional structures. All selected frms, regardless of their ownership nature and structures, are engaged in corporate giving and corporate volunteering activities.

However, our fndings indicate that ownership structures may affect both scope and intensity of community engagement. Family frms are particularly involved in comprehensive and strategic activities, such as steering group membership of local initiatives or participation in municipal strategy processes. This comprehensive engagement is driven by a sense of obligation and is coupled with efforts to generate returns for the local community – i.e., beyond mere economic returns. Since the frms' decision-makers grew up and still live locally, socio-emotional place attachment emerges as a distinct driver of frm engagement. Consequently, our study allows the further differentiation of previous fndings on the interrelations between frm engagement and ownership structures (e.g., Bürcher, 2017; Lengauer & Tödtling, 2010). We observe that while family involvement seems to affect the tendency of a frm to engage in corporate support activities (long-term orientation), sponsoring and volunteering activities are common in any frm regardless of its ownership structure.

# **Theoretical framework**

# *Family frms and embeddedness*

Regional embeddedness is about the involvement of (economic) actors in a geographically bounded social structure (Granovetter, 1973; Hess, 2004) whereby these actors become tied to spatial contexts (Pike et al., 2000). Through embeddedness, these actors shape the economic as well as societal dynamics of places and regions (Hess, 2004). Hence, embeddedness refects the frm-place nexus and constitutes an integral part of local and regional development beyond its economic dimension (Bürcher, 2017). In economic terms, embeddedness relates not only to transactions with regional suppliers, sub-contractors, and customers; integration in regional networks and intra-regional competition but also access to technology, human resources, and regionally specifc knowledge (Clifton, Gärtner, & Rehfeld, 2011; Lengauer & Tödtling, 2010). In social and cultural terms, embeddedness surfaces in linkages to local/regional administrative, intermediate, and civil society stakeholders (Baù et al., 2019; Bird & Wennberg, 2014; Bürcher, 2017) and a frm's integration into regional cultures, manifested in collective values, norms, and symbols (Clifton et al., 2011; Dequech, 2003). Thus, embeddedness facilitates the mobilisation of local resources, trust, and networks. Consequently, embeddedness operates as a determinant for corporate and regional competitiveness (Baù et al., 2019) but also conditions frms' intentions to pursue non-fnancial goals, such as philanthropy and community engagement activities (Bürcher 2017; Campopiano et al., 2014; Lengauer & Tödtling, 2010).

However, not all economic actors are similarly linked with their local communities. For instance, it has been mentioned that, compared to their non-family counterparts, family frms exhibit profound degrees of local embeddedness (Basco, 2015; Baù et al., 2017; Bird & Wennberg, 2014), especially regarding non-economic and social dimensions (Chrisman et al., 2012; Zellweger et al., 2019). Family reputation (Deephouse & Jaskiewicz, 2013), the fact that decision-makers typically grew up and live locally, shared regional mindsets (Dörhöfer, Minnig, Pekruhl, & Prud'homme van Reine, 2011), and family members being integrated into the community's everyday life (Lang, Görmar, Graffenberger, & Vonnahme, 2019) also play a decisive role in their engagement. Family frms are well embedded locally and beneft from this (Baù et al., 2019), especially in rural regions, where they "have the possibility to form alliances and build close connections with the

### 160 *Martin Graffenberger and Franziska Görmar*

community and are exposed less to the anonymity of urban areas" (Bird & Wennberg, 2014, 425).

It is important to note that embeddedness is not a static manifestation but an inherently dynamic concept (Hess, 2004). In this sense, Pike et al. (2000) highlight the importance of considering spatial processes of *embedding* rather than a state of *embeddedness* as well as counter-processes of *disembedding*. Thus, a frm's level of embeddedness might increase or decrease over time, as the characteristics of their ties to local/regional contexts alter for reasons such as changes in purchasing strategy, management, or ownership structure. In this sense, a frm's relation to economic actors, as well as administrative and societal stakeholders, fuctuates (Zellweger et al., 2019).

### *Corporate local and regional responsibility*

The "active involvement of frms in shaping the contexts and networks a frm is involved in" (Lengauer & Tödtling, 2010, 7) is widely understood as a key element of corporate engagement. It is a distinct form of and mechanism to strengthen socio-spatial embeddedness (Bürcher, 2017). Consequently, and due to their typically pronounced degree of local embeddedness, family frms are particularly receptive to corporate engagement (Campopiano et al., 2014) through their locally embedded decision-takers.

Research on corporate engagement lacks both distinct spatial elements (Lengauer & Tödtling, 2010) and ownership-related considerations (Campopiano et al., 2014). Even though activities under the umbrella of concepts like corporate social responsibility or corporate citizenship often have a (more or less explicit) regional orientation, the concepts themselves are non-spatial. To incorporate spatial considerations, this chapter adopts the notion of CLRR developed by Hohn, Kleine-König, and Schiek (2014). Departing from a holistic understanding, CLRR is about the "development partnership between companies and the region" (Kiese & Schiek, 2016, 10) in which frms engage voluntarily for economic, ecological, social, and cultural matters.

The associated activities in the partnerships related to CLRR can be categorised into three distinct, yet not mutually exclusive, mechanisms: corporate giving, corporate volunteering, and corporate support (Hohn et al., 2014; Vilain, 2010). Corporate giving refers to engagement through occasional or long-term fnancial and material donations. It emerges, for instance, as sponsoring of cultural and social events, institutions, associations, and sports clubs but also material support for (vocational; music) schools and kindergartens. Secondly, activities along the lines of corporate volunteering involve, for instance, leaves of absence for staff to participate in local fre brigades and social activities or days with schools and educational institutions. Furthermore, frms frequently provide their premises as venues for cultural events, such as concerts or readings. Finally, corporate support refers to strategic, i.e. rather formalised, and long-term engagement in local and regional development processes – e.g., through support of and participation in civic or regional management initiatives, local strategy formation, or the establishment of foundations. CLRR considers engagement primarily in the direct context of a frm's location. Consequently, corresponding activities can contribute directly and indirectly to various dimensions of small-town development – e.g., by securing diverse and high-quality social and cultural landscapes.

Generally speaking, proactivity, strategic (business) orientation, commitment, and resource input tend to increase from giving to support activities (Bürcher, 2017; Kiese & Schiek, 2016). Even if CLRR activities typically exceed immediate business goals, they might be coupled with and partially motivated by business-oriented considerations (Kiese & Schiek, 2016). For instance, local engagement can increase visibility and reputation of frms. Additionally, it has been highlighted that corporate responsibility can be seen as a specifc instrument to shape socio-spatial contexts and to maintain favourable local and regional living and business conditions – which might be particularly relevant for frms located in peripheral regions (Bird & Wennberg, 2014; Bürcher, 2017) but also result in elevated expectations regarding the engagement of (internationally successful family) frms. While frm engagement can support processes of local and regional development, associated activities are, due to their voluntary nature, highly volatile, dependent on frm success, and likely to be cut in economic downturns (Gallo 2004; Lengauer & Tödtling, 2010), such as the current COVID-19 crisis.

Research has so far largely neglected the particular effects of family frms in peripheral regions (Bird & Wennberg, 2014) and especially their wider societal implications (Zellweger et al., 2019). Hence, this study specifcally focusses on how frms located in peripheral regions engage in processes of (small) town development and, thereby, highlights a particular spatial dimension within the debate on corporate responsibility. With this chapter, we establish a direct link between CLRR and frms' immediate operational contexts, but also adopt its holistic understanding by investigating the linkages between CLRR and cooperative small-town development (Deutscher Städtetag, 2013).

# **Consciousness and strategy: frms as actors in small-town development**

### *Case studies – local contexts and frm embeddedness*

We investigate three case studies of Hidden Champion frms located in small towns in peripheral regions in Germany, namely EJOT and Berleburger Schaumstoffwerke (BSW) located in Bad Berleburg (North Rhine-Westphalia) and Holmer Maschinenbau located in Schierling (Bavaria). The case studies draw on 13 interviews conducted with frm representatives (owners, management), central administrative (mayors, administrative staff), and civil society actors (representatives of local clubs, associations and initiatives, regional management, etc.), as well as local media, during

# 162 *Martin Graffenberger and Franziska Görmar*

multi-day feld visits in 2018. Interview material was complemented with data gathered through desk research (from municipal documents, company websites, media etc.).

All three frms are important employers in their towns and contribute, with their economic interrelations and linkages, considerably to local and regional development. Hence, in all cases the economic core activities (through employment, taxes, purchasing power, expansion, etc.) are considered their main pecuniary contribution to local development. In the following section, we briefy portray the companies and give a short insight into their respective local contexts.

**EJOT** and **BSW** frms are both family-owned and in the third generation of family ownership. EJOT, founded in 1922 by the great uncle of today's managing director, currently employs more than 3,000 persons, approximately 1,300 of whom work in Bad Berleburg and its surroundings. The main goods produced by the company are special screws and other connecting elements. BSW was founded in 1954 and employs about 650 persons, of which 470 work in Bad Berleburg. The main product of BSW is foam material utilised for sports mats or running tracks (Box 8.1).

# **BOX 8.1: CONTEXT INFORMATION ON BAD BERLEBURG**

Bad Berleburg, North Rhine-Westphalia


Bad Berleburg faces ongoing demographic challenges that manifest in selective out-migration and population ageing. Both trends are of signifcance to local frms, in regards to their needs for qualifed workforce. Infrastructural access, both road and railway, are deemed highly problematic, resulting in the local "Route 57" lobbying initiative.

In the past, the municipality experienced a diffcult fnancial situation. However, the city council re-oriented its urban planning and development strategy, deliberately involving economic and civil society stakeholders in strategy formation processes and project activities.

In the past, the healthcare sector was highly important for Bad Berleburg but has, due to regulatory changes and economic restructuring, somewhat lost signifcance. Currently, Bad Berleburg builds on a diversifed industrial base characterised by medium-sized and family-owned companies – including the Hidden Champions **EJOT**  and **Berleburger Schaumstoffwerke (BSW)**.

### *Family frms and corporate responsibility* 163

Both frms are highly embedded in Bad Berleburg in economic as well as social and cultural terms. Economically, they constitute a great share of local employment and signifcantly contribute to local tax revenue and purchasing power. Additionally, they strengthen the local economy through subcontracts and the concentration of their economic activities within the region. Committing themselves to Bad Berleburg as their headquarters' location, both frms have expanded in recent years, partly by reusing former buildings of the healthcare industries and industrial brownfelds. Both companies have their headquarter buildings in central locations, refecting economic relevance and importance for the urban landscape.

In social and cultural terms, Bad Berleburg is not only the headquarters of the frms but also residential place of the founding families, who grew up in this region and are engaged privately in various local associations and initiatives. The same applies to their employees, who originate mainly from the surroundings. Both frms show strong loyalty towards their employees and are themselves engaged in multiple ways in the town's societal life. Thus, it can be expected that traditions and regional values are shared by both managers and employees (Dörhöfer et al., 2011), resulting in strong regional ties. Both frms understand themselves as rooted in and interwoven with the region where they fnd appropriate workforce and human resources to successfully manage their corporate development.

Human resources are a great potential and capital of our company and that certainly ties us to this location. It is certainly not the local infrastructure, which is a real disadvantage.

(Interview BSW, August 8, 2018)

As a consequence, the two frms' leaders are very much motivated "to keep the town and the region attractive for employees" (Interview EJOT, August 08, 2018) and engage in its development.

The Hidden Champion **Holmer** was founded in 1969 and produces sugarbeet harvesters and other agricultural machines. Of the frm's 420 employees, 330 work in Schierling. Holmer has witnessed several changes in the ownership structure. In 2004, the founder Alfons Holmer sold the enterprise to a private equity company, which in turn sold it to a second private equity company in 2006. Since 2013, Holmer has belonged to the French family-owned holding Exel Industries, who are familiar with the agricultural machinery business. Hence, Holmer can be considered a management-led enterprise with a family-owned enterprise in the background (Box 8.2).

Both the municipality and the company management consider the private equity ownership phase marked by uncertainty and turbulence. Since the acquisition of the company by Exel Industries, the relations between the town administration and the company improved considerably, strengthening, once more, the base for communication and trust.

# 164 *Martin Graffenberger and Franziska Görmar*

Ten years in private equity were certainly not favourable for our company, particularly because the identity was missing. […] It was a diffcult period. Finally, in 2013 we could indirectly get back into family ownership.

(Interview Holmer, November 6, 2018)

By acquiring a property of 14 hectares in a newly established industrial area, Holmer committed itself to the location of Schierling, which was a long-term and strategic decision by the management.

The development of the new location was a long-term and strategic decision, knowing that we are not able to relocate anywhere else. This is just not possible due to our product and our self-image. Our knowledge carriers are located here and we are deeply rooted. Therefore, we do not want to leave.

(Interview Holmer, November 6, 2018)

# **BOX 8.2: CONTEXT INFORMATION ON SCHIERLING**

Schierling, Bavaria


 Schierling has continuously gained population in the past few years. These population dynamics are partly due to its location in the commuting area of Regensburg. Schierling is well connected in terms of infrastructure. Recent infrastructural improvements have also facilitated the development of a commercial industrial site, which attracted a major automotive supplier and led to several local/regional frms, such as the Hidden Champion **Holmer Maschinenbau GmbH,** expanding their operations.

Due to these recent dynamics, the fnancial situation of Schierling is rather positive. Beginning in 1999, a number of urban development concepts were elaborated and implemented, resulting, for instance, in the establishment of a centrally located service centre containing local supplies, a medical centre, and the public library as well as space for meetings. Currently, a major project of the city council is the development of a new hotel.

Regarding its regional embeddedness, the economic aspects are clearly considered most important. The company is one of the biggest employers in the region and is an important partner in education. This generates purchasing power in the region and has increased tax revenue for the municipality, opening up considerable room for manoeuvring urban development.

As far as possible, Holmer organises its supply chains regionally, thereby strengthening local and regional economic structures. Further, the commitment of Holmer to stay in Schierling has substantially infuenced the development of the municipality. A new commercial estate was developed as result of Holmer's expansion plans, a process which was characterised by continuous communication between the city council and the frm. Both sides emphasise that the size of the town may also have a facilitating infuence on their mutual communication.

Socially and culturally, the enterprise is apparently less embedded than other companies in Bad Berleburg. While a major share of the employees originate from the region and are active in local associations like the fre brigade, the persons responsible for the management mostly live in Regensburg, a nearby regional centre. This fact might explain the supportive but rather distant and professional relations between frm and community actors in Schierling. Although there are many well-functioning communication channels, formal and informal, between the municipality and the frm, both sides have precise ideas about the competencies and responsibilities of public and private actors in urban development processes, which are rather separated from each other.

# *Corporate responsibility: from club sponsoring to community building(s)*

All Hidden Champions investigated as part of this research exhibit distinct interests in assisting the development of their immediate (headquarters) environments. Consequently, they engage in various, wide-ranging activities, particularly regarding cultural and social initiatives, as well as education, environmental concerns, and sports. Activities along the lines of CLRR and their different mechanisms (corporate giving, volunteering, and support) were identifed in each of the selected cases, albeit with different levels of engagement.

All three companies provide fnancial and material support for a wide range of local initiatives and institutions – e.g., for cultural associations and events, a biodiversity project on wisents, sports clubs (e.g., football, athletics and biathlon), schools/kindergartens, and youth associations. Additionally, BSW and EJOT open their facilities as venues for events such as concerts or literary readings. At Holmer, employees can propose relevant projects and initiatives to the management for funding and even use the facilities and machinery of the frm for their own civic engagement. Generally, it can be observed that, through these corporate giving and volunteering activities, the companies aim at making a conscious contribution to the overall locational attractiveness of their headquarter sites. In this sense, a certain diversity and quality of local social and cultural landscapes are considered relevant locational factors when it comes to binding and attracting workforce (Table 8.1):

We [think about engagement] every time we have the impression that some of our employees may beneft.

(Interview EJOT, August 8, 2018)


*Table 8.1* Selected CLRR activities of case frms

In addition to fnancial and material support, all companies grant leaves of absence for employees to participate in fre brigades and the Red Cross, for example. Furthermore, they cooperate closely with local/regional schools and vocational training centres, organise internships or open days to present themselves to future members of the workforce, and even enter into formal partnerships with neighbouring schools. A very important aspect in this regard is the high value that they place on training and apprenticeship. EJOT and BSW, for example, are strongly involved in the regional structures for professional education, being part of a regional educational centre (Bildungszentrum Wittgenstein GmbH) as founding and currently managing members.

Regarding strategically oriented corporate support processes, the intensity of engagement differs between the three companies. Both EJOT and BSW have been actively involved in strategic local and regional planning processes such as the municipal mission process ("Leitbildprozess"). However, the managing directors referred to their participation in these processes not as representation of their frms, but as private citizens of Bad Berleburg. Yet, a distinction between these roles seems not always to be clearly drawn, as private and corporate functions might easily overlap in public perception.

I participated but I perceived it as private meeting. It makes a difference whether the minutes of the meetings say "the representative of BSW or Mr. [family name]" said something. […] But in a small town like Bad Berleburg all of that blurs.

(Interview BSW, August 8, 2018)

Besides this local engagement, BSW and EJOT take on responsibilities in regional development processes. For instance, as a member of the steering groups of the LEADER region Wittgenstein, the regional Chamber of Commerce, and economic associations and initiatives lobbying for better regional infrastructure connections.1

One central project in Bad Berleburg in which both local Hidden Champions have been actively engaged is a newly built regional community centre integrating both touristic and social functions.<sup>2</sup> The centre constitutes a central, strategic urban development project of Bad Berleburg. At its core, it contains a multifunctional event and meeting place where several central organisations of the town (municipal youth welfare, Wisent world, tourist offce, etc.) are bundled and rooms for further associations and initiatives are available. The total investment is about €2.5 million, of which 70% was fnanced by federal and state funds. Through the donation of €200,000 each by BSW and EJOT, the municipality was able to co-fnance its 30% share of total project costs. Additionally, the companies contributed also to the planning process, by sharing their expertise in terms of event management and the likely specifc needs of such a place. This corporate engagement is

*Figure 8.1* Roofng ceremony of community centre in Bad Berleburg. Source: Pahl + Weber-Pahl Planungsgesellschaft mbH & Co. KG.

one of the reasons why the community centre is considered a "best practice example of regional engagement" (Prasse, 2018). Yet, there are also specifc and frm-related motivations that guide the support. First, BSW and EJOT can use the centre as a venue for their own events – something they had lacked. Second, the centre offers dedicated space for projects and events for adolescents to increase the attractiveness of Bad Berleburg for families, an important target group for both the municipality and local frms.

I was convinced [by the project] rather quickly, because it is good for Bad Berleburg. [We get] a representative building where receptions and other events can be arranged, even bigger ones.

(Interview EJOT, August 8, 2018)

In contrast, Holmer's corporate support activities are less pronounced than those of the companies in Bad Berleburg. Both Schierling's administration and its management underline the strict distinction between municipal and entrepreneurial competencies and tasks. This understanding may be one of the reasons why long-term engagement does not play a big role in Holmer's voluntary activities, with the exception of its participation in the climate council of the municipality and its membership in the local economic association. Holmer's activities manifest in rather low-threshold activities such as project and event sponsoring. Still, the frm's management is interested in current urban development and planning processes and supports them ideationally, but it does not actively engage in strategy formation or project implementation:

There is no strategic cooperation regarding urban development. We do not meet with the city council and jointly develop a long-term plan.

(Interview Holmer, November 06, 2018)

### **Discussion: from corporate engagement to cooperative action**

Using the examples of three Hidden Champions in Germany, the study examines the local effects of successful operating frms located in small towns in peripheral regions. In economic terms, the three selected companies are important local players and are well embedded within their regions. The frms contribute signifcantly to municipal budgets, offer diverse employment opportunities, and strengthen regional value chains through strategic purchasing.

Going beyond this economic dimension, our results suggest that successful and locally embedded frms are generally willing to provide substantial (monetary and non-monetary) resources for community purposes. We fnd evidence of the most common activities related to CLRR, such as cultural, social, and educational matters. Therefore, we validate activities identifed in previous studies as central dimensions of frm engagement, which are widely accepted and appreciated activity felds by local/regional initiatives as well as policy-makers (Lengauer & Tödtling, 2010; Vilain, 2010). However, we fnd that frm engagement is driven by a set of diverse motivations that is implicitly and explicitly coupled with their own interests (see Figure 8.2). Aspects such as recruitment/retention of skilled employees and development strategies for young professionals are closely intertwined with challenges of locational attractiveness, which are of particular importance in peripheral regions.


### 170 *Martin Graffenberger and Franziska Görmar*

Considering family frms and non-family frms, the study allows for refections on the specifc role of family ownership for CLRR activities and to expand previous research fndings (e.g., Bürcher, 2017; Lengauer & Tödtling, 2010; Vilain, 2010). Firms generally seem to have a high level of awareness that their engagement may operate as a mechanism to maintain and increase social, cultural, and educational infrastructures and, thereby, actively and consciously shape the attractiveness of place and region. Given the particular context of small towns in peripheral regions, such soft factors are highly relevant for frms, municipalities, and civil society actors, as they constitute determining factors for wider development dynamics of both frms and small towns. Since the spatial focus of engagement is mainly on the immediate frm environments, diverse CLRR activities related to corporate giving and corporate volunteering have particularly pronounced implications for small-town development. CLRR activities, especially in the felds of corporate giving and corporate volunteering, constitute substantial supplements to the activities and investments of municipal and civil society actors and, thereby, open room for manoeuvres to shape and strengthen local and regional structures. We fnd that the ownership nature of the frm (family or non-family) seems not to affect the types of engagements related to corporate giving and corporate volunteering – i.e., all frms use these activities regardless the nature of their ownership.

However, the case studies provide various indications that ownership structures might affect scope and intensity of corporate engagement. Decision-makers of the family-owned Hidden Champions are particularly engaged in terms of comprehensive and strategic corporate support activities. Such comprehensive forms of engagement – e.g., steering group membership in local institutions or participation in municipal strategy processes – can be partly attributed to personal motivations that emerge due to family frms' specifc and individual factors as well as socio-emotional dimensions of embeddedness (Bird & Wennberg, 2014; Campopiano et al., 2014). In this sense, engagement is driven by a sincere sense of obligation and responsibility, resulting in efforts to provide positive returns to the local community. Place related and socio-emotional attachment emerges in the motives of family frms. As decision-takers grew up and still live locally, they have more profound motivations to contribute to the development of their hometowns and to build diverse cultural and social landscapes. These specifc socio-emotional motivations are less pronounced for managers in non-family frms, especially if decision-makers do not live locally or if a high fuctuation can be observed among them.

Furthermore, our study highlights the importance of adopting a procedural understanding of embeddedness (Hess, 2004; Pike et al., 2000). Events such as ownership changes may affect the degree of embeddedness, the perception of a frm's local and regional responsibility, and, consequently, the intensity of CLRR activities; more generally, it can alter the nature of ties to stakeholders (Zellweger et al., 2019). Such aspects become apparent in Holmer's acquisition by and long-term affliation with an international private equity fund. During this period, both frm embeddedness and CLRR activities were substantially reduced. However, the acquisition of Holmer by a French family holding in 2013 facilitated the process of gradual (re-)embedding by defning new trustful ties with the municipality and other local actors. Unlike Holmer, BSW and EJOT refect a high degree of continuity as both frms are third-generation family frms. We interpret continuity as a constantly ongoing embedding process, which facilitates the emergence of shared mindsets and high trust levels (Dörhöfer et al., 2011) that result in institutionalised engagement and joint development of projects and initiatives along the lines of corporate support.

