Productive, Unproductive and Destructive Entrepreneurship: A Theoretical and Empirical Exploration

This is pre- print version of Sauka, A. (2008) Productive, Unproductive and Destructive Entrepreneurship. A Theoretical and Empirical Exploration. Peter Lang. Printed book might in parts contain different content. Please find the book on: https://books.google.lv/books?id=PDoy