These aspects further illustrate the (potential) role of family involvement in frms, which seems to point to a specifc corporate culture, including intentions to build diverse connections with community stakeholders (Campopiano et al., 2014; Zellweger et al., 2019) – even if family involvement relates to the macro holding-level. On the one hand, these results highlight the diverse and distinct local and regional values of maintained-family ownership of medium-sized frms such as Hidden Champions. On the other hand, our results pose the question of how these values can be effectively secured – e.g., as part of (family-external) company succession or (international) merger and acquisition processes.

Despite their generally positive attitude towards corporate engagement, all frms emphasise that the level and intensity depends on their (current) economic success, which highlights the processual nature of corporate engagement itself (Gallo, 2004; Lengauer & Tödtling, 2010). The latter aspect also has an important implication for municipal and civil society actors, who must avoid being dependent on voluntary sponsoring and support related CLRR activities in case frms restrain their efforts.

# **Contributions**

This chapter offers useful perspectives for policy-makers in the context of small-town development. As frms' and municipalities' goals substantially overlap, opportunities for cooperative action emerge. In this context, the identifcation of joint goals between frms, municipalities, and civil society actors is an essential driver of successful cooperation and steering development processes. Corporate engagement provides a specifc development resource to activate and strategically exploit – especially for peripherally located municipalities and small towns where the maintenance of attractive living environments faces complex and specifc challenges, compared to larger agglomerations. To mobilise these potentials and to tap new/additional resources (fnancial, time, ideas, etc.) municipalities need to fnd ways to channel the obviously existing willingness of frms to engage in local and regional matters. The identifcation and implementation of concrete joint projects that involve private, public, and civil society actors (like the

### 172 *Martin Graffenberger and Franziska Görmar*

community centre in Bad Berleburg) offer promising mechanisms to strategically engage frms based on common goals, envisaged outcomes, agreed responsibilities and resources, and clear time frames.

Thereby, frms can be actively and strategically involved in local development processes, to create mutually accepted outcomes. As we observe in our research, family frms are comprehensively engaged in local development processes. Besides strategically organising the corporate support activities of family frms, a central challenge for municipal actors can be seen in the activation of non-family frms or subsidiaries of larger companies to also engage in these processes. Their purposive activation could mobilise additional resources and facilitate the emergence of distributed and multi-actor CLRR development structures involving both family and non-family frms. When it comes to building such distributed structures, our study highlights that small towns might operate as particularly effective environments: the number of potential frms to be targeted is limited, decision-makers within frms are usually not anonymous and are accessible, and communicative distances are rather small, i.e., associated issues and activities can be discussed quickly via formal and informal channels.

### **Acknowledgments**

The empirical part of this chapter was conducted as part of the project "Hidden Champions – Stabilisierungs- und Entwicklungsfaktoren von Kleinstädten in peripheren Lagen" [Hidden Champions – stabilisation factors and development drivers of small towns in peripheral locations], funded by the Federal Institute for Research on Building, Urban Affairs and Spatial Development (BBSR). We thank our interviewees for their time and interest in the project; Thilo Lang, Lukas Vonnahme, and Lars Porsche for constructive discussions; and the book editors for helpful feedback on earlier stages of this chapter.

### **Notes**

 


### **References**


### 174 *Martin Graffenberger and Franziska Görmar*

*innovativer Unternehmen für Kleinstädte in peripherer Lage* [*Hidden Champions and Urban Development: The Economic and Societal Importance of Innovative Companies for Small Towns in a Peripheral Location*]. Bonn: Federal Institute for Research on Building, Urban Affairs and Spatial Development (BBSR).


# **Part IV Evidence around the world**

# **9 Comparing family and nonfamily frms' strategic effects on regional development**

Evidence from Kenya

*William Murithi and Kassa Woldesenbet Beta* 

# **Introduction**

Academic interest in family frms and regional development has grown recently in Western context (Acquaah, 2016; Basco, 2018). Despite the well-recognised contributions of family frms to regional economic and social development (Basco, 2015; Memili, Fang, Chrisman, & De Massis, 2015; Stough et al., 2015), the understanding of the nature and the type of contribution made by family frms in developing and emerging economies remains under studied (Kolk & Rivera-Santos, 2018). For instance, the African continent remains largely unexplored in spite of family frms' prevalence at the national and regional levels (Khavul, Bruton, & Wood, 2009). There are three main reasons why the phenomenon of family frms in Africa has attracted less academic attention. First, policy-makers and governments are not well aware of family frms' contribution to economic and social development. Second, there is a lack of offcial data (e.g., longitudinal and cross-sectional data) to describe, explore, and show the business ownership structures and family involvement and participation in economic activities (Kolk & Rivera-Santos, 2018). Third, the fnancial commitment for academic research on family frms is inadequate.

We believe that Africa provides a unique context for investigating the signifcance of family frms in regional economic and social development (Khavul et al., 2009; Stough et al. 2015; Zoogah, Peng, & Woldu, 2015) and poverty alleviation (Bruton, Ketchen Jr, & Ireland, 2013). This chapter attempts to explore the impact of family and non-family frms in the context of Kenya. Drawing on a quantitative analysis of 307 frms operating in Kenya, our study fnds that differences in the strategic behaviours (i.e., entrepreneurial behaviours, decision-making strategies, and external social capital) are the underlying mechanisms that may explain how both types of frms differ in their levels of contributions to regional development in developing economies, such as Kenya, and in sub-Saharan Africa. More specifcally, we fnd that entrepreneurial orientation and bridging social capital had positive and signifcant effects on managers' self-perception on frm contribution to regional development. However, there is a negative effect of participative decision-making process on the self-perception of their regional development contribution. That is, family and non-family frms differ in the extent to which they self-perceived their contribution to regional development.

# **Family frms in Kenya**

Family frms form the backbone of Kenyan economic and social development, with prominent medium and large family-run businesses contributing substantially to GDP, employment opportunities, and wealth creation (PricewaterhouseCoopers, 2014). The actual estimation is that family frms account for 60–80% of all employment in Kenya (National Baseline Survey, 1999; Waweru, 2014). According to a PricewaterhouseCoopers report on private frms, 95% of Kenyan family frms that predicted growth were confdent of achieving it, with 32% aiming for aggressive growth over a fveyear period and another 56% expecting steady growth (2014, 6). Contrary to family frms in Western economies, family frms in Kenya (and sub-Saharan Africa in general) face unique ownership, management, governance, and succession structures. Family frms are shaped by a highly collectivist culture (Hofstede, 2001), diverse social structures (Kamoche, Siebers, Mamman, & Newenham-Kahindi, 2015), and the coexistence of both formal and informal institutions (Murithi, Vershinina, & Rodgers, 2020) that may boost and/or constrain economic activities. Further, as opposed to Western countries where studies refer to the nuclear family as a majority owner, Kenyan family frms have members from the extended family and communities that infuence business behaviour and performance (Khavul et al., 2009).

As a developing economy, Kenya's political, economic, and culture patterns differ from other developing Western or Asian economies (Zoogah et al., 2015; Vershinina, Kassa, & Murithi, 2018). The weak political and institutional environments found in Kenya encourage informal economic activities (Khavul et al., 2009; Murithi et al., 2020; Murithi, 2019). Most small and medium frms operate at the intersection of formal and informal institutions (Waweru, 2014; Murithi, 2019). Like other countries in sub-Saharan Africa, Kenya has a unique socio-cultural landscape, deep rooted in traditions and cultural contexts that infuence wider management and entrepreneurial practices (Zoogah et al., 2015; Vershinina et al., 2018). In particular, the "harambee spirit" with its various aspects can enable or constrain entrepreneurial behaviours in Kenya (Vershinina et al., 2018). Therefore, the country presents a unique context to explore the differences between family and non-family contributions to regional development.

# **Theoretical framework and hypothesis**

# *Strategic behaviours and regional development*

Strategic behaviours are defned as the actions taken by frms or their toplevel managers that are intended to infuence the market environment in which they compete (Kuratko, 2012). The strategic behavioural differences between family and non-family frms can be understood by investigating the infuence of controlling family on business decisions, strategy making, and performance (Abdellatif, Amann, & Jaussaud, 2010). Family and nonfamily frms differ in their strategic behaviours because family involvement alters their goals (Basco, 2017). Family frms, like their non-family counterparts, pursue business economic goals but also family-oriented ones, such as employment for family members, support to local communities, and the intention to transfer the family frm from one generation to another.

The differences in strategic behaviours between family and non-family frms are likely to determine their outcomes and, therefore, the way both types of frms contribute to regional development. In general, the strategic behaviour of frms is associated with decision-making and its implementation to achieve desired goals and outcomes. It is in this process that family involvement and control affects the frms' strategic entrepreneurial behaviours, such as wealth creation and advantage-seeking actions (Hitt, Ireland, Sirmon, & Trahms, 2011). We conceptualised strategic behaviours of the frm as the strategic posture involving entrepreneurial orientation (Covin, Green, & Slevin, 2006), participation in strategic decision-making (Eddleston & Kellermanns, 2007; Basco, 2013), and developing and using resources embedded in social networks (Acquaah, 2012; Mani & Durand, 2019). This chapter examines whether family and non-family frms differ in these three dimensions of strategic behaviour and how such differences, if they exist, affect frms' self-perception of their contribution to regional development in context of Kenya.

### *Firm entrepreneurial behaviour*

Strategic entrepreneurship theory, which focuses on the entrepreneurship and strategic management aspects of opportunity recognition and wealth creation, predicts that family frms are the major contributors to economic development (Hitt et al., 2011). This is because family frms are able to orchestrate resources that lead to individual-, frm-, and regional-level outcomes (Hitt et al., 2011). In addition, the literature on corporate entrepreneurship has highlighted the differences between family and non-family frms concerning their entrepreneurial orientation (Casillas & Moreno, 2010). Studies that used a multi-dimensional approach (focusing on the dimension of proactiveness, risk-taking, innovativeness, competitive aggressiveness, and autonomy) found that family frms score low on the fve dimensions of corporate entrepreneurship (Zellweger, Muhlebach, & Sieger, 2010). Short et al. (2009) also found that although family frms do exhibit language consistent with all entrepreneurial orientations, their entrepreneurial orientation language in relation to autonomy, proactiveness, and risk-taking is lower than in non-family frms.

However, as most of these studies focused on either the individual (ownermanager) or frm-level analysis, they underestimated the infuence of the family stakeholders' entrepreneurial orientation on family frms (Dyer, 2006). Zellweger, Nason, and Nordqvist (2011) thus argued for the consideration of the family as a unit of analysis, as this would help further our understanding of the ability of family frms to generate transgenerational value. In addition, Basco and Pérez Rodríguez (2009) also called for a holistic view of the family frm by integrating the family and business systems. These authors recommend that researchers should focus on four important aspects of the family frm—strategy, human resources, governance, and succession—in order to understand how family infuences the frm's strategic decision-making.

Indeed, the current conceptualisation of entrepreneurial orientation at the frm level is limited in its explanatory power to show the frm's infuence at regional level. We know more about the effects of entrepreneurial orientation on frm performance, such as the growth and proftability (Gupta & Gupta, 2015), but less on regional levels. Further, evidence from emerging economies (e.g., China) shows that not all the dimensions of entrepreneurial orientation have equal levels of signifcance because entrepreneurial orientations could be shaped by contexts (Welter, 2011; Basco, 2018). Such limitations led to calls for further exploration of frm level entrepreneurship in developing economies (Kantur, 2016). Businesses operating in developing economies also face increased uncertainty, imperfect competition, and hostile environments, which require the deft use of entrepreneurial orientations. Recent studies have also found a positive link between frm-level entrepreneurship within developing economies and frm performance (Cai, Liu, Deng, & Cao, 2014). We argue that the frm level of entrepreneurship will affect regional development, through performance and social engagement. Our frst hypothesis is as follows:

*H1: Firm entrepreneurship orientation effect on regional development outcomes is higher in family frms than in non-family frms.* 

### *Firm strategic decision-making process*

The basic strategic management processes between family and non-family frms may look similar (Miller & Le Breton-Miller, 2011). However, there are differences on the goals to be pursued, the process of resource allocation to achieve those goals, and who participates in the decision-making (Chrisman, Chua, & Sharma, 2005; Basco, 2013). The involvement of the family in businesses and the goals they pursue account for differentiated outcomes in both types of frms (Lee, 2006). In family frms, family members are the main stakeholders likely to infuence the strategic management process in comparison to non-family frms, where different shareholders infuence the decision-making process and where they follow the proft maximisation goal. Furthermore, family stakeholders play an important role in strategic renewal, entrepreneurship, frm growth, innovation, and performance of family frms (Mazzi, 2011). It is a taken-for-granted fact that family (as individuals or as a group) are willing to retain the ownership and control of the frm in order to remain autonomous. Families always seek autonomy for control purposes over ownership and management, even in the context of poor fnancial performance (Gómez-Mejía et al., 2007). In doing so, a controlling family coalition pursues family-centred fnancial and non-fnancial objectives aimed at creating generational wealth (Carney, 2005; Aparicio, Basco, Iturralde, & Maseda, 2017).

According to Covin et al. (2006, 59), "Strategic decisions are made through consensus-seeking versus individualistic or autocratic processes by the formally responsible executive". Prior studies have found that strategic decision-making through teamwork participation, interaction, and regular consultation with employees, as well as free and open exchanges have positive infuence on frm performance (Eddleston & Kellermanns, 2007). Drawing from the stewardship theory, Eddleston, Chrisman, Steier, and Chua (2010) found that participative strategic decision-making and the focus on long-term orientation positively enhance the level of entrepreneurship. Firms that encourage information and knowledge sharing about their specifc process tend to be more innovative and effcient. Sharing information with other team members encourages a collective responsibility that allows members to participate in the development of organisational strategies. Therefore, stronger frm performance is likely to be associated with a participative process and frm involvement that will promote regional development. Based on the discussion above, our second hypothesis is as follows:

*H2: The effect of participative strategic decision-making on regional development is higher in family frms than in non-family frms.* 

# *Firm bridging social capital*

Family frms are an integral part of the community and institutions in Africa. We argue that their embeddedness in community and institutional environments would allow them to make signifcant contribution to socioeconomic development. Through their engagement in businesses, socially oriented enterprises, and many economy-related activities, business families in sub-Saharan Africa are in a better position to contribute to socioeconomic development. This is because they combine different sociological and cultural qualities to overcome institutional voids prominent in such economic contexts (Murithi et al., 2020). Empirical evidence from family frms in emerging economies has shown that top-level managers develop strong social capital with managers from other frms, communities, and political leaders, which increases their community involvement (Acquaah, 2012; Mani & Durand, 2019). Family frms represent both the family as an institution and the business as an organisational form when engaged in entrepreneurial activities. Family infuence in the African, particularly in Kenyan context, goes beyond the nuclear family. "Family" involves the kin relationships and those in industry member associations and government agencies. Hence, such developed social networks provide strategic resources for entrepreneurial and socially oriented activities with positive effects on development (Adusei, 2016). Therefore, family frms would be able to contribute to regional development outcomes in the African context.

Murithi et al. (2020) argue that business families are in a better position to infuence the economic development in sub-Saharan Africa because of their ability to navigate institutional boundaries. Hence, family frms in Africa can manage underdeveloped and dysfunctional formal institutions by using informal institutions such as social relationships and networks (Acquaah, 2012). As family frms operate at the intersection of family and community, and formal and informal institutions, there is a greater chance of them contributing to regional development in comparison to non-family frms. It is important to consider the network resources accessible because of a family's social capital. Familial ties and societal relationships help in bringing in important resources for business. McGrath, O'Toole, Marino, and Sutton-Brady (2018) argue that entrepreneurs and family frms must actively pursue related connections, "beyond early social network ties, to gain access to and use a broader pool of resources and capabilities external to the frm" (p. 523). To access external resources, family frms must be willing to collaborate with other economic and social actors (Dyer & Singh, 1998, 672), to have the ability to coordinate competencies and combine knowledge across (organisational) boundaries, and to sustain its innovativeness by creating and managing the overall network architecture over time (Capaldo, 2007, 585). Therefore, we posit the following hypothesis:

*H3: The effect of bridging social capital on regional development outcomes is higher in family frms than in non-family frms.* 

# **Methodology**

To explore the differences and potential strategic behaviours that may explain the extent to which family and non-family frms contribute to regional development based on their self-perception, the study deployed an exploratory survey, which is best suited to this kind of study (Hair et al., 2006). Using a structured survey questionnaire, quantitative data was collected from top-level managers of 307 privately held Kenyan frms.

### *Measures*

The dependent variable is the contribution that the frm makes to regional development outcomes. We defne regional development as the application of economic processes and resources available to the region that result in sustainable development and desired economic outcomes for the region (Stimson, Stough, & Roberts, 2006). Bishop, Mason, and Robinson (2009) conceptualised frm contribution to regional development as "direct effects of individual frms to employment, aggregate output (goods and services), and productivity growth as refected in their investments in resources and development capabilities". In this study, we use three regional outcome measures, drawing on the existing literature: contribution to GDP, employment, and wealth creation (Thurik, Wennekers & Uhlaner, 2002). Each item was measured using 7-point Likert scale (7 for the highest degree of agreement and 1 for the lowest). These are subjective measures as perceived by the top-level managers about their frms' contribution to regional development.

The independent variables are at the frm level and focus on the dimensions of strategic behaviours: frm entrepreneurial orientation, participation in strategic decision-making processes, and bridging social capital relationships. Entrepreneurial orientation is measured using fve dimensions: innovativeness, proactiveness, risk-taking, competitive aggressiveness, and autonomy (Casillas & Moreno, 2010). The participative strategic decision-making processes were measured using fve items drawn from extracted from Eddleston and Kellermanns (2007). Finally, bridging social capital items were drawn from Acquaah (2012). All items were captured using 7-point Likert scale.

We control for frm age and frm—both categorical variables. Firm age was measured with six categories: less than 5 years, 6–9 years, 10–14 years, 15–19 years, 20–24 years, and above 25 years. Firm size was measured using the number of employees in the following categories: micro (1–9), small (10–49), medium (50–249), and large (250 and above). Finally, the geographic distribution of sales was measured using fve levels: county, national, regional, African, and global.

### *Data analysis*

To analyse the dataset, we adopted contemporary structural equation modelling techniques (Hair et al., 2006). We classify a frm as family or non-family with the following two criteria. First, the respondent's self-identifcation as either a family or non-family frm (Westhead & Cowling, 1998). Second, the level of family involvement (Chua, Chrisman, & Sharma, 1999). In the later classifcation, fve variables were used to collect the responses based on the frm ownership, management, and governance, involvement of transgenerational family members and intrafamily succession. Of the 307 frms included in the sample, 40.4% (*n* = 124) self-identifed as family frms, whilst 59.6% (*n* = 183) self-identifed as non-family frms. Consequently, frms with high family involvement represented 31% (*n* = 96), while those with low family involvement represented 69% (*n* = 211).

Following the recommendation made by Bryman (2016), a pilot study was conducted using a sample of frms drawn from Strathmore Enterprise Development Centre, which contained frms of the similar characteristics as the main sample used. The analysis of the pilot study indicated that the items measure the expected constructs. We also tested for internal reliability with a Cronbach alpha coeffcient. All values are above the required minimum of 0.6 (entrepreneurial orientation = 0.88, decision-making strategy = 0.765, bridging social capital = 0.83, and regional development = 0.878).

The analysis of the demographic characteristics of the sample shows that the majority of the family frms are older (more than 25 years old) while non-family frms are younger (less than 5 years old on average). In terms of size, as measured by the number of employees, the majority of the frms in both categories are medium- to large-sized frms. Most family and nonfamily frms operate in national and regional markets.

Inferential statistical analysis was conducted using exploratory factor analysis and confrmatory factor analysis (Hair et al., 2006). The frst stage of analysis was done using the former, which relies on estimation of the factors and the contribution of the variables (items) to the factor loadings as a basis for identifying variables for subsequent analysis (Steinmetz, Davidov, & Schmidt, 2011). Table 9.1 shows the four latent constructs identifed with their communalities, Cronbach Alpha, composite reliability, and AVE coeffcients. To assess the measurement model, we used the goodness of ft indices (Table 9.2) as specifc evidence of construct validity (Hair et al., 2006).

After specifying the measurement model, we adopt the competing equivalent model structure. That is, a single model (i.e., set of relationships) is specifed and then alternative formulations of the underlying theory are identifed (Hair et al., 2006). Firstly, the model (Table 9.3) shows that the three independent variables (entrepreneurial orientation, decision-making process, and bridging social capital) have direct relationships with regional development outcomes. Secondly, all the three independent variables reported some degree of interaction among themselves that would give rise to their own inter-relationships.

 Table 9.3 reports the hypothesised relationships on the differences between the family and non-family frms in their contribution to regional development, based on self-perception. In relation to entrepreneurial orientation, family frms reported a higher and signifcant contribution to regional development (*β* = 0.278, *p* = 0.002) compared to non-family frms, which reported a positive but not signifcant effect on regional development (*β* = 0.109, *p* = 0.287). However, participative decision-making within family frms had a statistically negative effect on regional development compared to non-family frms (*β* = −0.107, *p* = 0.201), hence indicating differences between both types of frms. In terms of the bridging social capital effect, non-family frms make a statistically signifcant contribution (*β* = 0.291, *p* < 0.01) to regional development when compared to family frms (*β* = 0.138, *p* = 0.097).

Based on the aforementioned fndings, we confrm that entrepreneurial orientation, participative decision-making, and bridging social capital provide a holistic explanation on the differences between family and non-family frms' contribution to regional development.


*Table 9.1* Indicator loadings and composite reliability of variables and indicators

### 186 *William Murithi and Kassa Woldesenbet Beta*


*Table 9.2* Fit indices, measurement model (1), structural model (2), interaction model (3) and model criteria



Notes: \*\*\**p*-value < 0.01; \*\**p*-value < 0.05; \**p*-value < 0.10.

### **Discussion and conclusion**

The differences between family and non-family frms in terms of their contribution to regional development can be explained by their strategic behaviours. Even though both types of frms show that their entrepreneurial orientation is positively related to regional development, family frms reported a signifcant positive effect compared to non-family frms. Thus, this study's fndings, while consistent with the previous fndings that suggest that family frms are less entrepreneurial than non-family frms (Gómez-Mejía et al., 2007) at the frm level, show that family frms' entrepreneurial orientation effects on regional development is positive and signifcant. The longterm orientation of family frms and concern for socio-emotional wealth (Gómez-Mejía et al., 2007) prompt entrepreneurial decisions aimed at sustainability rather than short-term gains and this is likely to contribute to regional development (Zahra, Hayton, & Salvato, 2004). Hence, family frms may prefer to forgo short-term performance (fnancial gains) in order to maintain their ownership control over the frm and this consideration may create sustained value for regional development.

Our research goes beyond establishing that family frms have differentiated contribution to regional development outcomes; it also shows the underlying factors that contribute to the observed differences. Our study is the frst of its kind in the Kenyan and sub-Saharan African contexts that empirically shows the differences on the level of entrepreneurial orientation by family and non-family frms and its effect on regional development. It provides an interesting insight into the greater effect family frms' entrepreneurial orientations have on regional development in comparison to non-family frms in Kenya.

Our fndings can be explained by the level of family involvement. The interaction of the family, business, and context may enhance family frms' contribution to regional dimensions, compared to non-family ones. Even though this study shows a positive and sustained effect of family frm entrepreneurial orientation on regional development, we caution against hasty conclusions, as there may be circumstances where such positive contributions don't exist. These situations may include confict within a family such as diffculties in succession or lack of strategic renewal upon the exit or death of a founding family member(s).

Regarding participatory and interactive decision-making (Eddleston and Kellermanns, 2007), this study establishes that family frms' behaviours have a signifcant negative effect on regional development compared to non-family frms. Therefore, these fndings contradict recent studies, which argue that the participation of family owners in decision-making would positively enhance regional development (Basco, 2015; Stough et al., 2015). These negative effects on regional development can be explained by two factors. First, the participation of extended family members in the decision-making process slows down proactive strategies aimed at resource confguration and innovative activities for organisational performance and its effect on regional development. Second, a consultative strategy-making process, typical in family frms in developing economies, culminates with the founders' blessings in such a way that it promotes family-centred economic and non-economic goals.

The fndings show that bridging social capital has a direct, strong, and positive effect on regional development outcomes. However, its effect on non-family frms was more pronounced than on family frms. Family frms' utilisation of network social capital for navigating institutional voids and strengthening their social bonds may not necessarily equate with non-family frms' proactiveness in drawing strategic resources aimed at frm performance with a potential impact on regional development. Hence, this study speculates that family frms' bridging social capital may have long-run effect in comparison to short-term effects on regional development. Another potential explanation for the above fndings may be that the family becomes an institutional investor with a capacity to leverage institutional resources to generate wealth in the long run. Further, their strong embeddedness in community network structures and pro-social objectives might not be captured by the study—the effect of which could amplify their role in local socioeconomic development in the long-term. Instead, family frms might use bridging social capital to cope with institutional voids, as a buffer from uncertain markets, to make up for the lack of professional management, and to preserve family wealth.

### 188 *William Murithi and Kassa Woldesenbet Beta*

According to Murithi et al. (2020), focusing on the family (instead of family frms) in sub-Saharan Africa shows that the family and the business are not competing sub-systems, as argued in some existing studies. Instead, they complement each other and enable businesses to navigate the complex institutional environment. For instance, the involvement of the extended family becomes a major source of capital, expertise, and information, which can be a result of the informal entrepreneurial activities at the family level (Murithi et al., 2020). Evidence from small frms in Kampala and Uganda showed that the larger the entrepreneur's network (i.e., kinship networks), the more resources they could raise. However, this comes at a higher cost (Khayesi, George, & Antonakis, 2014), showing that the effects of social capital on frm performance can be either positive or negative because of the need to maintain family ties and network relationships, which is very costly in an African context. The family patriarch or matriarch is responsible for the maintenance of family members and contribution to community welfare because of the expectations of the collectivist culture. The high-level dependence on family frms by family members and communities may limit their ability to accumulate resources aimed at frm growth and, hence, restrict their contribution to regional development. Such an analysis suggests that the pathways in which the family frms contribute to regional development are complex and intertwined.

### **Contributions**

This chapter presented comparative evidence on family and non-family frms' contributions to regional development in Kenya. The key conclusion is that the family context is a signifcant moderator between frms' strategic behaviour, frms' self-perception, and its contribution to regional development.

Theoretically, we can highlight three main implications. First, the conceptualisation of family frms in the African context. The way in which family frms are defned in the Western context is less likely to capture the essence of African family frms because these families extend beyond the nuclear family and include kinships and extended family relationships. Hence, the way in which the family frms are conceptualised in the African context can have several implications for management, governance, succession, and strategic choice literature. Second, African family frms operate not only at the intersection of formal and informal institutions but also in environments characterised by institutional voids. Such institutional contexts merit further theoretical and empirical work to understand how family frms navigate the duality of institutions and institutional voids, how they respond to or cope with volatile changes in the institutional environments, and how family frms perceive their contribution to regional development. Third, our fndings show that family and non-family frms contribute differently to regional development in the Kenyan context based on their self-perception and that these differences can be explained in reference to their strategic behaviours.

Given the signifcant contributions family frms make to regional economic and social development, policy-makers should pay attention to how family and non-family frms contribute to regional development. A better understanding of how economic actors behave and react to the environment can help policy-makers tailor their regional and local policy and increase their impact in communities. National, regional, and local public institutions should be aware of the importance of having reliable data to improve the knowledge that we have about our economies. Public decision-making needs information to develop policies and to alleviate regional disparities or inequalities by engaging local actors.

### **References**


### 190 *William Murithi and Kassa Woldesenbet Beta*


# **10 Family frms and regional entrepreneurship**

The European evidence

*Riccardo Cappelli, Marco Cucculelli and Valentina Peruzzi* 

### **Introduction**

Family frms are the most predominant form of business organization and account for a large proportion of employment, business turnover and gross domestic product (Bjuggren, Johansson, & Sjögren, 2011). This relevance has led researchers to focus on the behavioral differences between family and non-family frms, and on the role played by some family-specifc attributes in explaining the former's proftability and productivity (Anderson & Reeb, 2003; Sraer & Thesmar, 2007), fnancing and investment decisions (Anderson, Mansi, & Reeb, 2003; Anderson, Duru, & Reeb, 2012) and internationalization and innovation activities (Minetti, Murro, & Paiella, 2015; Pongelli, Caroli, & Cucculelli, 2016).

The literature has mainly focused on micro-oriented research at the frm and family levels, whereas the role played by family ownership and industry attributes at the aggregate level has received less attention (Basco, 2015). This is particularly evident in the case of regional entrepreneurship, as the extant literature has scarcely explored the connections between local entrepreneurial activity and specifc exogenous industry characteristics, like the industry structure and the density of family-owned frms.

The aim of this chapter is two-fold. First, it aims at analyzing whether family frm density, together with industry scale, affects entrepreneurship at the regional level. Second, it investigates whether the development stage of the economy plays a role in this relationship. As this last issue is central to our analysis, we explore the different stages of development of European regions to identify the infuence of structural industry characteristics on regional entrepreneurship rates across European geographical areas.

The rationale behind this research study is to connect three complementary streams of analysis: the industry scale, the ownership structure of companies and the stage of industrial development of European regions. For the frst, previous studies have demonstrated that a high regional concentration of large scale industries – i.e., the industry scale – negatively affects entrepreneurship (Stuetzer et al., 2016), whereas the prevalence of small businesses has been associated with a higher level of entrepreneurship (Reynolds, Storey, & Westhead, 1994; Lee, Florida, & Acs, 2004; Fritsch & Falck, 2007).1 This makes the industry structure a signifcant external boundary condition in assessing the development of entrepreneurial activity at the regional level.

As for the second line of research, we argue that in addition to the industry scale, the industry composition – in terms of the prevalent type of ownership – also matters for entrepreneurship, as it can infuence the incentives behind individual decisions and, in turn, local entrepreneurial activity (Villalonga & Amit, 2010). A large presence of family frms makes the long-term orientation a pervasive feature of incumbent companies, thus supporting risk-taking in the economy, which, in turn, fosters entrepreneurship (Zahra, Hayton, & Salvato, 2004). Similarly, family frms are very well suited to promote interaction between founders and workers, an aspect that makes it easier for the latter to learn how to start and run a frm.

Finally, as for the third research area, recent evidence on the impact of family ownership on innovation helps to explain why the infuence of industry factors on entrepreneurship may run differently depending on the stage of regional development (Cucculelli & Peruzzi, 2020). In regions that are in an early stage of industrial development, foreign direct investments may fuel economic growth because of employment opportunities and other spillovers that are poured into the local economy by external large players. Multinational frms – typically large-sized frms – may also generate positive spillovers on host regions in terms of diffusion of technologies and management practices, thereby fostering local entrepreneurship. Moreover, large external companies may also stimulate local entrepreneurship through the sub-contracting of productive tasks from a vertically integrated process to small-sized satellite businesses. All these aspects make the stage of development of the local economic environment a signifcant factor in shaping the relationship between industry characteristics – including company ownership – and the process of new frm creation (Smallbone & Welter, 2001).

In this context, the chapter aims at checking whether and to what extent selected industry-level variables – i.e., the average frm size (regional industry scale) and the share of family owned frms (regional family frm density) – matter for regional frm birth rate (regional entrepreneurship), and whether this relationship is facilitated or inhibited by the stage of development of the local economic environment. To test these intuitions, we drew information from two main datasets. First, the Eurostat regional database, which provides the main regional economic and demographic variables (e.g., the frm birth rate and the average frm size in the manufacturing sector) for European countries. Second, the BvD-Amadeus dataset, which provides frm level information (e.g., ownership information) about a large sample of European frms. By merging the two, we ended up with a fnal sample of 121 regions, located in 13 European countries: Bulgaria, Czech Republic, Estonia, Hungary, Romania, Poland and Slovakia (classifed as Eastern European); Denmark, Spain, Finland, France, Italy and Portugal (classifed as Western European).

To summarize our fndings, estimation results indicate that the average frm size and the share of family frms in the regional industry are negatively and positively associated, respectively, with higher regional frm birth rate. This means that entrepreneurship reacts positively to a more pervasive presence of family frms and negatively to the presence of large companies. However, these results only hold in Western European regions, whereas in Eastern Europe, entrepreneurial initiatives are positively and signifcantly driven only by the average frm size and the density of family ownership does not play a substantial role.

# **Theoretical background**

### *Industry structure and entrepreneurial activity*

Few theoretical and empirical studies have investigated the relationship between industry structure and entrepreneurial activity (e.g., Bauernschuster, Falck, & Heblich, 2010; Wyrwich, 2013; Glaeser, Kerr, & Kerr, 2015; Stuetzer et al., 2016). In this research area, the classical approach to entrepreneurship is the one put forward by the literature on Industrial Organization (IO). According to this perspective, new frm creation depends on industry characteristics such as the level of capital intensity, the average minimum effcient size (MES) and the level of irreversible investment (i.e., the sunk costs in the case of exit or failure). For instance, large-scale industries are typically associated with high levels of capital intensity, large amount of irreversible investments and the existence of large minimum effcient size, all of which are negatively related to frm entry. Therefore, a key fnding of this framework is that large-scale industries only provide a limited number of entrepreneurial opportunities (Siegfried & Evans, 1994; Geroski, 1995).

Within the IO approach to new frms and startups (Santarelli & Vivarelli, 2007), one strand of research has focused on the role of frm size in explaining entrepreneurial phenomena (Parker, 2009; Elfenbein, Hamilton, & Zenger, 2010). These studies, which are closely related to IO literature because of the association between average frm size and industry scale (Beesley & Hamilton, 1984), have built on the idea that the opportunity cost of starting a frm is relatively low when working in a small business, due to low wages, reduced job security and limited opportunities for professional advancements.

In this framework, a number of reasons have been suggested that push people into small frms and then promote their entrepreneurial behavior. First, employees with personality traits that are conducive to entrepreneurship self-select into small frms because of the close similarity in the work environment (Stuetzer et al., 2016). Second, entrepreneurial skills that emanate from the ability of people to perform a broad spectrum of diverse tasks and challenges are higher in small enterprises where the division of labor is low and workers are not highly-specialized (Wagner, 2004; Hyytinen & Maliranta, 2008; Bublitz & Noseleit, 2014; Lechmann & Schnabel, 2014). Third, the propensity to start one's own frm might be stimulated by direct interactions with an entrepreneurial role model, which is more likely to occur in very small frms than in large enterprises. Through such interactions, young workers learn about the process of starting and running a frm, thereby improving their entrepreneurial human capital (Minniti, 2005; Fritsch, 2013).

Although large-scale industries affect entrepreneurial activity at the industry level, there is some evidence that they also determine the regional level of entrepreneurship. Davidsson (1995) analyzed regional entrepreneurial activity in Sweden and shows that small frm density is positively and signifcantly associated with new frm creation because of good supply of entrepreneurial models, relevant work life experience, and access to practical know-how on how to get a business going. Keeble and Walker (1994), who examined the geographic distribution of new frms and closing frms in the United Kingdom for the period 1980–1990, suggest that, in addition to population growth, housing wealth, professional expertise and urbanization, the presence of small frms fosters entrepreneurial initiatives. Similarly, Reynolds (1994), who looked at 382 labor market areas in the United States, shows that the impact of small frms on the annual birth rate of independent businesses is positive and statistically signifcant, but tends to vary for different economic sectors and the type of labor market area.

The aforementioned fndings have been confrmed by more recent studies. By analyzing labor market areas in the United States, Armington and Acs (2002) found that new frm creation is substantially explained by regional differences in income growth, population, industry intensity and frm size. Similar results are shown by Stuetzer et al. (2016), who analyzed the entrepreneurial activity of British regions and found that the ones with high employment shares of large-scale industries in the 19th century have lower entrepreneurship rates and weaker entrepreneurship culture today. All these contributions make regional heterogeneity in the endowment of small and large frms a crucial issue to be investigated, to understand how entrepreneurship develops within a region.

### *Family ownership and regional entrepreneurship*

Most empirical studies on regional entrepreneurship only partially consider the heterogeneous nature of frms behind the entrepreneurial activity. We believe that a key distinction here might be between family and non-family frms, because the former exhibits a stronger social network within their region and are themselves more likely to be infuenced by their environment's community (Stough et al., 2015).

Recently, a growing number of studies have started to analyze the intersection between the entrepreneurship and the family business felds by examining the impact of the family environment on entrepreneurial activities and processes. This literature considers the family frm a highly entrepreneurial organizational context, that is dynamic and change-oriented because of the long-term mindset of family owners and managers. Consistently with this view of family ownership, by using data from 536 U.S. manufacturing companies, Zahra et al. (2004) showed that family frms' culture is an important strategic resource that gives these frms a competitive advantage over their rivals by promoting and sustaining entrepreneurial activities. Similarly, Kellermanns and Eddleston (2006), who focused on a sample of 232 family businesses in the northeastern United States, indicated that a willingness to change and technological opportunity recognition are positively related to entrepreneurship in family frms. Finally, Aldrich and Cliff (2003) found that family dynamics is the most important factor in recognizing entrepreneurial opportunities, because of the exclusive access to resources, norms, attitudes and values of family frms.

Berger and Luckmann (1967) argued that the family can be considered the strongest social institution when it comes to passing on values and attitudes to its members. In the context of family frms, the values that are transmitted to family members are specifc knowledge, skills, abilities, attitudes and motivations on how to run a business. Because of their involvement in the frm's ownership and management, family members become endowed with a higher tolerance of failure and risk-taking behaviors. However, the transmission of human and entrepreneurial capital in the context of family frms is also extended to non-family employees and managers. Through close and frequent interactions with the family frm's founder, young workers learn about the process of starting and running a frm, thereby becoming more likely to start their own in the future. Consistent with this view, Fairlie and Robb (2007) found that in the United States, children of business owners are substantially more likely than others to become self-employed. Moreover, Alsos, Carter, and Ljunggren (2014) highlighted the importance of the entrepreneurial household in business creation and growth by demonstrating that farm households play an important role in business incubation by providing resources, knowledge and emotional encouragement.

So far, studies on the relationship between family ownership and entrepreneurial activities have been mainly micro-oriented, thus neglecting the role of family frms in explaining regional outcomes (Stough et al., 2015). Considering the regional density of family frms, we provide an empirical evidence to shed light on the importance of family frms for regional entrepreneurship.

#### *Testable hypotheses*

The aim of our chapter is to investigate the role played by industry scale and family ownership prevalence at the regional level in promoting entrepreneurial initiatives in European regions. In line with the aforementioned theoretical and empirical literature, we expect that the industry scale, i.e., the average frm size at the industry level, is negatively associated with the birth rate of frms at the regional level, whereas the density of family frms has a positive impact. Hence, we formulate our hypotheses as follows:

Hypothesis 1: *Average frm size is negatively associated with the frm birth rate at the regional level.*  Hypothesis 2: *Family frm density is positively associated with the frm birth rate at the regional level.* 

However, as far as the infuence of the industrial structure is concerned, we expect that the relationship between average frm size and entrepreneurial activities may be different in European regions at a lower industrial development stage. Chinitz (1961) used this intuition to explain the different level of entrepreneurship between New York and Pittsburgh, which resulted from the dissimilar processes of past economic development of the two cities. The historical concentration of industries characterized by large, massproduction frms (e.g., the steel industry) hindered entrepreneurial activity in Pittsburgh. On the other hand, the development of decentralized industries (e.g., the garment industry) characterized by the presence of small independent frms fostered entrepreneurship in New York. Using this evidence and exploiting the same underlying rationale, a recent paper by Stuetzer et al. (2016) showed that entrepreneurship activity is actually shaped by the industry structure in the UK, where the presence of large-scale industries in a region is associated with a lower rate of entrepreneurial activity in that area.

Borrowing from Stuetzer et al. (2016), we therefore assume that the industry average frm size may affect entrepreneurial activity in European areas differently according to their stage of industrial development. Large foreign frms that are mainly responsible for the increase of the average frm size in Eastern European regions actually play a crucial role in these areas because of the positive externalities – in terms of diffusion of technologies and management practices – on local businesses, which, in turn, can foster local entrepreneurship. Moreover, we may expect that also the outsourcing of production activities to local agents by large foreign and state-owned enterprises may drive new entrepreneurial activities, along a process of division of labor spurred by the re-organization of vertically-integrated plants. Hence, we test the following additional hypothesis:

Hypothesis 3: *The relationship between average frm size and frm birth rate may be reversed in Eastern European regions.* 

# **Data, variables and method**

### *Dataset*

To construct regional variables, we rely on two databases. The Eurostat regional database provides the main economic and demographic variables (e.g., the regional frm birth rate and the regional share of employees in the manufacturing sector) for European NUTS2 regions (Eurostat, 2011) for the period 2011–2016. The Bureau van Dijk–Amadeus database provides information on ownership and headquarter location about European companies, which we use to compute the regional share of family frms in 2009. Combining the two databases, we obtain a fnal sample of 121 regions of 13 European countries, i.e., Bulgaria, Czech Republic, Denmark, Estonia, Spain, Finland, France, Hungary, Italy, Poland, Portugal, Romania and Slovakia.<sup>2</sup>

### *Dependent variable*

We measure annual regional entrepreneurship in terms of frm birth intensity in the industry sector (NACE Rev. 2, B-E) (*Regional frm birth rate*), calculated as the ratio between the sectoral regional number of frm births 3 for each year of the reference period 2011–2016 and the sectoral regional number of existing frms in the same year. Due to data constraints, the number of observations on the annual regional frm birth rate (expressed in percentages) ranges from 1 (for 18 regions) to 6 (for 56 regions).

Map 10.1 shows the distribution of the average annual frm birth rate over the period 2011–2016 in the 121 regions considered in our analysis. It shows

 *Map 10.1* Regional birth rate of industry frms (*Regional frm birth rate*) – average values for the period 2011–2016.

# 200 *Riccardo Cappelli et al.*

that regions with the lowest levels (less than 6%) of frm birth rate are located mainly in Italy, Finland and Spain, while regions with the highest levels of frm birth rate (higher than 10%) are located mainly in East European countries like Poland, Romania and Slovakia. Overall, a strong country effect emerges, which will have to be accounted for in the empirical analysis.

# *Industry scale*

To measure the scale of the manufacturing sector (NACE Rev.2, 10–33) at the regional level, we use the average frm size (*Regional frm average size*). This variable is computed as the ratio between the regional number of employees in the manufacturing sector in 2009 and the regional number of frms in the same sector in 2009. Map 10.2 shows the distribution of the average frm size in the 121 regions with available data. Regions with the largest frm sizes (higher than 20 employees) are located mainly in Slovakia and Romania, while those with the smallest frm sizes (less than ten employees) are located mainly in the Czech Republic, Portugal and in the southern regions of Italy, France and Spain.

# *Family frm density*

Family frm density at regional level is computed using information on ownership and the headquarters location of frms in the manufacturing sector,

 *Map 10.2* Regional average size of manufacturing frms (*Regional average frm size*) in 2009.

 *Map 10.3* Regional density of family manufacturing frms (*Regional family business density*) in 2009.

as provided by Bureau van Dijk–Amadeus. First, to distinguish family frms from non-family frms we use the ownership equity share. We consider a frm as a family frm when the largest shareholder (top 1) in 2009 was an individual or a family. Additionally, we use the NUTS 2 code of the company's headquarters location to assign it to a particular region. Regional family frm density (*Regional family business density*) is the ratio between the regional number of family frms in the manufacturing sector and the regional total number of frms in the manufacturing sector.

Map 10.3 shows the distribution of family frm density in the 121 regions with available data. Regions with the highest shares of family frm density (higher than 0.4) are located mainly in Bulgaria, Italy, Poland and Spain, while regions with the lowest shares of family frm density (less than 0.2) are located mainly in Denmark, France, Portugal and Romania.

### *Econometric specifcation*

We model regional entrepreneurship using the annual frm birth rate in the industry sector (*Regional frm birth rate*). Our two main variables of interest capturing the industry structure of regions are average frm size (*Regional frm average size*) and family frm density (*Regional family business density*) in the manufacturing sector. We controlled for several factors that

### 202 *Riccardo Cappelli et al.*

affect frm *Regional frm birth rate* and that could be correlated with *Regional frm average size* and *Regional family business density.* The GDP per capita in 2009 (*Regional GDPpci* ) is included to control for the overall level of economic development.4 Moreover, the share of employees in the manufacturing sector (*Regional share of manufacture*) controls for the degree of industrialization of regions. Finally, a set of country dummies (*C*) and year dummies (*T*) are included to control, respectively, for all country-level unobserved characteristics and for time specifc events. As results, *Regional frm birth rate* is modelled using the following equation:


Equation (1) represents our main model. As we were also interested in exploring different patterns between Eastern and Western European countries due to the stage of industrial development, additional regressions have been run for the sub-sample of 48 regions belonging to East European countries (i.e., Bulgaria, Czech Republic, Estonia, Poland Hungary, Romania and Slovakia) and the sub-sample of 73 regions belonging to West European countries (i.e., Denmark, Finland, France, Italy, Portugal and Spain).

 All estimates are performed using OLS pooled regressions with clustered standard errors. To facilitate the comparison of the regression coeffcients, continuous independent variables were standardized, dividing them by two times the sample standard deviation, while the dichotomous independent variables are centred around their sample mean (i.e., demeaned) (Gelman, 2008). The adopted linear rescaling changes the coeffcient values of the independent variables, but it does not change the associated *t*-statistics and *p*-values.

### **Estimation results**

Table 10.1 shows the descriptive statistics and matrix of correlations of the unstandardized regression variables. Table 10.2 shows the results of the OLS pooled estimates for regional frm birth rate.

In Table 10.2, models in columns 1, 2 and 3 show the baseline specifcations with controls only for the country (Italy as reference category) and year fxed effects. Model 4 extends the basic models by adding all other control variables – i.e., *Regional GDP pc* and *Regional Share of manufacture*.

The coeffcient of *Regional frm average size* is negative and signifcant in Models 1, 2 and 3. This means that the presence of large-scale frms negatively affects entrepreneurship. A two-standard deviations increase in *Regional frm average size* (i.e., an increase of 16.94 employees per frm)


*Table 10.1* Descriptive statistics and correlation matrix


*Table 10.2* Determinants of regional frm birth rate: OLS pooled regressions


 Notes: Clustered standard errors in parentheses; \*\*\**p* < 0.01, \*\**p* < 0.05, \**p* < 0.1; Year dummies (included in the models) are not reported for the sake of clarity.

produces an average decrease of 1.55% in regional frm birth rate (see Model 3). These results support Hypothesis 1a. However, when we control for *Regional GDP pc* and *Regional share of manufacture* (see Model 4), the negative effect of *Regional frm average size* decreases and becomes not signifcant. This means that the relationship between *Regional frm average size* and entrepreneurial activity at regional level may be dependent on the level of industrial development of the region. Eastern and Western European regions differ in terms of stage of industrial development and the role of the manufacturing industry as an engine of growth, thus making the infuence of the variable *Regional frm average size* different between the two groups of regions. We turn back to this issue below.

In all models, we fnd a signifcant and positive effect of family frms (*Family business*). A two-standard deviation increase in *Regional family business density* (i.e., an increase of 0.38 in the share of family frms) produces an average increase of 1.1% in regional frm birth rate (see Model 3). These results support Hypothesis 1b and suggest that a high density of family frms is conducive to entrepreneurial attitudes and cultures. With regard to the other control variables, we fnd a negative and signifcant effect of *Regional Share of manufacture* on the entrepreneurial activity. This means that regions with an economic structure characterised by a strong presence of manufacturing industries show lower regional frm entry rate. Moreover, country dummies are almost always statistically signifcant: the frm entry rate are particularly high for Poland, Romania and Slovakia as compared to Italy, the reference category.

As argued in Hypothesis 1c, we can expect that *Regional frm average size*  reduces regional frm entry rate only in Western European regions, while an opposite effect is expected for Eastern Europe. Table 10.3 reports the results of OLS pooled regressions for the sub-sample of Western European regions (Model 5) and the sub-sample of Eastern European regions (Model 6). All regressions in Table 10.3 include the full set of control variables.

In line with Hypothesis 1c, the coeffcient for *Regional frm average size* is negative for Western European regions (Model 5) and positive for Eastern European regions (Model 6). Interestingly, the results show that the coeffcient for *Regional family business density* is positive for Western regions (Model 5), but it is not statistically signifcant for Eastern regions (Model 6). We believe that this evidence, i.e., the scarce or null infuence of family business density on entrepreneurship in Eastern European regions, is due to the signifcant dependence of local industrial sectors on non-local actors, i.e., multinational or other foreign frms. However, further research is needed to comprehensively explain this result.

Additional checks have been conducted to validate the robustness of our main results.5 First, we performed tobit regressions to take into account the fact that the dependent variable is left censored. The results are very similar to those discussed above. Secondly, we performed additional regressions using different subsample periods to account for heterogeneous macroeconomic conditions (crisis and recovery periods). Finally, we assessed the



Notes: Clustered standard errors in parentheses; \*\*\**p* < 0.01, \*\**p* < 0.05, \**p* < 0.1; Country dummies and year dummies (included in the models) are not reported for the sake of clarity.

weak causality of the cross-sectional model by using a rolling-reference window for the dependent variable, from the time interval 2011–2016 to the time interval 2014–2016 and 2015–2016. In all cases, estimated results confrm the main evidence of the paper.

# **Conclusion**

In this chapter we investigated the impact of industry scale and family frm density on new entrepreneurial initiatives at the regional level in Europe. By drawing information on 121 regions located in 13 European countries from the Eurostat regional database and the BvD-Amadeus database, we found that the regional industry scale (i.e., the average frm size at the industry level) and the regional density of family frms positively affect regional entrepreneurship (i.e., the regional frm birth rate). However, when the level of industrial development is taken into account – and European regions are classifed either as Eastern or Western – these results only hold for Western European regions, whereas in Eastern European regions the industry scale variable has a positive (instead of negative) infuence on entrepreneurship and the contribution of family frm density almost disappears.

The main contribution of this chapter is to explore behavioral differences between family and non-family frms in a regional context. Existing literature highlights the importance of family ownership in explaining frms' economic and innovative behaviors, but partially neglects the role played by family frms within the geographical context. Family frms can beneft from a strong social network and they are more embedded in the industrial setting of the region than non-family frms. This book chapter provides an

# 206 *Riccardo Cappelli et al.*

empirical evidence to shed light on the impact of family frms on entrepreneurship as a regional outcome.

Our analysis has a number of limitations that also presents questions for further research. First, the specifc mechanisms that drive the different results for Eastern and Western European regions remain a black box to a large extent. In addition to large multinational and foreign frms, these outcomes can be also related to national and regional institutions and policies. In general, the analysis could be enriched by incorporating additional institutional, economic and social variables in the framework and identifying their role more precisely. Second, we focus on entrepreneurship, but the analysis of the impact of industry scale and family frm density can be extended to other regional economic outcomes. Third, since we use OLS cross-sectional regressions, we could only show associations and not causality in the relationships. Therefore, it might be important to strengthen the analysis using more appropriate and robust estimation methods.

# **Acknowledgment**

We thank the editor Rodrigo Basco for his guidance in the preparation of the paper and his helpful comments on a previous version. Usual disclaimers apply.

# **Notes**


# **References**


# **11 Family frms and regional development**

Evidence from China

*Xinrui Zhang, Junsheng Dou and Hanqing "Chevy" Fang* 

### **Introduction**

In recent years family business literature has started recognizing the importance of context (Gomez-Mejia, Basco, Müller, & Gonzalez, 2020; Morck & Steier, 2005) to better understand family frm behavior. This is because context might affect the size and structure of the business family, resource endowment, and feasible governance mechanisms that business families use to protect their economic and non-economic endowments (Chen, Fang, MacKenzie, Carter, Chen, & Wu, 2018; De Massis, Ding, Kotlar, & Wu, 2016; Janjuha-Jivraj & Woods, 2002; La Porta, Lopez‐de‐Silanes, & Shleifer, 1999).

Since opening up to global trade and implementing free-market reforms in 1978, China's economy has experienced tremendous growth, with a 9.5% average annual growth rate in gross domestic product (GDP). Indeed, China has grown from a low-income economy with a GDP per capita of less than \$200 to a middle-income country with about \$8,800 GDP per capita. Market reforms have also nurtured a large population of entrepreneurs, who have become the driving force behind the increasingly important private economic sector in China (Tsang, 1996). As China's economic system continues to mature, entrepreneurs who were able to capture the opportunities of the reform in the 1980s have reached an age where they need to consider the possibility of passing their business to next-generation family members (Pistrui, Huang, Oksoy, Jing, & Welsch, 2001).

Nonetheless, most research in family business is contextless and scholars in this feld have recognized that more studies are needed to better explore the interaction between family frms and context (James, Hadjielias, Guerrero, Discua Cruz & Basco, 2020). At the same time, despite the strong interdependence among family businesses, business families, and the Chinese context, we still lack a thorough understanding of how the Chinese context might contribute to unique specifcities of Chinese family frms. It is an emerging economy but at the same time is also a transitional one. Hence, federal and regional governments play an important role in designing policies that eventually impact how family frms behave and perform.

Our evidence shows that, just like in other economies, family frms are a relatively common phenomenon in China across regions. Additionally, we observe that even though there is a prevalence of family frms, its presence varies across geographical regions. Finally, our fndings show that, comparable to other national and regional economies, Chinese regional development seems to have substantial effects on family frm behaviors in terms of R&D investment, industrial diversifcation, and internationalization.

# **The Chinese context, institutional transition, and family entrepreneurs**

Just like in other economies, family frms are highly prevalent in China, if not more so (Sharma & Chua, 2013). Different from other cultural environments, business leadership in China is inextricably tied to the central role that family plays in business (Liden, 2012), largely stemming from the Confucian norms and values deeply embedded in Chinese society. In China, the idea of "family" (Chinese: 家) refers to a unit of members related to each other by blood, marriage, or adoption, characterized by culturally defned role differentiations and solidarity among members, the owning of common property, the sharing of common production and common consumption, as well as the practice of common social activities. Here, "family" would go beyond simply a biological unit bonded by marriage and blood. In fact, Chinese family can be conceptualized as a social unit in which family members' roles and solidarity relations are defned by underlying Confucian values; it is an economic unit in which living members produce and consume in common and also a cultural unit where the family is responsible for performing certain anthropological rituals such as ancestor worship for the wellbeing of both living and deceased members.

Note that, Confucian norm specify a unique relationship between the family and the individual. In the West, the family often serves the basic function of raising and preparing an individual before he/she goes out into the world and becomes a full member of society. As a result, families often experience a "break up" when children reach adulthood, depart from the original family, and start to build their own nuclear families. In other words, the family system in Western society exists to nurture junior family members. The situation in the traditional Chinese context is the exact opposite: the purpose of family members, especially junior males, is to nurture and continue the family, which is often aligned with a shared family surname, history, and antecedents. As Baker puts it,

There is an underlying assumption in Chinese thinking on the family that there is such a thing as a "Continuum of Descent"…Decedent is a unity, a rope which began somewhere back in the remote past, and which stretches on to the infnite future . . . the individual alive is the personifcation of all his forebears and all his descendants yet unborn. (1979, 26)

Hence, it is not a surprise to see that Baker concluded that the "individual was dominated by the family (in China)" and "the actions of individuals were geared to the requirements of the (Chinese) family" (1979, 27).

### 212 *Xinrui Zhang et al.*

Given its prevalence and strong infuence, the family-centered culture in China often makes family business the default form when entrepreneurs start their ventures (Aldrich & Cliff, 2003). Also, family members are socially and culturally obligated to support each other's business initiatives and family-endowed fnancial, social, and human capital are often used in the family system (Arregle, Hitt, Sirmon, & Very, 2007). In addition, business leaders in China are often considered "father fgures", who are expected to have proven capability and integrity, yet be considerate, and are expected to take good care of even non-family managers and employees. Taken together, it appears reasonable to conclude that the prosperity and idiosyncrasies of Chinese family frms partially stem from traditional Confucian culture, which emphasizes the inseparable connection between the individual and the family system.

Besides informal institutions such as culture and value systems, formal institutions also play an important role. Indeed, China is still a transitional economy undergoing changes in its economic and political systems. The most relevant part for our discussion is the adolescence stage of the private sector, where economic activities led by entrepreneurs started during 1980s and 1990s (Pistrui, Huang, Oksoy, Jing, & Welsch, 2001). Prior to this, China was a planned economy largely controlled by stateowned frms, with very few opportunities left to private enterprises. It is family entrepreneurs/founders who started the frst wave of privately owned businesses immediately after the 1978 Open-Up and Reform era. In addition, given the fact that formal institutions such as the protection of property rights are still not fully developed in China, family governance is often used to "fll the voids". Here, the family's reputation often serves as a reliable substitution for effective commercial laws and the family network might help reduce the cost of searching for business partners and maintaining business relationships (i.e., transaction costs).

Finally, because of the emerging nature of Chinese economy, all family entrepreneurs and family frms in China are relatively young. In fact, in Western economies some family frms have already been successfully passed from the founding members to the second or later generations (Chua, Chrisman, & Sharma, 2003; Morck & Steier, 2005). In comparison, the majority of Chinese family frms have never experienced intra-family, trans-generational succession (Janjuha-Jivraj & Woods, 2002). Furthermore, entrepreneurs and business families in China still lack experience with family governance as well as talent in the family labor pool (Morris, Allen, Kuratko, & Brannon, 2010). Finally, the global economy has entered a period of recovery and China's economy, which had been on the rise so far, is facing a slowdown. In contrast to mature family frms in Western countries, which have already experienced various economic cycles (La Porta, Lopez‐de‐Silanes, & Shleifer, 1999), family frms in China are still struggling to leverage technological innovation and strategic renewal in order to maintain their positions in the market (De Massis, Frattini, & Lichtenthaler, 2012; De Massis, Ding, Kotlar & Wu, 2016).

To summarize, meaningful insights of family business in China cannot be decoupled from (1) cultural heritage stemming from Confucian culture and values; (2) institutional transitions that are strongly connected with economic growth in China; and (3) family founders/entrepreneurs who started the business. In the following sections, we will discuss the specifc characteristics of family entrepreneurs and family business, family governance, and strategic and environmental issues in family business. In our discussions, we will elaborate on the unique "Chinese characteristics" of family business in the country, which either result from or contribute to the nation's economic growth and institutional transitions.

### **History and overview of Chinese family business**

#### *Data sources*

All the following discussions are based upon information from two data sources. The frst is the China Stock Market & Accounting Research Databases (CSMAR), which covers all publicly listed frms in China. In line with existing research (Chua, Chrisman, & Sharma, 1999; La Porta et al., 1999), we classify family business as those with (1) at least 15% family ownership and (2) at least two family members who currently serve as top managers and directors. The second data source is the All-China Federation of Industry and Commerce (ACFIC) database, which is based on a survey effort of private enterprises in China from 1995 to 2016, with data collected once every two years. We also include a data collection on the "health and sustainability of Chinese family frms" jointly conducted by Zhejiang University and the All-China Federation of Industry and Commerce in 2015 and 2016. The nature of public and private family frms vary due to frm size and goals that the dominant coalition pursues. Therefore, we used data from public and private family frms and we separated the analysis because both types may differ in terms of behavior and performance.

Furthermore, according to the China market index, we classify provinces and areas in China into three categories: highly-developed, moderatelydeveloped, and under-developed. In general, most provinces along the eastern and southern coast fall into the highly developed category. Provinces in the middle and mid-western parts of China are mostly moderately developed, whereas those in the west or south-western parts are under-developed. This observation is consistent with the idea that the coastal areas are more likely to be exposed to economic opportunities, hence more likely to experience higher economic development.

In the following sections, we draw attention to regional differences with regard to family frm behavior, structure, and performance. Such an effort refects our research intention of explaining the spatial distribution of Chinese family frms as well as their differences and similarities across regions. Additionally, when the data information allows, we will explore the regional infuence on Chinese family business.

# *Private enterprises and family business in China: an overview*

According to the National Bureau of Statistics of China, in 1996 there were 443,000 private enterprises. In 2017, the number reached 14.4 million with an increase of 31.4 times and an average annual growth rate of 18.0%. The proportion of private enterprises increased from 16.9% to 79.4%. In addition, as of 2017 private frms have contributed to more than 60% of China's GDP, 50% of national tax, 70% of investment in technological innovation and new product development, and 80% of newly created jobs. According to China's National Bureau of Statistics, the number of private enterprises in China has dramatically accelerated over the past few years (Figure 11.1). Among them, the number of private enterprises in highly-developed areas have been much higher than that in moderately-developed and under-developed areas. As shown in Figure 11.2, it appears that the level of regional economic

*Figure 11.1* The number of private firm registration in China by regions (unit: million, 2010–2017).

 *Figure 11.2* Publicly listed family frms in China and regional development (unit: million, 2005–2018).

*Figure 11.3* Average family ownership of Chinese listed frms.

development is positively associated with the number of private sectors/enterprises, especially family-owned ones.

In the following part we briefy discuss the general features of family involvement in ownership across publicicly-traded companies in China. According to the CSMAR database, average family ownership has increased over the past 14 years. Family ownership is highest in highly-developed areas, second in moderately-developed areas, and lowest in under-developed areas. However, the gap between moderately developed and underdeveloped areas has been narrowing in recent years, maybe because of the overall improvement of the institutional context across regions in China (Figure 11.3).

#### *Family entrepreneurs in China*

As mentioned above, any discussion on family frms in China cannot neglect the important role of family business leaders, who, in most cases, are the founders of the business. Yet, in contrast to their descendants, family entrepreneurs/founders might lack the formal knowledge, skills, and/or education to run the business, partially due to the under-developed educational system of China before the Opening-Up and Reform era. Nonetheless, in alignment with free-market reforms, education has become increasingly important and family entrepreneurs have been active in obtaining education in order to develop/maintain their competitive edges in the market. Indeed, ACFIC data from 1995 to 2016 shows that the education levels of family entrepreneurs have signifcantly improved over time (Figure 11.4).

Additionally, consistent with existing studies on family leadership in business (Chen et al., 2018), female family leaders have been on the rise. According to ACFIC data from 1995 to 2016, the proportion of female family leaders has increased from 10.40% in 1995 to 20.40% in 2016 (Figure 11.5). Such a change has also been observed in Western economies, which might refect a global trend where the function of female family members has

*Figure 11.5* Proportion of female entrepreneurs in China.

transitioned from a supportive role in the family system into a leading role in the intersection between the family and business systems (Curimbaba, 2002; Nelton, 1998).

In the absence of a mature market system, family entrepreneurs in China often choose to develop political connections and/or join political alliances to better exploit opportunities and absorb uncertainties stemming from the context (Morck & Yeung, 2004). In fact, business families' active political participation (the communist party) might facilitate access to valuable opportunities and resources, help overcome the liability of "newness" and lack of legitimacy in entrepreneurial activities, and potentially expand formal and informal networks (Tsang, 1996; N. Xu, X. Xu, & Yuan, 2013). Indeed, according to the ACFIC survey, the proportion of family entrepreneurs with membership in the Communist Party of China (CPC) increased dramatically from the 1990s to early 2000 (Figure 11.6). Nonetheless, we

*Figure 11.6* Political background of entrepreneurs. Note: CPC is short for Communist Party of China.

also observe strong fuctuations and the proportion has started to decline in recent years.

Due to data limitations we were unable to specify the variation of family private entrepreneurs across regions. Future researchers might want to collect additional data to explore these differences and causes, and how they might vary across regions.

### **Current status**

This section intends to discuss critical issues related to current challenges in Chinese family business. The topics cover governance mechanisms, innovation, diversifcation, internationalization, and frm performance in Chinese family frms across regions.

### *Corporate governance*

Just like their Western counterparts, Chinese business families are inclined to use governance mechanisms in enhancing their control over the business. Certain mechanisms such as a board of directors and chair positions are often used. These not only help the owning family supervise and control operational activities, but also mitigate the tension of conficts of interest among shareholders and stakeholders and may help maximize the value of the frm in the longrun (Carney, 1998, 2005; Chrisman, Chua, Kellermanns & Chang, 2007).

Starting from 1995, the percentage of family frms establishing boards of directors, boards of supervisors, and shareholder meetings has been increasing (Figure 11.7). According to the ACFIC survey, from 1995 to 2016 this trend can be best described as a "rise and fall". In 1995 only 29% of respondents had a board of directors. This number jumped to 74% in 2004, and then declined into 43% in 2016. Similar trends have also been observed in the establishment of board of supervisors and shareholder meetings.

*Figure 11.7* The governance institutions of Chinese private frms.

 *Figure 11.8* Family members in top management teams, supervisor boards and director boards by regions.

Nonetheless, executive, supervisory, and director positions are often reserved for family business founders or their relatives (Figure 11.8). In fact, the proportion of family members in the executive team (10–20%) and the board of directors (15–25%) are relatively high.

We notice substantial regional differences. In the more developed areas, we tend to have higher proportion of family members in the executive team and boards. This phenomenon may be due to the positive correlation between family ownership and regional development (Figure 11.3), as higher family ownership might motivate the family to assign more members to leadership positions. This might also stem from the improved education and capabilities of family members in highly-developed regions, hence making them more capable of management. Future researchers might want to collect additional data to explore these possibilities.

### *Innovation*

Innovation is the driving force behind business growth and superior performance. Research and development (R&D) intensity, measured as the R&D expenditures/Sales percentage, is often used to capture a frm's resource investment in technology innovation. According to the European Union standard, frms with more than 5% R&D intensity are considered to have a high level of R&D investment, with 2–5% a medium-level, and those below 2% a low-level.

Using the CSMAR database, we track R&D investment in 307 listed Chinese family frms from 2009 to 2015. In general, half of them have an R&D intensity between 2% and 5%; 15.72% have an R&D investment intensity higher than 5%; and 21.21% below 1%. In addition, frms whose founders are aged 41 to 50 have the highest level of R&D in business (4.01%). Also, family founders who have at least a bachelor's degree tend to invest more in their businesses compared to those with lower levels of education. In addition, family frms with both family owners and family managers have slightly higher R&D intensity (3.69%) compared to those with family involvement only in ownership (2.82%). Furthermore, those with second or later family generation members in top management positions have higher R&D investment compared to those without.

Note that the abovementioned fndings are somehow different from the Western experience. As Fang et al. (2018) and Memili, Fang, and Welsh (2015) show, compared to the founding generation, second- or later-generation members in Western family frms are often more risk averse, refected in lower levels of R&D investment and internationalization. Our fndings might imply that the transitional nature of the institutional and economic context in China, coupled with better overseas education among junior family members, might motivate late generation-controlled family frms to be more innovative and risk-taking compared to the founding generation-controlled family frms.

It appears that there is a positive correlation between a family frm's R&D investment and regional development as those in developed regions tend to have higher R&D investments (Figure 11.9). Such a phenomenon might be due to the prevalence of institutional voids in under-developed regions. Hence, family frms in these areas might be more motivated to invest resources in political connections rather than R&D and technology innovation (Morck & Yeung, 2004). It is also possible that under-developed areas

*Figure 11.9* R&D investment intensity of listed family frms in China by regions.

*Figure 11.10* Average patens and invention patents of listed family frms in China by regions.

are dominated by low-technology industries and that the demand for R&D is not urgent. In addition, such a fnding might be due to the fewer colleges and universities, as well as the lack of supportive government policies, in under-developed regions. While we are unable to test these propositions given the data limitation, future studies may certainly further explore the specifc mechanisms behind these differences.

It has been found that the number of patents in highly-developed areas is the highest, while those in moderately developed areas are the lowest (Figure 11.10). Given the fact that a patent is an outcome measure of innovation, this result might suggest that family frms vary in terms of innovation productivity across regions. Future researchers might want to pay more attention to this possibility.

### *Diversifcation*

Diversifcation happens when the company expands its product lines and/ or enters a new geographic/industrial market. According to the CSMAR database, for 305 listed Chinese family frms from 2009 to 2015, the overall diversifcation index was 0.237, a moderate level of diversifcation. Its calculation takes into account the number of different products, the distribution of products in the production line, and the dissimilarity or heterogeneity of products, applying equally well to plants/frms/industries and bounded between zero and unity (Gollop & Monahan, 1991; Jacquemin & Berry, 1979). Data from the United States in 1982 shows that the diversifcation index is between 0.049 and 0.482 (Gollop & Monahan, 1991).

The proportion of non-related diversifcation is over 78%, meaning that Chinese family frms tend to focus on market areas that are less related to their "core" business. Unrelated diversifcation can help neutralize risks from same business areas, but may also increase the cost of resource allocation and operational coordination (Jones & Hill, 1988). It is our belief that the reason behind this unrelated diversifcation choice is the rapid growth of China's economy, which may create many new yet unrelated opportunities that business families intend to exploit. Additionally, unrelated diversifcation may be infuenced by political connections (Faccio, 2006). A series of studies fnd that political connections might result in better frm performances in emerging economies (Li & Zhang, 2007; Peng & Luo, 2000). This is because when market and political institutions are immature, political connections can help companies better seek help from the government and obtain valuable resources such as fnancial support, novel technology, and monopoly permits (Peng, Lee, & Wang, 2005). Furthermore, political connections are not bound to specifc industries, hence this encourages companies to enter new industries to avoid idle resources (Li, He, Lan, & Yiu, 2011).

There are also notable regional differences. Family frms in underdeveloped and highly-developed areas tend to have higher diversifcation compared to those in moderately developed regions (see Figure 11.11). One possible explanation is that family frms in highly-developed areas choose to diversify due to acquisition and leveraging of superior knowledge stemming from regional development. Conversely, family frms in underdeveloped areas diversify into certain industries that are not developed yet in order to exploit the frst-mover advantage. In other words, in highlydeveloped regions, family frm diversifcation is driven by the "supply" side (capable of diversifying), while in under-developed regions it is driven by the "demand" side (the industry is emerging or has not been exploited yet).

In addition, different types of family involvement may affect the degree of business diversifcation. In fact, family involvement in both ownership and management is coupled with lower level of diversifcation compared to cases where families only serve as the owner (Figure 11.12).

*Figure 11.11* The degree of diversifcation of family businesses in different regions.

*Figure 11.12* Types of family participation in business and enterprise diversifcation.

 *Figure 11.13* Total overseas business revenue of listed family enterprises in China (billion yuan).

Due to data limitations, we were unable to specify the particular causes of family frm diversifcation across regions. Future studies might explore the relationship between types of family participation in business, business diversifcation, and regional contexts.

### *Internationalization*

Internationalization refers to a strategic choice in which companies expand the business territory into foreign countries. According to the CASMR database, from 2009 to 2015 there were, in total, 2,053 frm-year observations for publicly listed family frms in China. Among them, about 74.9% (1,538) had revenue from internationalization, resulting in a total of 768.794 billion yuan (about \$109.83 billion).

As shown in Figure 11.13, the incomes of Chinese family businesses from foreign operations has been increasing over time. The total overseas revenue in 2015 was about four times higher compared to that in 2009. In terms of regional differences (Figure 11.14), it appears that family frms in highly developed regions tend to have higher incomes (547 million yuan/year) from internationalization compared to those in moderately-developed (310 million

 *Figure 11.14* Overseas business revenue of family frms in different regions of China (billion yuan).

*Figure 11.15* Annual sales of listed family frms by regions (billion yuan).

yuan/year) and under-developed (225 million yuan/year) areas. In China, highly-developed areas are located in coastal areas where geographical and transportation advantages may create more international opportunities. In addition, these regions might equip family members with better education and knowledge, which might help family frms beneft more from globalization. Finally, there might be more supportive government policies in highly-developed regions compared to others, stimulating more international activities in family frms. Indeed, future researchers might want to collect more data to explore the specifc mechanisms behind our fndings.

### *Firm performance*

In the following analysis, we focus on two performance measures: sales and net proft. Figures 11.15 and 11.16 report change in sales and change in net proft in publicly traded family frms from 1999 to 2015. In general, family frms in China have experienced a state of dramatic growth in frm performance especially in highly-developed areas, which might be driven by

*Figure 11.16* Annual net proft of listed family frms by regions (billion yuan).

the improvement in institutional contexts and the embracing of free-market reforms in those regions. Such a result may also be driven by the increasing supply of high-quality human capital, as highly developed areas might attract more and better non-family professionals. Finally, as we discussed above, emerging economic opportunities in highly-developed areas might contribute to performance differences among family frms. Once again, future studies are encouraged to explore the specifc reasons behind these fndings.

# **Planning for the future**

In this section, we will discuss issues related to the future of family business in China based on regions, including intra-family, trans-generational succession and social responsibility in family business.

# *Succession*

The family founder's willingness to pass the business to later-generation family members constitutes a major driving factor behind the succession process. Nonetheless, according to ACFIC data 1997–2010, only about 32% of family founders are willing to initiate the succession process (Figure 11.17). In fact, in 1997 there was a higher proportion of family founders (42.8%) who were willing to pass control to their children. This number declined to 24.8% in 2006 and then increased to 34.1% in 2010. The fuctuation might be due to the 2008 fnancial and economic crisis, which might have suppressed the intention of family succession. In fact, family decision-makers, especially the founder, might choose more conservative ways to pass on the family's wealth, such as through family trusts, fnancial investments, etc.

*Figure 11.17* Succession intention of entrepreneurs.

*Figure 11.18* Succession intention of the next generation.

The children's willingness to take over the business is also important. According to the survey data in 2002 and 2010 (Figure 11.18), very few late-generation family members were willing to become the successors of the business (15.2% in 2002; 16.8% in 2010). This might be because they plan to start their own businesses or because the owning family prefers a more traditional way (i.e., fnancial investment) to protect the family's wealth.

 Even so, according to CSMAR, a large number of second- or latergeneration family members has chosen to work in family frms as chairs, CEOs, top executives, and/or directors. There are certain regional differences (Figure 11.19). The second- or later-generation involvement is highest in moderately-developed regions, followed by highly-developed and underdeveloped regions. As the speed of economic reform varies by regions, the economic opening of less-developed regions was relatively late, which might explain why we have lower later-generation family involvement in less-developed regions. Future researchers can further explore succession issues in Chinese contexts as well as cross-cultural comparisons. For example, what are the unique factors that motivate family founders to initiate the succession process? What factors might enhance the second generation's

*Figure 11.20* CSR in Chinese family frms.

willingness to inherit the business? Affected by frequent changes in political systems and economic policies, entrepreneurs of different ages may have different cognitive models and, hence, we might need more "microfoundation" studies. Indeed, more studies are needed to explore crossindividual, cross-border, and even cross-cultural possibilities.

# *Corporate social responsibility*

Family frms in China have gradually increased their emphasis on corporate social responsibility (CSR). According to the ACFIC survey (Figure 11.20), 56.2% of family frms consider CSR very important, 24.78% important, and 12.56% relatively important.

According to the China Charity Donation Report, in 2016 there were in total 139.294 billion yuan made in donations, with private family frms

 *Figure 11.21* Trend of charitable donations made by Chinese private frms in different regions (thousand yuan).

contributing nearly 50%. Figure 11.21 reports the temporal trend in different regions. In general, the annual amount of charity donation has been increasing and the amount in 2015 was almost fve times higher than in 1996.

One important part of CSR is the company's intentional effort to reduce damage to the environment and/or propose additional policies to enhance the protection of it. In fact, a large share of the environmental pollution in China is caused by commercial organizations. As the ACFIC survey shows, family frms with 1) better educated family founders, 2) more effective governance mechanisms, and 3) more international operations tend to have heavier investments in environmental protection.

It appears that family frms in under-developed areas have higher environment-related costs (Figure 11.22). This might be because family frms in these areas often pollute more due to weak law enforcement, and higher costs on environmental protection simply refect the money paid for violations of environment-related laws and regulations. This may refect the difference in the embeddedness of family frms between regions. Due to data constraints, we cannot explore the relationship between family embeddedness and CSR. Future researchers can collect data in this area for further research. Indeed, whether family frms have more or less CSR has been under debate for a long time (e.g., Campopiano, De Massis, & Chirico, 2014; Ge & Micelotta, 2019). Some studies have concluded that family frms might actively engage in CSR activities in order to maintain their family reputation (Campopiano, De Massis, & Chirico, 2014), whereas some claim that family owners are more "self-interested" and might refuse to invest (Cruz, Larraza-Kintana, Garcés-Galdeano, & Berrone, 2014). Under the condition of imperfect market competition, CSR of Chinese family frms may stem

 *Figure 11.22* Environmental protection costs of private frms in different regions (thousand yuan).

from government pressure. More studies are needed to explore the specifc causes of family business CSR in the Chinese context and how these might differ from their Western counterparts.

# **Discussions**

### *Theoretical implications*

### *Comparing to the Western context*

When comparing Chinese family frms with their Western counterparts, we notice many similarities but also some substantial differences. Indeed, business families in China often prefer to have stronger control in ownership, governance, and management than their Western counterparts. This implies that the owning family's power in business is one of the fundamental features in family frms all over the world, in both developed and emerging economies.

In the particular context of China, succession continues to be a critical issue, as second-generation family members often refuse to take a leadership role in business. This lack of motivation might stem from various reasons. For instance, the rise of business opportunities after the 1978 Open-Up and Reform might motivate later generations to start their own businesses rather than succeeding their parents in family frms. In addition, strong dynamics in the institutional and cultural systems might affict traditional Confucian norms, hence "generational confict" might be further pronounced in contemporary China, as senior generations often assume traditional values whereas younger ones might place more weight on realizing their individual merits and might refuse to fulfll their family obligations. Of course, the lack of motivation among second-generation junior family members may also contribute to the rise of professional non-family management in Chinese family businesses.

### *Spatial difference and family business in China*

Another issue worth noting is that family frms in China show a high level of heterogeneity, which is largely captured by regional differences. In fact, China is a country with the third-largest national territory in the world. Thus, it represents a unique context to explore the interaction between business families and external factors that result in heterogeneity in family frm behaviors and performances.

As mentioned in the section "The Chinese context, institutional transition, and family entrepreneurs" and throughout the chapter, the similarities and differences can be best explained by the economic, political, and cultural context in China. On one hand, given the long history of the country, power structures and intra-family relationships in business families often refect the patriarchal traditions rooted in Confucian culture (Fei, Chang, & Ward, 1946). On the other hand, the free-market system in China remains relatively young and the political and institutional systems are still evolving. Such an issue is refected in the "unbalanced" distribution across regions, as highly-developed areas might have better institutional context, more economic opportunities, and higher-quality human capital and knowledge resources that family frms can leverage in order to pursue superior frm performance.

As an example, it is found that Chinese family frms located in lessdeveloped regions are less motivated to take risks in strategic actions. It is our suspicion that such regional differences might be related to unbalanced development in institutional and political systems. Indeed, family frms in these regions might be more motivated to develop political connections and make profts from "rent seeking" (Morck & Yeung, 2004) rather than from the Schumpeterian type of innovation and entrepreneurial activities. It is also found that family frms in highly-developed regions tend to have more revenue from international trade and better frm performance overall, which might stem from the higher prevalence and quality of economic opportunities in these areas. Finally, we notice that family frms in highly-developed regions have stronger control of the business, with more later-generation family members involved as chairs, CEOs, top managers, and directors.

From the summary above, we can conclude that there might be two different types of family frms in China and in other developing and transitional economies according to frm's location: family frms in developed areas and family frms in under-developed areas. This fnding seems to suggest that even in one (developing/transitional) economy there still exists a high level of heterogeneity among family frms, largely captured by economic and institutional developments across regions. We encourage future researchers to further explore regional differences based on the characteristics and

# 230 *Xinrui Zhang et al.*

behaviors of family business, as well as factors that might contribute to these differences. Future researchers can further compare this "regional differences" in China with other economies, hence we might be able to develop better understandings of how country-level factors such as culture, economic development, and institutional stability might contribute to regional differences and family frm heterogeneity in each country.

# *Family business and context*

As entrepreneurship literature and the theory of endogenous development remind us (e.g., Audretsch & Keilbach, 2004; Feldman, 2001), entrepreneurs and business persons are adaptive agents who often actively adjust their strategies to best capture opportunities and neutralize threats stemming from the context, refected in policies, fnancing, labor, and other aspects (Acs & Varga, 2005; Black, Meza, & Jeffreys, 1996; Johnson & Parker, 1996). While arguably both family and non-family frms are affected by context, the uniqueness of family businesses lies in its deeper "embeddedness" in the local community through family-centered and/or family-member-based business connections and social relations (Basco, 2015). Such a high embeddedness might imply that, compared to their non-family counterparts, family frms might be more "altered" to valuable information from the context, hence are more "adaptive" as they are able to better and more quickly identify upcoming opportunities and threats in the context. Another important insight is that, given an upcoming opportunity or threat, the strong family-centered control in business can ensure that business families are capable of responding more quickly without the interference from other decision-makers. That means that not only can family frms better identify unusual signals in the context, but that they are also capable of responding even more quickly than non-family frms.

# *Practical implications*

This chapter has some practical implications for designing government policies and nurturing next-generation family leaders. To begin, economic development not only depends on the stock of capital, also on who owns and uses the capital (Morck & Yeung, 1998). As mentioned above, family businesses and business families have been playing an important role in China's economic development. Indeed, given the strong focus on "family" in the traditional Confucian culture, as well as the prevalence of family frms in all regions, federal and local governments in China should maintain a high degree of tolerance for family businesses and create an environment that is conducive to business families.

Given substantial regional differences, local governments should design specifc family business-related policies that ft local context. For underdeveloped regions, it might be important for local governments to increase transparency and reduce the costs of accessing and obtaining external resources. In another word, local governments should fnd alternative measures to "substitute" political connections and establish a more mature market system, so that the competitiveness of local family frms can be further improved. Under such circumstances, local governments should open more channels to communicate with family frms, and provide them with more support in terms of expertise, infrastructure, and debt fnancing. Similar measures have been implemented in large and more-developed cities such as Beijing, Shanghai, Guangzhou, Shenzhen, and others.

It is important to note that, according to the ACFIC survey, most second-generation family members refuse to take the leadership role in family businesses partially due to the explosion of entrepreneurial opportunities and their ambition to start their own businesses. Given such a context, it is important for the government to develop specifc policies, such as those highlighting family tradition and the legacy of family founders, in order to prepare future generations for assuming the leadership transfer.

Indeed, second- or later-generation family members grew up during a period of great social change in China and might possess a completely different mentality compared to their parents. Furthermore, the educational background of the leadership seems to be important even in the family business context. Hence, local governments need to design policies to motivate second-generation family members to be more actively involved in their family frms and design educational systems that nurture these junior family members such that their knowledge and skill structures can match severe market competition.

# *Limitations*

There are limitations to the analysis conducted in this chapter, which might shed light on future studies of Chinese family frms. First, we report our fndings based on two databases: the China Stock Market & Accounting Research Databases (CSMAR) public database and the All-China Federation of Industry and Commerce (ACFIC) database. While these two cover both publicly-traded and private family frms, future researchers might want to use alternate databases, especially those with broader coverage of family systems and members to better explore family business and regional development in China.

Second, although we intend to compare Chinese family frms with their Western counterparts, the two databases we used do not contain any information related to family frms in other economies. Future researchers might want to develop some comparative studies with a focus on how the respective economic and institutional contexts of China and Western countries might contribute to the similarities and differences of family frms in these regions.

### 232 *Xinrui Zhang et al.*

Finally, it is important to note that our discussion is bounded by a lack of historical longitudinal data. In fact, CSMAR started to include family business data just a few years ago, ACFIC's earliest survey data is from 1995, and even data related to private frms offered by the National Bureau of Statistics of China only started from 1999. Future researchers might want to collect richer historical data, especially those related to family businesses before and after the Opening-Up and Reform era that started in 1978.

### **References**


### 234 *Xinrui Zhang et al.*


**Part V A policymaker perspective** 

# **12 Family frms and corporate spatial responsibilities in Germany**

Implication for urban and regional planning and management

# *Hans-Hermann Albers and Lech Suwala*

### **Introduction**

The planning and management of urban and rural areas has always consisted of a mixture of state‐market relations and civil society. This trend has accelerated over the preceding decades as liberalization, deregulation and privatization of former state-performed tasks has increased, the institutional capacities of smaller communities have declined, and as regionally consolidated public administrations have become less effective (leading in extreme cases to failed states). During this time, much responsibility has shifted to non-state actors in general to participate in 'place-based' economic engagement; private sector and family frms in particular have been encouraged to become more involved in regional development and other governance schemes (Harrison, 2014; Suwala, Kulke, & Gade, 2018; Basco, Stough, & Suwala, 2020). This chapter attempts to analyze this phenomenon by offering a unique perspective on urban and regional engagement as it is driven by family frms. It does so by connecting two evolving discourses: On the one hand, on spatial distributions and characteristics of family frms (Basco, 2015; Stough et al., 2015; Basco & Suwala, 2020a, b), and on the other hand, on corporate spatial responsibilities (CSpR) – a spatial expansion of the well-known corporate social responsibility phenomenon (Albers, 2011; Knieling, Othengrafen, & Preising, 2012; Albers & Suwala, 2018; Albers & Suwala, 2020a; Suwala & Albers, 2020).

This chapter summarizes selected existing case studies of such family frm-driven urban and regional engagement from literature and asks how (forms) and to what extent (their spatial impact/intensity) family frms implement CSpR initiatives. In other words, the chapter seeks to establish whether cases from the one discourse can be translated into the conceptual schemes of the other; can family frms be viewed as "spatially responsible"? To operationalize this analysis, we apply the so-called 'CSpR maturity model' (Albers & Suwala, 2018) to family frms involved in spatial development in their areas. The cases in this chapter are pulled from data in existing literature, from the authors' previous research and from a personal selection of case studies of spatially engaged family frms (e.g., Albers, 2011; Albers & Suwala, 2018). We intend to use "purposive sampling" (Patton, 2001) and focus on "polar types" (Meredith, 1998) in order to showcase the diversity and wide variety of applicable cases.

Our results show that family frms engage with their environments in many ways and with varying degrees of spatial commitment. Family frms sponsor the development of social, educational and recreational infrastructure and facilities. They participate in district renewal, offer privately owned resources for public use, contribute to business improvement districts, are involved in town center management and lead master plan initiatives, among other activities. At the same time, they participate in spatial responsibility to different degrees of intensity. Some initiatives, similar to traditional corporate social responsibility measures, are limited in their spatial impacts (for example, donations, sponsorships, patronage and charity involvement); these are still the most common forms of outreach for family frms. Others engage in more complex measures, including joining public-private partnerships, cooperating on projects with public and civil sector actors and investing in long-term interinstitutional projects. These latter cases demonstrate the potential for targeted spatial engagement by family frms. In some cases, family frms even take state functions and exert place leadership, as in cases of frm-driven master plan initiatives.

This chapter sheds light on the importance of family frms as active stakeholders in urban/regional planning and management, a topic widely neglected in related academic literature. In this vein, policy-makers should not only consider family frms as an equivalent member in the enlarged orchestra of stakeholders involved in regional planning and management, but also need to understand and anticipate their interest in participating in regional policies and the advantages that can derive from these initiatives. Family frms are already widely 'entangled' in their regions, with personal and professional ties and commitments to their surroundings. Our research proposes strategies for unleashing the signifcant and often latent potential of family frms, in particular by recognizing shared interests and developing common objectives between them and the regions in which they are situated.

The chapter is organized as follows: Section 2 summarizes the 'state of the feld' with regard to family frms in spatial contexts and CSpR, and points to synergies while linking both disciplines; Section 3 encompasses the main analysis by showcasing different CSpR types and intensities of family frm-driven engagement and their implications for urban and regional planning and management; and Section 4 presents our conclusions and underlines our contributions to contemporary academic discussion.

# **Review of concepts – spatially embedded family frms and corporate spatial responsibilities**

# *Family frms in spatial context*

Although every economic entity is somehow situated or embedded in contexts (e.g., organizational, social, institutional) (Granovetter, 1985), spatial contexts have received very little academic attention in the realm of family frms, with some notable exceptions in recent years (Basco, 2015; Dana & Ramadani, 2015; Seaman, 2015; Stough et al., 2015). The contemporary rediscovery of space in family business studies has led to emerging concepts of 'regional familiness' (Basco, 2015) and later, 'spatial familiness' (Basco & Suwala, 2020a, b). According to Basco (2015, 267), regional familiness

incorporates the regional level into the concept of familiness by defning it as the embeddedness of family businesses in social, economic, and productive structures within a spatial context and the type of connections that emerge and interact with regional factors (…) and regional processes (…) through regional proximity dimensions.

Whereas 'regional familiness' favors a particular aggregational level (derived from the original concept of Habbershon & Williams, 1999) and scale (obviously, the regional), 'spatial familiness' offers a much broader understanding of the phenomenon by considering the recursive relationship between family frms and spaces at different levels (e.g., individual, frm, frm groups). Spatial familiness also recognizes the simultaneous overlapping of multiple scales (e.g. local, global), as well as a broader scope of defnitions of space that include locations, places and landscapes, among others (Suwala, 2019; Basco & Suwala, 2020a, b).

Apart from that, different topics have been investigated in recent years, linking family frms to certain spatial entities like locational factors (e.g., Kahn & Henderson, 1992); internationalization (e.g., Gallo & Pont, 1996); specifc spatial contexts like emerging markets (Basco, 2018) or transition economies (e.g., Duh, Tominc, & Rebernik, 2009); different scales such as local (Baù, Block, Cruz, & Naldi, 2017), regional (Bird & Wennberg, 2014) and global (Yeung & Soh, 2000) or various understandings of space (Suwala & Schlunze, 2019). Therefore, it is not surprising that family frms are often tied to a specifc spatial entity as the home region (e.g., Pongelli, Calabrò, & Basco, 2019). Moreover, very few studies have considered family frms and their importance for spatial planning and/or policies (Basco & Bartkevičiūtė, 2016; Albers & Suwala, 2018) This relationship between family frms and spatial entities may exert various effects on the surrounding community in terms of philanthropy (Campopiano, De Massis, &

### 240 *Hans-Hermann Albers and Lech Suwala*

Chirico, 2014), community citizenship (Berrone, Cruz, Gomez-Mejia, & Larraza-Kintana, 2012) or even regional business-led community development (Enright et al., 2016; Albers & Suwala, 2020a). All efforts have resulted in a growing, but not coherent, body of research focusing on the nexus between family frms and spaces.

### *Corporate spatial responsibility*

Whereas research on family frms is only beginning to systematically include spatial contexts, research in the feld of corporate spatial responsibility (CSpR) has yet to dedicate much attention to the unique aspects of family frms. Corporate spatial responsibility (CSpR) is an extension of the well-known concept of corporate social responsibility (CSR). As a concept, CSpR emphasizes not only corporate engagement within a broader social context, but also within a particular spatial – usually urban or regional – setting. CSpR focuses mainly on cases where frms make social and ecological commitments to their surrounding environments that go beyond their core economic competencies (Albers, 2011; Albers & Suwala, 2018, 2020b). In many ways, CSpR aligns with CSR principles embracing corporate ethics, social enterprise, corporate civic leadership, corporate and voluntary self-commitment and sometimes even corporate citizenship, to tackle manifold problems primary not connected with the principal economic activities of the frm (albeit those measures can be utilized for corporate objectives) (Hanson et al., 2010). What differentiates CSpR from CSR is the explicit spatial dimension and the respective commitment for locations or places at various scales (local, regional, urban, rural) (Knieling et al., 2012). We understand CSpR as a combined term for 'Corporate regional responsibility' (Schiek, 2017), 'Corporate urban responsibility' (Albers & Hartenstein, 2017) and 'Corporate regional engagement' (Lengauer & Tödtling, 2010). These predecessors locate CSR principles at particular spatial scales (Werna et al., 2009), whereas CSpR attempts to cover cases within any spatial context and any spatial process. Furthermore, existing disciplines tend to vary in their respective areas of focus, including the revitalization of the built environment (Albers, 2011), enhancing community life (Bürcher & Mayer, 2018) or fostering sustainable urban development (Suwala & Albers, 2020).

Earlier forms of the CSpR concept can be found in the period of industrialization when corporate or family frms fnanced urban (infra-) structures, supported local cultural and social institutions or constructed affordable housing for employees (e.g., Margarethenöhe [Germany, Ruhr by Alfred Krupp company dwelling estate] or Ford Homes in Detroit and Dearborn, US). Although traditional CSR instruments such as donations, sponsoring, patronage and charities are still widespread, novel measures such as public-private partnerships, corporate cooperation with the public sectors or civil society and long-term inter-institutional projects are on the rise (Albers & Suwala, 2018). These complex instruments have often an explicit

### *Family frms and spatial responsibilities* 241

spatial dimension and impact (Albers & Hartenstein, 2017). The variety of measures is far-fung, starting from simple frm-driven social, educational or recreational object-based building infrastructure (e.g., child-care centers) (Albers, 2011), revalorization (e.g., privately-owned spaces for public use, Kayden, 2000), relocation (e.g., corporate re-urbanization, Mozingo, 2011) or alteration of corporate premises (e.g., district renewal through cultural creative industries) (Suwala, 2015). Some of the most embedded practices range from involvement in business improvement districts or town center management schemes to more comprehensive private-sector-driven and business-community-led spatial development models (Hoyt, 2003; Coca-Stefaniak et al., 2009; Enright et al., 2016). These latter cases are particularly prevalent where there is a lack of consolidated government bodies or where public authorities display little effectiveness in pursuing development goals (Enright et al., 2016; Suwala et al., 2018).

### *Family frms and corporate spatial responsibility*

Our aim in this section is to cross-fertilize insights from both emerging discourses in family frms in spatial context and in CSpR to enhance the understanding of family frm-driven engagement in various spatial settings. For this undertaking, we have to revisit literature that has already made a connection between family frms and CSR/CSpR to gauge in what circumstances and with what tools these approaches might overlap. Although caution should be exercised when linking concepts to new areas of application, family frm studies and regional engagement enjoy enough affnity to suggest a successful merging of disciplines. Regional engagement driven by family frms is a centuries-old activity with different degrees of commitment and versatile applications depending on institutional frameworks and economic systems (e.g., decentralized systems, such as the US and Germany, are more benign to family frms or family leaders in certain regions) (Stimson, Stough, & Salazar, 2009; Hanson et al., 2010).

There is an elaborated research stream of family frms and CSR (e.g., Dyer & Whetten, 2006), with studies focusing on the differences between the relationships of family vs. non-family frms with CSR (e.g., Cruz, Larraza-Kintana, Garcés-Galdeano, & Berrone, 2014), the family ownership structure and CSR (Block & Wagner, 2014), different types of CSR (e.g., philanthropy [Campopiano et al., 2014], corporate citizenship [Astrachan-Binz, Ferguson, Pieper, & Astrachan, 2017], etc.) and family frms' social responsibility to communities (Niehm, Swinney, & Miller, 2008), amongst others.

Concerning the question whether family frms are more socially responsible, Cruz et al. found that, given their socio-emotional wealth bias, they have more pronounced relationships with external stakeholders, yet neglect internal social dimensions. Moreover, they do not necessarily comply with national standards and industry conditions the way CSR measures of non-family frms do, but, rather, choose other activities (Cruz et al., 2014). This aligns with the ideas that 'family frms disseminate a greater variety of CSR reports, are less compliant with CSR standards and place emphasis on different CSR topics' (Campopiano & De Massis, 2015, 511) and that they report less information on their CSR duties than non-family frms (Nekhili, Nagati, Chtioui, & Rebolledo, 2017). The same study found that family frms would rather beneft from communicating their commitment to CSR, as they could obtain shareholders' support more easily than non-family frms (Nekhili et al., 2017, 41). In general, fndings on whether family frms are more socially responsible or not are ambivalent depending on the size of the frm or feld of commitment, among other factors (Cruz et al., 2014). With regard to family frm engagement in communities, Niehm et al. (2008) point to three dimensions – commitment to the community, community support, and sense of community – as striking features of family frm involvement.

In times of growing interactive and collaborative modes of governance, more attention needs to be paid to the socially responsible leadership question and its spatial context, especially at the regional or municipal level. Hence, space becomes a crucial factor. In this realm, family frms and CSR affects regions, cities, or rural communities and are affected by those spatial entities (Albers & Suwala, 2018, 2020a). Devinney even calls for a

rethinking the meaning of space and place (…) as the rise of CSR is part and parcel of a more general phenomenon that is redefning the fundamental meaning of sociopolitical and economic geography. As corporations and economies have globalized, the sociopolitical structures have lagged behind.

(Devinney, 2011, 329, 339)

 Interestingly, empirical studies dealing with family frms and CSR have largely ignored space or place with regard to both where family frms implement CSR and the impact of such initiatives on spatial entities (Albers & Suwala, 2018). Space was rather considered a by-product, albeit a favorable one (e.g., in small rural communities [Niehm et al., 2008]; small business communities [Peake, Davis, & Cox, 2015]; etc.). There are only few studies that explicitly deal with the relationship between family frms and CSpR – e.g., Albers and Suwala (2018, 2020b) and Graffenberger and Görmar (2020) considered family frm-driven engagement in and for spatial entities – leaving room for more research on the nature of 'space-based corporate responsibility of family frms'. For our research, the following guiding questions are of interest here : Firstly, *how do family frms contribute to CSpR?* Secondly, *are there different intensities of family frm-driven engagement and how do those intensities relate to each other?* 

# *Methodology*

In light of this theoretical background, we provide selected case studies of particular German-based family frms in order to shed light upon the great variety of their CSpR initiatives (frst research question) and their impact/ intensity on spatial entities (second research question). Hereby, we utilize the CSpR maturity model that differentiates between 'degrees of responsibility/ spatiality' of CSPR measures (Albers & Suwala, 2018, 55). Based on the CSR maturity model (Schneider, 2012), Albers and Suwala (2018) identifed four different types of CSpR that vary according to the 'degrees of responsibility/ spatiality'. CSpR 0.0 and CSpR 1.0 encompass low-threshold/conventional engagement activities resulting from compliance with laws/rather accidental effects (CSpR 0.0) or philanthropic engagement (CSpR 1.0) with rather passive spatial impacts; both remotely (if at all) align with corporate objectives. CSpR 2.0 might possess a more systemic design that targets purposeful regional economic and societal synergies between the city/region and the family frm. Finally, CSpR 3.0 activities proactively interfere in duties once assigned to the public authority, or in matters of government with a clear spatial and societal impact (Albers & Suwala, 2018, 2020b). Examples of CSpR 3.0 measures include, for example, the development of corporate spatial master plans or initiatives that cross-cut multiple areas of policy within the territorial domain of public authorities.

The analysis of the relationship between family frms and CSpR, both in the form this relationship takes and in terms of the intensity of its spatial impact, is based on a selection of case studies of corporate spatial engagement of German family frms. This selection rests on a continuous monitoring of CSR activities in the context of urban and regional development over the past 10 years; the evaluation of relevant studies, databases and research work and the authors' participation in several research projects. For the case study selection, three methodological steps were taken into account. First, the choice of examples rested on 'purposive sampling' (Patton, 2001) where selected examples were either easily accessible or where personal participation was involved. Second, the relevance of examples was checked by analyzing their frequency in daily press and academic journals. Third, the variety of examples was ensured by the 'go for polar types' approach (Meredith, 1998) in order to identify different degrees (e.g., extreme cases, typical cases, maximum variation of cases, intensity sampling, critical cases, politically important or sensitive cases, etc.) of CSpR initiatives of family frms. Since our background is in regional and CSpR studies, we grouped different CSpR measures of family frms according to their degree of 'degrees of responsibility/spatiality'. For this assessment, the pyramid in Figure 12.1 gives insight into how we can link and classify different types and intensities of family frm CSpR. We have elected to pursue a plausible rather than a representative approach, in which cases were chosen based on their accessibility rather than their claim to representativeness.

# **Analysis**

The following analysis presents selected case studies of family frms engaged in urban and regional initiatives. It attempts to classify those examples into

*Figure 12.1* Family frms and the corporate spatial responsibility model. Source: Adapted from Albers and Suwala (2018), 55 based on Schneider (2012).

four types of engagements based on the above-mentioned CSpR maturity model (Figure 12.1). The frst step of this analysis will be to show different examples and forms of CSpR initiatives by family frms. In a second step, these examples will be classifed based on spatial intensities. The analysis is not intended to be exhaustive, but to illustrate plausible links between family frms and CSpR initiatives.

# *Family frms and the sponsoring of social, educational or recreational infrastructure*

One traditional model of regional engagement by family frms is regional philanthropy, such as the fnancing or promoting buildings for social, educational or recreational purposes (Feliu & Botero, 2016). In the 19th century, donations for art and cultural buildings helped cities develop a new civic identity. This tradition is still widespread among business patrons and family frms and, in addition to prestige for donors, is also intended to enhance the image of cities and their centers (Rectanus, 2002). A modern interpretation of this mechanism is the so-called the 'Bilbao Effect' (named after the Guggenheim Museum built in the Basque city of Bilbao by Frank Gehry) (Plaza, 2008). The term describes the general strategic use of icons (symbolic buildings) in order to revive a city economically and culturally (Suwala, 2014). This effect acts as a role model and has been also used for partly or fully family frm-fnanced art projects since then. In what follows, we want to showcase some examples from the German context.

### *Family frms and spatial responsibilities* 245

In the city of Herford (North Rhine-Westphalia, 67,000 inhabitants), the local furniture industry, mostly driven by family frms, mutually commenced a corporate initiative with local politicians and the famous architect Gehry to realize the Marta Herford project (a Bilbao-inspired art museum) based on an impressive selection of works by contemporary artists of a local family frm owner and art collector. In Hamburg, the construction of the Elbe Philharmonic Hall (the city's new landmark) was partly founded on a donation initiative by the city's family entrepreneurs. Other cases particularly in smaller German cities show that family frms or their owners (frequently world market leaders in certain economic niches) are often the driving forces behind such initiatives. Cultural buildings sponsored by family frms are used to brand city centers (e.g., the Kunsthalle Weishaupt in Ulm, the Museum Barberini in Potsdam [Hasso Plattner, SAP] or the Knauff Museum in Iphofen). All these projects are based on a strong site solidarity of the CEOs or family members (with a prolonged locational entanglement) and have similar expressions in countries outside Germany (Albers, 2011; Basco et al., 2020; Suwala & Albers, 2020).

Apart from these philanthropic, branding initiatives, family frms are increasingly investing in local built infrastructure, especially in educational or recreational facilities for use by their employees and the surrounding citizenry. In one example, the company Stihl is building a daycare center in Waiblingen to serve not only their employee's children, but also children in the surrounding neighborhoods. These efforts are often meant to attract employees, enhance quality of life and recruit new graduates. A similar trend can be seen in the renaissance of company-fnanced and owned dwellings. In Memmingen (Bavaria, 44,000 inhabitants), Häussler, a family frm that runs a bakery chain, is planning to build employee apartments at the company headquarters. As the lack of affordable housing continues to be a growing problem in many German cities, companies whose employees fnd it diffcult to fnd nearby homes are investing in this housing themselves. Although many of these initiatives have a mediate effect on regional development, this broader impact is seldom the intention of the engagement.

### *Family frms and their impact on re-urbanization, district renewal and privately owned public spaces*

Another type of spatial engagement driven by family frms is founded by corporate support for urban renewal and individual public space initiatives and is best understood in light of changing urban spatial dynamics over the last few decades (Albers & Suwala, 2020b). Most contemporary cities host a series of competing urbanizations and increasingly complex urban migration patterns of both residential and commercial uses. Departing from the well-known 'doughnut effect' where many cities and municipalities are struggling with deteriorating centers and where retail moves to shopping centers, industry is migrating to business parks and new housing is being created in many places on the outskirts (Sternlieb & Hughes, 1980); setting up activities around cultural creative districts such as the aforementioned familydriven commitment to art and cultural buildings is a frst step (Suwala, 2015). A more developed approach, however, is family frm re-urbanization, where companies move back to city centers, directly or indirectly leading to local economic development, for example, for the revitalization of centers (e.g., corporate family frm campuses or universities). There are instances of big U.S. corporations, some of them family frms, that have recently returned to city centers after decades of corporate suburbanization (Mozingo, 2011). Although frms often have clear business-related incentives for this geographical return, in particular to attract young workers, particularly interesting cases emerge when corporations move into certain areas with the intention of contributing to local economic and community development and city revitalization (Sutton, 2010).

One example from Germany can be found in Mertingen (Bavaria, approx. 4,000 inhabitants). The town center of Mertingen is currently undergoing remodeling. Zott (a family frm founded in 1926 with around 3,000 employees currently and one of the leading dairy production companies in Europe) serves as an important economic player in the region. Supported by the Ministry of Construction of the Free State of Bavaria, the corporate initiative attempts to revitalize the town center and turn it into a lively and attractive place. Zott is modernizing parts of its inner-hamlet old premises, relocating their headquarters and administration here. In addition, the project initiates further developments such as the conversion of the main square or the refurbishment of adjacent buildings.

A further characteristic of some cases of re-urbanization by family frms is the integration and opening of corporate space for public use (Albers & Suwala, 2018). This phenomenon refects an old trend starting in the 1950s in the dense urban fringes of central business districts in global cities (New York, Tokyo, Hong Kong and new metropoles in China), called 'privately owned public spaces', where large corporations were opening up their properties in manifold ways to the public (e.g., as arcades, urban plazas, through blocks or covered pedestrian spaces, etc.) (Kayden, 2000).

# *Family frms' interest in business improvements districts, town-center management and family frm-led master plan initiatives*

The engagements that we have discussed so far have been focused on the implementation of spatial measures, either through the sponsorship of individual buildings or urban infrastructure or by participating in broader reurbanization-based development schemes. But some family frms are also involved in more strategic tasks of urban management in a variety of spatial contexts. These include direct involvement in planning processes on a district level, as well as in business improvement districts, in town center management schemes or in the preparation of spatial master plans for the city or region (e.g., Hoyt, 2003; Albers & Suwala, 2018). This is far from new. The Plan of Chicago or the Burnham Plan of 1909 is considered the most famous example of a master plan initiated by the private sector. The plan was commissioned by the Commercial Club of Chicago, however, it was only partially implemented. The principal aim and purpose of the initiative was to ensure the city's functionality and competitiveness (cf. Burnham & Benett, 1909). These examples are still relevant today and appear in manifold ways, varying in intensity and obligation and the extent to which family frms are involved.

A frst mode, originating in Canada and the U.S. from the 1970s on, are business improvement districts (given similar names depending on the country; e.g., 'downtown improvement districts' in Japan, 'main street associations' in New Zealand [Hoyt, 2003], etc.). Business improvement districts are among the most widespread initiatives in local governance and represent a geographically defned area where the majority of property owners and/or merchants agree to provide an enhanced level of public service by imposing an additional tax or fee on all the properties and/or businesses in the area (Mitchell, 2001). The idea is to 'channel private-sector energy toward the solution of public problems' (MacDonald, 1996, 42). Studies from Germany have shown that family frms are slowly starting to support business improvement districts by marketing downtown districts, providing additional infrastructure (e.g., sanitation and security services), advocating public policies that promote downtown interests and acting as drivers of urban regeneration (Faller & Wiegandt, 2010). A second similar but citywide mode is town-center management, which can be roughly defned as 'a coordinated pro-active initiative designed to ensure that (…) city centres are desirable and attractive places. In nearly all instances the initiative is a partnership between the public and private sectors and brings together a wide-range of key interests' (Coca-Stefaniak et al., 2009, 75). An interesting case study is 'Berlin Partner', which is a unique public-private partnership set up by the Berlin State Senate and over 280 frms (many of them family run), dedicated to promote marketing, city business and technology support for companies, investors and scientifc institutions in the German capital.

A third mode, and probably the most infuential one in family frm-driven regional engagement, are so-called private sector-led master plan initiatives (Morrison, Wilson, & Bell, 2012; Albers & Suwala, 2018). These initiatives inherit a 'combination of district, city and state authorities or government agencies on issues from business conditions to a city's broad long-term agenda' (Enright et al., 2016, 3). Here, the private sector or family frms initiate far-fung regional development measures in cities where there is discontent with planning frameworks but that have a robust business presence or weak/consolidated city authorities. Weak authorities manifest for various reasons (e.g., fragmented or localized authorities, subject to control from strong regional or national authorities, a lack of advanced infrastructure or insuffcient investment, tax income or spending power). Such a measure in the German context is the so-called business association 'Unternehmer für die Region Köln' (many of them family frms). Together with the Chamber of Commerce of Cologne, they charged a famous architectural offce with drafting a master plan for the city. Two further master plan initiatives initiated by the private sector have recently attracted considerable attention in Germany. In Mönchengladbach (North Rhine-Westphalia) a master plan was created in 2012/2013 with fnancing from local (family) frms (https:// mg3-0.de). All initiatives were subsequently confrmed by the city council after public hearings and have since then become the de facto 'offcial' urban development concepts.

Our fnal example encompasses a business community-led model initiated by a single-family frm and executed through a private sector-driven urban development agency. The Duderstadt 2020 initiative is the idea of the family entrepreneur Hans Georg Näder (owner & CEO of Otto Bock GmbH from Duderstadt, a mid-size town in Lower Saxony, Germany; the company is the world market leader in orthopedics). The project has been developed and organized by a team from the regional University of Applied Sciences (HAWK) since 2009. Against this background, a limited company (Duderstadt2020 GmbH) was founded to guide this initiative and to strengthen the attractiveness of the town (e.g., by boosting amenities, quality of life, employment). The process comprised the following consolidated actions: setting up a master plan, strengthening networks (e.g., corporate, public and civic among different stakeholders), stimulating citizen participation (such as discussion forums, future workshops, etc.) and fostering neighborhood development, among many others. In the meantime, Duderstadt2020 is regarded as the city's urban development agency that traces an 'integrative urban development management' within six areas of interest (urban marketing, tourism promotion, economic promotion, cultural promotion, neighborhood development and social interaction). Most recently, the Duderstadt 2030 city vision and the 'Futuring Duderstadt' master plan were presented (Albers & Suwala, 2020b).

### *Intensities of spatial engagement by family frms*

In this section, we will compare the types and intensities of the spatial engagements by family frms that were traced in the sections above. In the "Family frms and the sponsoring of social, educational or recreational infrastructure" section, we looked at examples of family frms and their sponsorship of social, education or recreational facilities within their local spatial setting.

These activities, although they sometimes have extenuated spatial impacts, are usually not pursued for their effects on the spatial environment. For this reason, they are more in line with traditional CSR activities. Spatial effects are usually dependent on the type of built infrastructure and its particular spatial radiance (compare, for example, an art museum and a kindergarten). Architectural icons are often guided by patronage and personal motives of the family entrepreneurs or their frms; they rather represent traditional philanthropic initiatives and, if they exert any spatial impact at all, this is usually indirectly, as fagship projects for promoting or reviving a city or district. These types of initiatives correlate to levels CSpR 0.0 or CSpR 1.0 from the CSpR maturity model (Figure 12.1).

In the "Family frms and their impact on re-urbanization, district renewal and privately owned public spaces" section, we highlighted projects where frms are engaged more comprehensively with their spatial surroundings. By offering private spaces for public use, or by committing to large-scale fnancial or locational investments for reurbanization and revitalization purposes, family frms, often in cooperation with municipalities, have a much clearer and substantial link to urban and regional development. In these cases, the needs of the city or district and the requirements of the company coalesce; the city's desire to ease the local housing market, for example, can be met by private frms supplying workers with frm-owned housing. Still, there are varying degrees of spatial impact here. Permitting public use of frm-owned spaces, for example, has a more locally confned spatial impact than relocating a corporate headquarter, which in many cases has a tremendous effect on entire villages or urban districts. The Mertingen case also shows a family frm acting as a main driver for urban revitalization, especially in situations where public funding, collaborative know-how and companies' sensitization for local issues abound. In the most integrated cases, these measures can unfold into purposeful regional economic and societal synergies and help to contribute to win-win development scenarios. These cases correspond to the classifcation CSpR 2.0. Often, however, the examples laid out in the "Family frms and their impact on re-urbanization, district renewal and privately owned public spaces" section still fall into the classifcation of CSpR 1.0.

The "Family frms' interest in business improvements districts, towncenter management and family frm-led master plan initiatives" section details a number of cases where family frms are engaged in strategic local or regional engagement and mid-term regional planning. Included in these cases are frms' involvement in business improvement districts, town center management schemes and private sector-led master plan initiatives. On average, these cases display a much higher impactfulness on urban and regional planning and management, as well as on the surrounding spatial environments. Moreover, all measures of this variety have an explicit spatial focus and fulfll many of the criteria outlined for CSpR 2.0. Often these initiatives are led by associations of proactive frms rather than individual companies and are willing and able to design, fnance and lead certain processes within their spatial setting. Generally speaking, this type of engagement by family frms unfolds much more effectively if the public sector is involved to advise and steer activities, at least in German cases. Participation of the civil sector is also an important indicator for general success. This not only leads to more comprehensive inclusion of stakeholders, but also helps private sector-driven initiatives to gain credibility for their actions and plans. Because many of these initiatives have a trans-sectoral focus (i.e. they target not only economic but also cultural, educational, social or developmental issues), they are usually most successful when they integrate a cross-institutional, systemic, long-term approach. Accordingly, they often lead to signifcant place leadership by the family frms involved. All master plan initiatives laid out in our analysis ft the criteria for intentional CSpR 3.0 measures. They have far-reaching implications for residential, commercial and industrial planning, as well as for land use, the built environment and far-reaching spatial implications within the wider regional community. The commitment is often sustained by the existence of independent private sector-driven urban development agencies set up as autonomous legal entities (see Figure 12.1).

Based on the cases above, we think that our classifcation of CSpR fts appropriately to family frms. Many possible exceptions exist, however, that would require a deeper case-by-base analysis. One could imagine, for example, a family frm engaged in frequent and extensive philanthropic sponsorship but without a comprehensive spatial thread or strategic aspirations, despite the great fnancial commitment. Furthermore, some spatially impactful initiatives, such as the allowance of public usage of private land or involvement in a business improvement district, may be legally mandated by local zoning or building code ordinances and may therefore not fulfll the criteria of CSpR at all on account of being non-voluntary.

# **Conclusion**

Our goal was to answer two questions about family frms and corporate spatial responsibility, namely how (forms) and to what extent (their spatial impact/intensity) family frms implement CSpR measures, or in other words whether family frms are spatially responsible. We have been able to show that family frms are spatially engaged in manifold and unique ways and with varying intensities. Examples above ranged from CSpR 0.0, engagement which demonstrates little to no spatiality, to CSpR 3.0, where corporate engagement has strategic, far-reaching and long-lasting spatial impacts. CSpR adds to the regional policy toolbox by establishing alternative models for local and regional engagement, development and governance. This becomes even more spatially salient when family frms are involved in the design, implementation, participation and even leadership of these activities. The main feature of this engagement is the regional orchestration and framing of activities by family frms – in tandem with state authorities – in order to foster economic development or other favorable outcomes in the region. These models align with the rise of non-state 'place-based' economic development strategies at a time when state and society at large are more reliant on non-state actors (Albers, 2011; Harrison, 2014). CSpR initiatives by the private sector and by family frms are constantly exposed to nuanced and manifold critiques in accordance with their placement within rising neoliberal tendencies of spatial planning and management and an increasing hesitancy surrounding corporate infuence of the public sphere (e.g. Ward, 2006). Our examples demonstrate this fear is in part comprehensible (take, for example, the widespread infuence of the family entrepreneur in Duderstadt), especially if the civic sector is not appropriately included or even intentionally is neglected from strategic planning measures (as was the case in Masterplan Cologne).

Notwithstanding, we think that a coordinated and appropriate level of regional engagement by family frms can create synergies for communities and the spatial setting in which frms operate. Moreover, in small towns or rural areas, or in jurisdictions lacking consolidated government bodies or demonstrating low effectiveness of public authorities, family frm-driven community development may be without alternatives (Horlings & Padt, 2013; Suwala et al., 2018). By considering family frms and their corporate spatial responsibilities in Germany, we have sought to contribute to theoretical, practical and policy-based conceptualizations of this phenomenon. We have provided examples for different degrees of corporate spatial responsibility as performed by family frms, expanding the spatial dimension within research of family frms and corporate social responsibility (Campopiano et al., 2014; Campopiano & De Massis, 2015). Practically speaking, understanding the role of family frms in CSpR initiatives adds to the regional policy toolboxes that stakeholders use to enact urban and regional change. Still, our analysis provides only an overview of possible spatial embeddings of family frms' engagements (for other types, see Selcuk & Suwala, 2020). In cases of practical application, this analysis needs to be appropriately tailored to the respective modes of governance, economic systems, social norms and regional particularities. Finally, this chapter demonstrates the strategic nature of family frms' engagement with CSpR and the resulting implications this can have for urban and regional planning and management. We emphasize the importance of tri-sectoral negotiations (between family frms, the state and civic society) to ensure the effectiveness and value of these initiatives; only transparency, communication, reciprocity and a mutual sensitivity to the needs of partners will allow problems to be solved. By enlisting family frms into CSpR policy, policy makers have the ability to increase confdence within the private sector and unleash the potentials of the strong connection family frms tend to have with their home regions (Bird & Wennberg, 2014; Basco et al., 2020). Future lines of thinking and research could either concentrate on some of the specifc ways in which family frms can engage in CSpR, or could conduct a cross-country comparison of best practices in this feld to provide policy makers with a better understanding of when, why, how and under which circumstances family frms should be integrated within spatial development agendas.

# **References**

Albers, H.-H. (2011). *Corporate Urban Responsibility*. Frankfurt: Campus.


# **13 Place-based approach and family frms**

The Tatula Programme in Lithuania

*Rodrigo Basco and Inga Bartkevičiūtė*

### **Introduction**

The process of globalisation has brought benefts for less developed regions and enabled them to become a part of the broader systems of production, distribution, and consumption. However, globalisation *per se* does not reduce the inequality gap among regions. Although some regions have benefted from the opportunities arising from it, in general, social and geographical polarisation has increased (Pike, Rodríguez-Pose, & Tomaney, 2006); that is, most regions around the world are struggling to identify and utilise their potential to be part of the globalisation process.

Throughout this process, policymakers have tried to understand and correct the inequality gap. They have implemented, with different degrees of success, specifc actions to overcome the problems due to which regions miss the benefts of globalisation, by using the traditional approach rooted in the classical economic growth theories. These strategies were traditionally designed to tackle regional disparities by applying the so-called 'topdown' approach, that is, national governance institutions played a critical role in developing strategies. Top-down strategies tend to focus on defciencies associated with a region's economic problems, applying a universal approach to tackle these problems, regardless of the local context. However, such traditional regional development solutions have been only partially successful, encouraging academia and policymakers to rethink the concepts of regional development.

Since the late 1990s, the place-based approach to economic development has been related to a more holistic interpretation of the causes and consequences of local and regional economic development. During this time, local and regional economic development theories and practices have witnessed a shift from a frm-oriented sectoral approach towards a more in-depth consideration of unique local factors, resources, and actors (Rogerson & Rogerson, 2010), recognising the importance of bottom-up actions characterised by comprehensive analysis of regional micro-foundations. This change in approach to local and regional development was also prompted by the process of decentralisation and the growing empowerment of sub-national governments all over the world (Rodríguez-Pose & Wilkie, 2017).

Even though important theoretical and practical advances have been made in framing and understanding the place-based approach—within the micro-foundations perspective that accounts for local socioeconomic actors and the participative processes to create and implement actions aimed at energising less developed geographical areas or regions—local family frms have received less academic attention (Basco & Bartkevičiūtė, 2016). More specifcally, the importance of local family frms as a node in the relational space has not been recognised. Therefore, this chapter aims to explore the role that local family frms may play when place-based approach initiatives are implemented in lagging regions. We present, explore, and analyse the case study of the Tatula Programme in Lithuania.

We can summarise our results as three important characteristics of family frms for place-based approach. First, the entrepreneurial orientation of family frms leverages place-based programme actions by using existing local knowledge to reconfgure resources in different and innovative ways. Second, the long-lasting embeddedness of family frms, characterised by their social capital, embraces other socioeconomic actors in their initiatives to unfold a participative developmental approach. Finally, the connection between family frms and their territory creates a sense of belonging to generate the intention to stay in the region, favouring local action in the long term. This chapter contributes to both theory and practices. We expect that this exploratory case study will open the door to better understanding the micro-foundations of place-based approach as a lens to promote regional socioeconomic development. Additionally, the case study that we present highlights practical contributions by illustrating the importance of encouraging local actors—in this case, family frms—to create long-lasting conditions that boost economic dynamism.

# **Framework**

# *Traditional approach*

One of the main challenges for policymakers in terms of regional development is to tackle economic disparities between different regions. Economic growth theories and some policies, in the second part of the 20th century, were dominated by a macroeconomic approach (Scott & Storper, 2003; Vasquez-Barquero, 2010) and designed for national income growth (Vasquez-Barquero, 2010). Traditional policies to reduce the inequality gap between leading and lagging areas were designed and implemented by applying a top-down approach. Such top-down strategies tended to tackle the most important development shortcoming that was seen as the main impediment for economic development, thus ignoring other relevant factors (Pike et al., 2006) such as local culture, human resources, or technological context. These policies aimed at reducing location defciencies by replicating the regional development path of now-developed regions. Most strategies focused on infrastructural investments to build highways, railways, and telephone lines, and on attracting large manufacturing companies to less developed areas (Pike et al., 2006). However, this approach did not produce the expected results of diminishing regional inequalities.

Since the 1960s, the regional planning policy framework in various countries (e.g., Italy and France) has been highly affected by well-known growth poles theory (Perroux, 1955), which suggests that a growth pole is formed by a group of industries connected through input-output linkages. Although the original theory is not related to a particular geographical space (especially one that is characterised as a country with its regions), and defnes growth pole as an economic category, other scholars proposed regional implications of the growth poles theory. Boudeville (1966) transformed the idea of a growth pole in terms of place and suggested the concept of geographic clustering of economic activity. In practical application, growth poles were treated as urban centres with the anticipation that an industrialised area will also stimulate the development of adjacent regions. Specifcally, during the 1960s and 1970s (Higgins & Savoie, 1988), this approach manifested mainly as an attraction of industrial frms to urban centres located in lagging regions. Most of the strategies dealing with regional development were designed at the national level and had a tendency to be replicated across different regional contexts.

The implementation of regional development policies based on poles brought some positive impact on the reduction of regional disparities, as it contributed to an increase in employment and income (Vasquez-Barquero, 1999). However, the critical question lies in the sustainability of the generated impact, that is, to what extent the policies induced self-sustained local development.

An interesting example is the long-lasting effort to reduce differences in development between the southern and northern parts of Italy. Following the growth pole theory and the related approaches, a strategy to tackle the problems of underdeveloped southern Italy (Mezzogiorno) was implemented during the 1950s–1980s. Various locations in the region were designated as growth centres, to establish big state-controlled industrial frms and to attract smaller frms that would supply goods and services to the local market. (Pacione, 1982). However, these massive plants did not embed into the local context, which was characterised by weak endogenous industrial structures and was missing industrial networks that would foster sustainable local development and employment. Therefore, these plants became 'cathedrals in the desert' as most suppliers and consumers were not located in the vicinity (Trigilia, 2012).

Even though one of the problems in the Mezzogiorno intervention can be related to the loss of effectiveness during the implementation phase (Felice & Lepore, 2017), the leading causes of the failure seemed to point at insuffcient attention given to the local cultural, social, and economic context. For instance, in the context of an agricultural region, investments were designed to develop large and capital intensive industries. It created tensions at the very beginning of policy implementation because of the radical transition from an agricultural to an industrial economy. Even more, one of the major diffculties was the takeover of agricultural land—which may have belonged to a farmer's family for several generations—for industrial use (Pacione, 1982). Additionally, the newly established factories lacked skilled workers as regional human capital did not embody the required specifc knowledge, culture, and other attributes associated with the new industries.

The main limitation of the top-down approach and the universal regional development strategy is its spatial-blindness. This can be seen as one of the reasons why the practical application of standardised development policies in different conditions seldom generate the anticipated impact on sustainable long-term economic development (Storper, 1997) and have even created adverse effects (Pike et al., 2006). A lack of proper evaluation of the local context in many cases has led to the jeopardising of local entrepreneurial resources and local frms that were not able to compete with major industrial players attracted to the region (Vasquez-Barquero, 1999). Although it was expected that non-local, large, and technologically advanced frms relocated in less developed areas may induce local business development by generating new linkages, thereby triggering entrepreneurship, and expanding knowledge spill-overs, this was not always the case. In most cases, new frms relocated in less developed areas exploited the lagging regions simply as new labour supply markets (Pike et al., 2006).

### *Place-based approach*

The aforementioned lack of success in strategies to reduce inequality among regions prompted new discussions in the regional development feld during the late 1990s, offering greater impetus to explore alternative dimensions such as the location and its cultural, economic, institutional, technological, and political contexts. The fact that not all regions developed and innovated at the same pace raises a question about something underneath traditional developmental policies that either boosts or hinders prosperity. Thus, the idea that regional development today should follow the patterns of now-developed countries and rely mostly on the development of mega-urban regions was questioned (Barca, McCann, & Rodriguez-Pose, 2012). Although this approach is attractive—as it allows the building of universal development models looking at the experience of now-developed regions and locations, and analysing long-term dynamics of their development—the complete reliance on universally-presented tendencies carries a risk of misjudging the relationship between universal causes and local context (Storper, 2011).

What seems to be overlooked while promoting space-neutral interventions is the need to leverage local potential, which has become one of the main focuses of the place-based approach to regional development. This approach emphasises local potential and means to maximise it rather than only introducing universal economic growth solutions or simple administrative redistribution of wealth to achieve regional convergence. Thus, while the place with its characteristics is recognised as a factor in economic growth, the place-based approach considers it a relational one in which proximity has a meaning, not only in terms of geographical proximity but also in terms of institutional, cognitive, and organisational proximity.

Successful place-based policies cannot be written using a ready-made universal template. Instead, they should be developed in close cooperation with local and external actors, and identify and rely on locally embedded knowledge. Moreover, for example, policies aimed at increasing the local economy cannot merely tackle the competitiveness of local frms but should also evaluate all the contextual conditions, including human capital, physical infrastructure, and so on (Rodríguez-Pose & Wilkie, 2017); that is, they should analyse the region holistically. As Rodríguez-Pose and Wilkie (2017) state, in terms of economic growth the question should be 'not *if* different localities have the potential to achieve sustainable and inclusive economic growth via territorial approaches to development, but rather, *what* must be done to ensure that they do, and relatedly, *how* do they do it'.

Consequently, the application of the place-based approach leads to a greater variety of regional development strategies (Rodríguez-Pose & Wilkie, 2017), opening the door for more complex solutions that cannot be achieved in a simplifed solution based on a 'one size fts all' approach (Barca et al., 2012). In practice, the interpretation of the place-based approach varies from government to government. As it does not have theoretical underpinning, its practical application depends highly on local context and traditions. For instance, the European Union (EU) document 'Territorial Agenda of the European Union 2020' calls for greater awareness of the territory and aims to unleash the local potential using location-specifc assets and factors that contribute to the competitiveness of places. Following this approach, the European Commission uses dedicated instruments and requires member-states to develop bottom-up place-based strategies to receive funding under specifc EU funding programmes, recognising the place-based approach as an essential tool in enhancing territorial cohesion. Although the general requirements for the implementation of place-based instruments are common to all countries, the adaptation of the framework in different member states and its sub-regions suggests a vast array of local strategic solutions (CSIL, 2015; BGI Consulting, 2019).

However, theory and practice suggest some universal attributes of all successful place-based strategies. As the place-based approach is associated with decentralisation and the empowerment of sub-national governments, one of the main preconditions for its successful application is the vertical and horizontal integration of institutions and social partners (Barca et al., 2012). Therefore, place-based policies should be built on cooperation among national, regional, and local governments, and with socioeconomic partners.

#### *Local family frms in the place-based approach*

Place-based approach requires the local economic and social actors' involvement to develop actions within regional strategies. Their participation, engagement, and commitment are important characteristics to tailor actions that are going to affect their social and economic life. Endogenous actors are crucial contributors for developing, implementing, and evaluating regional strategies because they are the source of local knowledge. Moreover, they are part of the local social capital, which enhances strategic planning in terms of connections and networks. However, not all economic and social actors are the same and their contributions to successful design and implementation of regional strategies may depend on their specifc attributes.

Family frms pose specifc characteristics because of the active family participation in ownership, governance, and management decision-making. Because of this, family frms have a unique set of goals characterised by the coexistence of business- and family-oriented goals (Basco, 2017), altering the reference point for making decisions. For instance, while economic performance is an essential issue for any frm, family frms may accept lower expected economic performance at the expense of transferring the frm to next generation, to maintain the headquarters in their local territory or even to contribute and participate in the local and regional development through corporate local and regional responsibility actions (Lenz, 2020). In this sense, family frms become unique socioeconomic actors for policymakers to integrate into the debate on local socioeconomic development.

Local embeddedness of family frms could play an important role in applying the place-based approach for several reasons. First, family frms are guardians of local business culture, which could help interpret implicit codes of conduct, values, and beliefs when doing business. Second, family frms are important local nodes embedded in the socioeconomic network in the relational space through which information and knowledge fow. Finally, beyond the geographical proximity of family frms, they leverage other proximity dimensions such as organisational, social, and cognitive.

Even though family frms possess unique characteristics and a potential role in the local socioeconomic life, this specifc local actor has not been theorised in the micro-foundations of the place-based approach. Considering that family frms are the most common form of organisation in the EU (Ricotta & Basco, 2020), our research question is as follows:

*What roles do family frms play in place-based approach intervention?* 

### **The case study of the 'Tatula Programme'**

#### *Sustainable farming promotion in ecologically vulnerable regions*

The place-based approach can be applied in regional or local programmes varying in scale, topics, and policy instruments. Its main characteristic is

# 262 *Rodrigo Basco and Inga Bartkevičiūtė*

that it does not have a clear and universal pattern in terms of suitable interventions; instead, its philosophy is to carefully examine local circumstances to develop actions and implement them by embracing local actors and using the potential of local regional factors and processes. We aim to describe and explore a small-scale development programme that shows how locally tailored actions can help foster empowerment and unlock local potential in outlying areas.

# *Data*

We collected qualitative and quantitative data from different sources of information, such as the Lithuanian statistic offce, the certifcation institution 'Ekoagros', legal documents, academic papers, and additional information available in the national and local media. We also interviewed the management of the Tatula programme. Additionally, we used our observations of the Tatula programme development. For instance, we actively participated in organic farms to observe the interaction and action among Tatula participants and their customers. As customers of some Tatula farmers, we have witnessed their development.

# *The region*

Part of northern Lithuania is covered by karst landscape underlain by carbonate rocks (gypsum-dolomite). The dissolution of these soluble rocks produces underground drainage system with sinkholes. In karst landscapes, the precipitation infltrates into aquifer not only through the soil, which acts as a natural flter against some contaminants, but also directly through karst sinkholes that could be covered only with a thin layer of soil. Moreover, as underground sinkholes are often interlinked with each other, contaminants spread faster. These features are of particular concern when land is intensively used for agriculture purposes, because fertilisers, pesticides, and oil products remain in surplus in groundwater. Karstic landscape covers around 10% of Lithuanian territory, of which almost 30,000 hectares are characterised by intensive karstic processes. The geographical area of the intensive karst process overlaps with two local administrative districts located in northern Lithuania, namely, Birzu and Pasvalio.

# *The programme*

To reduce groundwater pollution, the frst initiative to implement environmental measures and to transform agriculture practices in northern Lithuania was launched back in 1993. The programme adopted by the Lithuanian government aimed at reducing groundwater pollution in the designated areas. A signifcant part of the programme was dedicated to the promotion of sustainable farming in the Birzu and Pasvalio districts and was implemented under the title 'Tatula programme'. The Tatula programme provides membership to any sustainable farming or processing entity, although the majority of its members are family-owned farms.

Although the funding of the Tatula programme has undergone various stages since its launch, activities promoting sustainable farming in northern Lithuania under the Tatula programme continue until today. The initial programme, which was adopted in 1993, was supported by national funding until the end of the 1990s. This period was characterised by the highest fnancial allocations to support farmers' transition to sustainable farming practices, promote environmental monitoring in the region, and construct wastewater treatment plants. Since 2000, the national government's attention to the specifc issues of the karst region has signifcantly diminished and various EU funds have emerged as the leading fnancial source for its implementation. In 2014, a new political initiative to revive the former strategy of environmental protection of vulnerable karst region was launched. Following it, in 2015, a national law on agriculture and rural development was supplemented by a new policy aimed at sustainable farming in northern Lithuania. Implementation of this policy is supported by triennial action plans (2016–2018 and 2019–2021), which include funds for Tatula programme activities.

### *Governance*

The Tatula programme is designed at the national level. However, local actors and agriculture scientists have actively participated in tailoring the programme's actions. The responsibility for implementing the programme was assigned to the Tatula programme organisation, which is governed by a multilevel governance structure formed by regional and local economic, social, and political actors such as farmers, agricultural frms, scientists, local politicians, representatives of non-government organisations, and other regionally acclaimed people. Since its launch, the Tatula programme has been organised in a form similar to cooperative, that is, its members (local farmers) have a vote and actively participate in the programme's development. The number of Tatula programme members has fuctuated, reaching a peak in 2000 with around 150 active members. The members' local embeddedness and prior connections have helped build a strong network based on existing social capital. It has not only become the programme's support network but also serves as a local knowledge and resource exchange platform amongst farmers, boosting the programme's positive impact on the region.

### *Programme's main activities*

The Birzu and Pasvalio districts are primarily dedicated to agriculture, which has been the main economic activity in the region. The vast majority of local agriculture entities were (are) small family-owned farms characterised by low business orientation. One specifc attribute of the Tatula programme, since its inception, has been to consider soft aspects of socioeconomic development as well as existing local knowledge when transforming economic practices. Thus, it suggested a means to enhance particular economic activities built on long-lasting local experience, rather than the reverse socioeconomic structure of the region. The programme was specifcally tailored to the needs of local natural conditions and economic activities.

The initial Tatula programme had a goal to convert 5% of all agricultural land in the two aforementioned districts into sustainable farming lands. It is important to highlight that in the 1990s, the practice of sustainable farming in Lithuania barely existed. Therefore, it was assumed that only the promotion of sustainable farming among local farmers through tax incentives, funding, or similar typical interventions would not be suffcient to achieve a long-term impact on farming practices. Consequently, along with subsidies for wastewater treatment facilities, soft loans for the development of agriculture business plans, and acquisition of equipment, among other classical instruments, the Tatula programme took a holistic perspective, including additional soft measures oriented to the particular needs of current and future farmers.

The programme began collaborating with agriculture scientists who were responsible for developing proposals on new farming practices, types of plants best suitable for the particular soil in these districts, use of organic fertilisers, and crop rotation. The programme funded the development of scientifc recommendations directly related to the specifcity characteristics of the karst region. Local farmers were also given multiple opportunities to directly consult with scientists specialising in karst landscape farming issues and to visit other farmers successfully applying particular methods. Importantly, the scientifc recommendations in multiple cases were developed in close collaboration with local farmers, whose long-running experience in farming in the particular landscape enriched scientifc approach with valuable pragmatic insights and increased its practical applicability.

Although the support for the development of new sustainable farming practices was an important aspect of the programme, another crucial issue was to induce the business orientation of small family-owned farms to convert their local agriculture practice into a viable economic activity. In this regard, the certifcation of sustainable agriculture farms and its production helped local farmers differentiate their products from agrochemical-based agriculture and access one specifc consumer segment in the market. In this sense, the programme leveraged the local entrepreneurial orientation by converting family farms into family business farms.

The national organic certifcation system was introduced in 1997. However, Tatula's management and co-working scientists soon realised that for some farms, local organic certifcation was still too early and too complicated. The problem was attributed to the complex requirements that the owner of organic certifcation had to adopt and the complicated process, which could become a burden to small, mainly family-owned farms. Therefore, since 2003, the Tatula programme has started introducing a regional certifcation system. It is based on a ground analysis of the local farming conditions and promotes the so-called sustainable farming system. This new certifcation entails the use of progressive farming methods that are less environmentally harmful than traditional ones but are easier to implement than organic certifcation. Some farms participating in the Tatula programme were certifed following the new system, even though the original sustainable farming certifcation system ceased to exist within a few years of its introduction.

Nevertheless, the system was benefcial to the region's environmental goals and organic agriculture businesses. It encouraged farms to progressively introduce less environmentally harmful farming practices and prepare themselves to turn their farms organic in the future. Additionally, this new alternative programme has opened several connections, support, and product distribution channels for farmers.

To ensure business continuity and new opportunities for farmers, the programme dedicated part of its resources to creating access to retail food markets. This initiative was important, particularly in the initial stages of the programme, because the regular sustainable farm-produce supply was insuffcient and could not adapt to the conditions of big retail chains. Therefore, sales through farmer's markets were identifed as the primary channel for creating business opportunities by meeting customers.

In the 1990s and early 2000s, the demand for organic food in Lithuania was low mainly because of a lack of awareness and the high price of organically grown products. The information regarding the characteristics of organic products and their benefts was scarce. The Tatula programme started organising and promoting farmers' markets under the Tatula brand in different locations of the country. Direct-to-consumer markets were seen as a good opportunity to create a new ecosystem of customers and farmers that understood each other in terms of needs and expectations. Although it began as occasional events, some locations have become meeting points where farmers and customers can interact regularly.

The Tatula programme employed various methods to encourage the link between farmers and customers. For instance, public funds were directed to cover farmers' participation expenses such as stall rent, market advertisements, and transportation costs. The transportation cost was introduced considering the geographical location of the target region and the distance from the capital, which is the leading market area for organic produce. Moreover, the Tatula programme actively promoted sustainable farming production on local and national media and, thus, made the Tatula brand visible. The programme's farmers have benefted from it and the 'Tatula programme' has become recognisable as a brand representing assured quality.

### 266 *Rodrigo Basco and Inga Bartkevičiūtė*

Some of the aforementioned activities of the Tatula programme continue today. Currently, public funds are used to further organise farmer markets, support new training programmes for organic farms, develop new organic products, and raise public awareness regarding multiple benefts of organic production. Additionally, the programme has increased its efforts in promoting farm-based processing of primary agriculture products and their certifcation.

### *Impact*

In Lithuania, organic farming barely existed in the 1990s. The volume of organic farming commenced gradual growth only in the late 1990s after the national certifcation agency was established and national organic farming support payments were introduced. By 2002, only 0.3% of the total agricultural land of Lithuania was occupied by organic farms, indicating the slow process of converting and convincing farmers to move from a traditional to an organic system. However, the Birzu district was characterised by a more rapid increase in the number of organic farms, accounting for more than 10% of organic farms in the country. This is mainly attributed to the Tatula programme's impact.

The aforementioned slow pace began to change after the entry of Lithuania into the EU in 2004, which led to an increase in fnancial support for organic farming. In 2012, already 5.4% of the country's agricultural land was used for organic farming. Birzu district also experienced rapid growth in organic farming volumes, with organic farms covering 14.4% of agricultural land during the same year. Although organic farming volumes continue to grow throughout the country even now, Birzu farmers remain amongst the leaders in the country's organic farming. The district has almost three times more intensive use of agriculture land for organic farming in comparison to other districts, on average.

The other type of impact, which can also partially be attributed to the Tatula programme, relates to the development of local agriculture farms in terms of production and farm viability. Constant support to diversify organic farms' products (including an increase in the variety of both raw and processed agriculture products) and established connections between local farmers and customers contributed to family-owned farms' long-term viability. This could not have been achieved in the conventional agricultural market, which is dominated by large businesses. Successful transition to organic farming was conditional on many families maintaining their businesses or even growing from subsistence farming to small businesses, thereby ensuring their fnancial income. Lasting family traditions and experience in organic farming encouraged some members of young generations to start their own organic farming business, while parents continued working on the old family farm. Moreover, the economic spin-off observed with the increase in organic farms prompted the establishment and local organic food production businesses (e.g., mills), which are closely interlinked with the region's organic farms and partially depend on their supply. All these circumstances are crucial to the region maintaining its cultural agriculture business ecosystem, which relies on local human resources and contributes to local socioeconomic development.

### **Results**

We applied our research to observe and understand the role local family farms have played in the Tatula programme and what could have been the key factors for its success. Based on our interpretation of the data collected, having local family farmers in the programme highlights several important consequences at three levels of analysis.

First, at the family level, business families were able to redirect their entrepreneurial efforts, generational knowledge, and physical resources into a sustainable economic activity. In this sense, the programme facilitated the reinvention of farmers' families and created competitive advantages. Active family participation in the economic activity created a sustainable business model diffcult for big corporations to imitate, which are more interested in proft maximisation at the expense of local and environmental conditions. Organic farming, the scale of farms that make it economically sustainable, and family participation are important conditions that implicitly or explicitly have made a generation of family farmers stay in their regions.

Second, at the local community level, the programme incentivised farmers to remain attached to their land, fnd alternative uses for local resources, and helped them avoid needing to migrate to urban areas. The programme did not use short-term incentives with the promise of fast results; instead, it succeeded by helping socioeconomic actors, such as family frms, fnd reasons to be embedded in their communities. Even though local family farmers were attached to their communities previously, the programme reinforced and renewed links and increased farmer family retention.

Third, at the national level, the programme helped connect rural and urban areas not only by meeting the supply and demand at a certain point in time and space but also by humanising the whole production system by facilitating physical meetings between producers (also sellers) and customers. This exchange of experience, needs, and expectations on market foors created a long-lasting relationship and mutual understanding. Direct links between producers, sellers, and customers materialise the sustainable economy.

Even though there are positive consequences from involving local family frms (business families) in the programme, family frms, because of their specifcities, could also bring unique challenges for the sustainability of the programme. To focus on the local needs and context, changing the situation of existing organic farms requires deeper considerations. Some local organic farms have been in operation for more than 20 years; hence, the problem of farming management succession, that is, the new generation, becomes more prevalent. Although are no statistics, our observations suggest that some successful family-owned organic farms whose founders or successors have died are changing the type of farming and returning to conventional methods that rely on synthetic pesticides, herbicides, and fertilisers. On the other hand, some family farms are able to successfully transfer the knowledge of organic farming practices to the second or sometimes even third generation of family members, proving the sustainability of organic production businesses. This observation is relevant to evaluate the programme in the long term and to redesign its future actions. Even though the programme has been successful so far, its sustainable impact is not guaranteed. The family dimension of the business would require addressing several challenges related to ownership and generational management changes.

## **Discussion and conclusion**

This chapter aimed to explore the micro-foundation of the place-based approach by focusing on family frms when developing and implementing initiatives to boost local socioeconomic development in lagging regions. We argued that family frms, characterised by the embeddedness of family members in the formal and informal institutional contexts, may act as essential nodes in the relational space, accelerating information fows, creating trust, and assimilating knowledge across generations.

Our main conclusion is that the place-based approach used in the Tatula programme was critical in creating incentives for long-lasting impact among local socioeconomic actors such as family farms. The family condition of the business target of the programme was important to its success because business families have several characteristics that leverage and reinforce programme actions, such as farming knowledge across generations, local embeddedness, local social capital, and the intention to stay in the region. The Tatula programme nudged the family entrepreneurial orientation to reconfgure resources and capabilities to give sense to the farmer activities and restore the association with the territory. It is diffcult to ascertain if such results could be achieved with non-family farms or multinational frms.

Even though the programme achieved positive results, several risks or challenges have to be addressed in the coming years. One of these is related to the generational and succession issues that any family frm has to overcome. In this particular case, family farmers have to navigate the ownership and management transition from one generation of family members to another. Several questions remain open: Is there any family business member who would like to continue with the farm? Has the current generation prepared for the coming generation to continue the leadership? How should the ownership and family wealth be distributed to guarantee the continuity of the family farm? Is it possible to distribute the family wealth when it is attached to the family frm or economic activity? How can the land be distributed among family members to continue farming? How can those family members that do not want to continue with the main economic activities be compensated? Is there any other alternative to compensate them without destroying necessary resources for farming activities? The future versions of the Tatula programme have to deal with these kinds of issues to create sustainable socioeconomic development across generations.

### *Theoretical and practical contributions*

The exploratory nature of our study could have several implications for theory and practice. First, our fndings have theoretical implications for those scholars attempting to explain the role that family frms play in local socioeconomic development. Our results reinforce the model proposed by Basco (2015), that is, the embeddedness of family frms in the relational space is an important characteristic that enhances the quality of proximity, not only in its classical geographical dimension but also in its organisational, institutional, social, and cognitive dimensions. Our results also highlight the importance of family frms for regional processes necessary to activate conditions for local development. For instance, in line with previous studies in Lithuania (Stangej & Basco, 2017), family frms can trigger entrepreneurial processes by reconfguring existing local resources, learning process by using and creating new knowledge, information exchange process by disseminating critical information and ideas, and spill-over process by encouraging family members or new actors to extend economic activities. The importance of family frms as economic and social nodes in the local economy has not been clearly understood, and future studies should investigate the micro-foundations of family frms in regional studies.

In addition to theoretical implications, our results also offer some practical recommendations for policymakers, who should pay attention to the soft local resources and processes that may boost or retard regional economic development while framing policies. The most important lesson is that economic actors are not rational in their behaviours and traditional incentives such as relocation of multinational factories through incentives (e.g., tax reliefs) do not necessarily work for all regions. To better understand the geographical, formal, informal, and historical aspects, local communities could help develop a unique solution to reactivate their economies. In this sense, embracing family frms could reinforce actions, leverage resources, and connect past, present, and future local knowledge for sustainable development. Future studies should explore other place-based approach programmes and actions to better understand the connection between family frms (as one socioeconomic actor) and the place-based approach. In this sense, future studies should focus on the micro-foundations of the place-based approach by considering factors, actors, and processes.

### **References**


### 270 *Rodrigo Basco and Inga Bartkevičiūtė*


# **Index**

Note: **Bold** page numbers refer to tables and *italic* page numbers refer to fgures. Page numbers followed by 'n' refer to endnotes.

ACFIC *see* All-China Federation of Industry and Commerce (ACFIC) Acquaah, M. 183 Acs, Z. J. 196 Adjei, E. K. 3, 83–99 ageing population 136 agency costs 85 agglomeration economies 67–68, 78, 90, 97, 98 Albers, H.-H. 4, 15, 237–251 Aldrich, H. E. 197 All-China Federation of Industry and Commerce (ACFIC) 213, 215–217, 224, 227, 231 Alsos, G. A. 197 Amato, S. 2, 3, 14, 65–80, 103–116, 140–153 Armington, C. 196 Astrachan, J. 7 Backman, M. 3, 140–153 Bad Berleburg 161–163, 167, 168, *169*  Baker, H. B. 211 Bartkevičiūtė,I. 4, 256–269 Baschieri, G. 14 Basco, R. 1–4, 7–23, 33–58, 65–80, 83, 180, 239, 256–269 Baù, M. 127 Becattini, G. 103 Berger, P. 197 Berleburger Schaumstoffwerke (BSW) 161, 162 'Berlin Partner' case study 247 Berrone, P. 154n2 Beta, K. W. 3, 177–189 Bika, Z. 154n2

Bilbao effect 244 Bird, M. 11 Bishop, K. 183 Block, J. H. 9, 11, 134 Boix, R. 105 Boschma, R. 88 Boudeville, J.-R. 258 Breusch-Pagan Lagrange Multiplier test 91 bridging social capital 181–182, 184, 187 Bryman, A. 183 BSW *see* Berleburger Schaumstoffwerke (BSW) Bureau van Dijk–Amadeus database 34, 194, 199, 205 Burnham Plan/Plan of Chicago 247 business-community-led spatial development models 241, 248 business improvement districts 238, 247, 249, 250 Calabrò, A. 2, 65–80 capitalism 33 Cappelli, R. 3, 193–206 Carter, S. 197 channel private-sector energy 247 Charron, N. 42 China Charity Donation Report 226–227 China market index 213 China's economic system 210 China's National Bureau of Statistics 214, 232 China Stock Market & Accounting Research Databases (CSMAR) 213,

215, 219, 220, 222, 225, 231

Chinese context 211–213, 229 Chinese family business: average family ownership *215*; corporate governance 217–218, *218*; CSR *226*, 226–228; cultural heritage stemming 213; data sources 213; diversifcation 220–222, *221*; family entrepreneurs 215–217; family founders/entrepreneurs 213; frm performance *223*, 223–224, *224*; innovation 218–220; institutional transitions 213; internationalization 222–223; limitations 231–232; overseas revenue *222, 223*; practical implications 230–231; private enterprises 214–215; private frm registration *214*; publicly family frms *214*; succession 224–226, *225*; theoretical implications 228–230 Chinitz, B. 198 Chrisman, J. J. 181 Chua, J. H. 181 Cliff, J. E. 197 CLRR *see* corporate local and regional responsibility (CLRR) cognitive proximity 19, 69 cognitive space 17 Communist Party of China (CPC) 216, *217*  community citizenship 125, 240 community engagement 158 community space 11–12 conceptual space 11 consultative strategy-making process 187 contingency space 11 continuum of descent 211 corporate citizenship 160 corporate engagement 160, 167–168, 171 corporate entrepreneurship 179 corporate giving 158, 160, 166, 170 corporate governance 106, 217–218, *218*  corporate local and regional responsibility (CLRR) 157, 160–161, 165, **166**, *169*, 170 corporate regional engagement 240 corporate regional responsibility 240 corporate responsibility in peripheral regions: CLRR 160–161; company specifcations 158; consciousness and strategy 161–168; contributions 171–172; from corporate engagement to cooperative action 168–171; engagement 158; facts 158; family frms and embeddedness 159–160; ownership structures 158–159

160, *226*, 226–228, 240–242 corporate social spatial responsibilities (CSsR) 237 corporate spatial responsibility (CSpR): economic competencies 240; goal 250; industrialization 240; intensities 15; maturity model 237, 243–244, *244*; measures 238; methodology 242–243; urban/regional setting 240; *vs.* CSR principles 240; *see also* family frms corporate support 160, 167, 171, 172 corporate urban responsibility 240 corporate volunteering activities 158, 160, 166, 170 Covin, J. G. 181 Cowling, M. 9 CPC *see* Communist Party of China (CPC) Cruz, C. 241 CSMAR *see* China Stock Market & Accounting Research Databases (CSMAR) CSpR *see* corporate spatial responsibility (CSpR) CSR *see* corporate social responsibility (CSR) CSsR *see* corporate social spatial responsibilities (CSsR) Cucculelli, M. 3, 11, 107, 193–206 cultural rigidity 115 cultural space 11 Davidsson, P. 196

corporate social responsibility (CSR) 15,

decentralisation 131, 256, 260 decision-making process 180–181, 187 degrees of responsibility/spatiality 243 Dekker, J. 154n2 developing/transitional economies 229 Devinney, T. M. 242 digital-driven disruption 115 digital innovation: descriptive statistics 110, **110**; empirical design 108–110; hypothesis 108; industrial districts 104; Industry 4.0 104; means *vs.*  Wilcoxon rank-sum test **111**; MID 104–108; pairwise Pearson correlation coeffcients **112**; physical and social spatial contexts 104; policy makers 115; regression results **113**; variables **109**  Dijkstra, L. 42 dis-embedding 135–137, 152–153, 160 diversifcation 220–222, *221*  Donckels, R. 10


family-frm territory nexus 78, 142, 151 family leadership 215 family ownership: industry attributes 193; regional entrepreneurship 196–197; social responsibility 134; third generation 162; transition 158 family participation: business diversifcation 221, *222*; corporate culture 171; enterprise diversifcation *222*  family spatialities 15–20, 23 family stakeholders 180–181 family *vs.* non-family frms' strategic effects: academic interest 177; attributes 193; bridging social capital 181–182; contributions 188–189; data analysis 183–184; decision-making process 180–181; family business research 65–66; frm entrepreneurial behaviour 179–180; ft indices 184, **186**; informal entrepreneurial activities 188; Kenya (*see* Kenyan family frms); measures 182–183; participatory and interactive decision-making 187; path and hypothesis analysis and control variables 184, **186**; poverty alleviation 177; productivity 66; regional development 177–179; strategic behaviours 178–179, 186; variables and indicators 184, **185**  Fang, H. "Chevy" 3, 210–232 FA regions 87 farm households 197 Fernandez, Z. 125 fnancial structure 106 Finnish Longitudinal Employer-Employee Data (FLEED) 146 Finnish Longitudinal Owner-Employer-Employee Data (FLOWN) 146 frm labour productivity 87 frm-level analysis 179 frm performance: annual net proft 223, *224*; annual sales 223, *223*; economic opportunities 224; entrepreneurial orientation 180 frm productivity *see* family co-occurrence frms' heterogeneity: economic actors 143; family business heritage 143; hypothesis 145; intensity 144; micro and macro levels 142; similarity 143; socio-emotional perspective 144; solidarity 143–144; sources 142; trustbased and reciprocal relationships 145 fxed-effects (FE) model 90–91

FLEED *see* Finnish Longitudinal Employer-Employee Data (FLEED) FLOWN *see* Finnish Longitudinal Owner-Employer-Employee Data (FLOWN) formal skills: competencies 88; relatedness 89; similarity 88; unrelatedness 89 fourth industrial revolution/Industry 4.0 104, 105, 109, 113–115 Functional Geography of the Firm 8 Galletto, V. 105 Gallo, M. A. 10 GDP *see* gross domestic product (GDP) gender inequality 99 generational confict 228 genetic ability 105 geographical context 86 geographical proximity 68–69, 127, 133, 260 geographic clustering 258 Gierusz, A. 13 globalisation: inequality gap 256; migration patterns 84; ongoing 136 global value chain (GVC) 116 Glückler, J. 128, 129 Golden, M. A. 53 Golikova, V. 12 Gómez-Ansón, S. 2, 65–80 Gordon, I. R. 97 Görmar, F. 3, 157–172; 242 governance: family labor pool 212; interactive and collaborative modes 242; Tatula programme 263 Graffenberger, M. 3, 157–172; 242 gross domestic product (GDP) 83, 140, 210 growth poles theory 258 GVC *see* global value chain (GVC) Hasso, T. 154n2 Henderson, D. A. 9 Hervas-Oliver, J. L. 114, 116n2 "Hidden Champions" 158, 161, 163, 168 Hohn, U. 160 Holm, E. 84, 92 Holmer, A. 163–165 *homo economicus* 15 human resources 163 human society 7 I-district effect 105 I-MID effect 114

impartiality 43

independent productive factor 140 individual-level analysis 179 industrial organization (IO) 195 industry-level variables 194 industry structure 195–196 inequality gap 256 inferential statistical analysis 184 informal family network 98 informational institutions 212 information symmetry 98 innovation 218–220 innovativeness 140 institutional proximity 19, 20 institutional theory 127 institutional ties *see* regional ties institutional transition 211–213, 229 integrative urban development management 248 inter-frm relationships 131, 133 inter-/multinational conglomerates 8 international fnancial crisis 115 internationalization 10, 222–223 inter-organisational linkages 135 interpersonal linkages 135 intra-district heterogeneity 103–105, 114 IO *see* industrial organization (IO) Isard, W. 7 Italian regions: demographic characteristics 51–53; descriptive analysis 48–53; family involvement **53**; frm age categories **52**; frm size categories **51**; industries 48–49; labour productivity *55*; quality of institutions 53–54, *54*; reasons 47–48; regional productivity, exports and innovation 54–57, *55, 56*; sector and geographical area 49, **50**, **51**; strata 48; Xth UniCredit-Capitalia survey 48 Jacobs externalities 68, 79

Kahn, J. A. 9 Kalantaridis, C. 154n2 Keeble, D. 196 Kellermanns, F. W. 109, 183, 197 Kenyan family frms: developing economy 178; estimation 178; "harambee spirit" 178; quantitative analysis 177 kinship/familial relationships 85 Kitzmann, R. 13 Kleine-König, C. 160 Kuznetsov, B. 12

labour productivity *55*, 70, 76, *76*, 78 Lambrecht, J. 10 Lapuente, V. 42 Lattanzi, N. 3, 67, 103–116 Lenz, R. 3, 125–137 Levinsohn, J. 59n6 Lindgren, U. 87, 92 Ljunggren, E. 197 local embeddedness: assumption 126; data and variables 145–147; defnition 141; descriptive statistics, variables 146, **147**; distance to frm 146; economic integration 144; empirical model and results 148, **149**, 150–151; frms' heterogeneity 142–145; locality 152; local production systems 153; place-based approach 261; place tenure 146; social relationships 152; stakeholders 152 local family frms 261, 267 location quotient (LQ) 90 low fertility rates 136 LQ *see* location quotient (LQ) Luckmann, T. 197

macroeconomic approach 257 Malecki, E. 13 management geography research stream 16 Marino, L. 182 Marshallian atmosphere 103 Marshallian Industrial District (MID): defned 103; digital innovation 104–108; family frms 106–108; intra-district heterogeneity 103–105; limitations 115–116; local production systems 114 Marshall's theory 13 Marta Herford project 245 Martyniuk, O. 13 Mason, G. 183 Massis, A. De 85 MAXQDA software 130 McCann, P. 97 McGrath, H. 182 McNee, R. B. 8 Memili, E. 219 MES *see* minimum effcient size (MES) Mezzogiorno intervention 258 MID *see* Marshallian Industrial District (MID) micro-oriented research 193, 197 minimum effcient size (MES) 195 Mohnen, A. 136 Mückenhausen, V. 136

multinational frms 194 Murithi, W. 3, 177–189 Murphy, F. 109 Näder, H. G. 248 Nason, R. 180 national organic certifcation system 264 nepotism 83, 98 Niehm, L. S. 242 Nieto, M. J. 125 Nordqvist, M. 180 North, D. C. 154n1 Oinas, P. 15, 16 OLS *see* ordinary least squares (OLS) OLS pooled regressions 202, **203**, 204, **205**  ongoing globalisation 136 ordinary least squares (OLS) 111 organic farming 266, 267 organisational proximity 19 O'Toole, T. 182 participative decision-making process 177–178, 187 Patuelli, A. 3, 103–116 Peltonen, J. 3, 140–153 Pérez, P. F. 10 Pérez Rodríguez, M. J. 180 Peruzzi, V. 3, 193–206 Petrin, A. 59n6 philanthropy 239, 244 Picci, L. 53 Pike, A. 160 Pini, M. 14 place-based economic engagement 237 place-based approach: economic development 256; local family frms (*see* family frms); mega-urban regions 259; micro-foundation 268, 269; 'one size fts all' approach 260; ready-made universal template 260; space-neutral interventions 259; theoretical and practical contributions 269; universal attributes 260; *see also* Tatula Programme polar types 238, 243 policymaker's perspective 136 private-sector-driven 241 private sector-led master plan 247 productivity *see individual entries*  pairwise Pearson correlation coefficients **74**, **112**

product *vs.* process innovation 111 professionalisation 115, 131 proximity:cognitive 19, 69; dimensions 145; geographical 68–69, 127, 133, 260; institutional 19, 20; organisational 19; social 19, 69, 97 public funds 265, 266 public policy implications 152–153 Pucci, T. 107 purposive sampling 238, 243 random-effects (RE) model 72, 90–91, 148, **149**  Raposo, N. P. 10 R&D *see* research and development (R&D) R&D investment intensity 218–219, *219*  regional business-led community development 240 regional certifcation system 265 regional competitiveness: exporters and innovative behaviour 46–47, *47*; exports 45, *45*; innovation 46, *46*; productivity 44–45 regional entrepreneurship: aims 193; average frm size 200, 204; birth rates 199–200, 202, **203, 205**; business density 200–201, 204; datasets 194, 198–199; dependent variable 199–200; descriptive statistics and correlation matrix 202, **203**; econometric specifcation 201–202; estimation results 202–205; familyowned frms 193; family ownership 196–197; industrial development of European regions 194; industry scale 193–194, 200; industry structure and entrepreneurial activity 195–196; limitations 206; ownership structure of companies 194; testable hypotheses 197–198 regional familiness model 18–20, *20*, 66–70, 78, 125, 239 regional orchestration 250 regional planning policy framework 258 regional ties: dis-embedded family frms 135–137; effects on family frm 131–135; effects on inter-frm relationships 133; family frms and regional context 126–127; family frms and regional distinctiveness 129; interviews conducted **129**; policymakers 136–137; qualitative case study 129–130;

relationships, institutions and time 127–128; successors *vs.* predecessors 131; temporal dynamics 137 relational infrastructure 128 relational space 140 rent seeking 229 research and development (R&D) 218–219 reurbanization-based development schemes 245 reverse causality 150 Reynolds, P. 196 Ricotta, F. 2, 12, 33–58 Robb, A. 197 Robinson, C. 183 Rodriguez-Pose, A. 260 Rojas 114 Samphantharak, K. 12 Sarathy, R. 109 Schamp, E. W. 15 Schiek, M. 160 Schierling 163, 164 Schneider, A. 244 Seaman, C. 11 Sempere Ripoll, F. 114 Sforzi, F. 116n5 Short, J. 179 similarity 88, 89, 143 small and medium-sized enterprises (SMEs) 13, 47, 85, 98, 99, 103, 115 small-town development: from club sponsoring to community building(s) 165–168; dimensions 161; immediate frm environment 158; local contexts and frm embeddedness 161–165; policy-makers 171 SMEs *see* small and medium-sized enterprises (SMEs) social capital 83, 107, 108, 116 social networks 8, 127, 140, 182 social proximity 19, 69, 97 social relations 230 social space 17, 140 social status/reputation 157 socioeconomic development 2, 181 socio-emotional wealth 186, 241 Soleimanof, S. 127 solidarity 143–144 space-based corporate responsibility 242 spaces of familiness 23 Spanish manufacturing frms 125 spatial contexts 8, 10

spatial entities: concepts 11–12; contexts 10; factors 9; policies and planning 12; processes 10; scales 10–11; settings 11; spatial structure and distribution 9–10 spatial familiness 239 spatial family management model 15–18 Spiegel, F. 9, 11 Steier, L. P. 181 stewardship theory 181 Storai, D. 11, 107 Storper, M. 128 Stough, R. 1–4, 15, 44, 56, 66, 150 strategic behaviours 178–179, 186 strategic entrepreneurship theory 179 Stuetzer, M. 196, 198 succession 135, 224–226, *225*  successors *vs.* predecessors 131 Survey on Business Strategies 70 sustainable farming certifcation system 265 sustainable farming promotion: data 262; governance 263; impact 266–267; programme 262–263; programme's main activities 263–266; region 262 Sutton-Brady, C. 182 Suwala, L. 1–4, 7–23, 237–251 Sveen, J. 10 Swedish labour market 97 Swedish Standard Industrial Classifcation 2002 (SNI02) 86 tacit knowledge 107 Tatula Programme, Lithuania: family entrepreneurial orientation 268; family level 267; farmers and customers 265; local community level 267; national level 263, 267; sustainable farming promotion 261–266 Territorial Agenda of the European Union 2020 260 territorial innovation models (TIM) 11, 13, 19, 142 testable hypotheses 197–198 TFP *see* total factor productivity (TFP) TIM *see* territorial innovation models (TIM) time-invariant heterogeneity 71 topdown approach 256, 259 total factor productivity (TFP) *44*, 44–45 town-center management 247 traditional approach 257–259 transgenerational value 180

280 *Index* 

unbalanced distribution 229 urbanisation economies 68, 70, 79 urban management 245 urban revitalization 249 urban spaces 68

variables 87–90

Wagner, M. 134 Walker, S. 196 Welsh, D. H. 219 Wennberg, K. 11 Wenzhou model 13 Western context 177, 188, 228–229 Westhead, P. 9 Wilkie, C. 260

Zahra, S. A. 197 Zellweger, T. 180 Zhang, X. 3, 210–232 Zhou, Y. 13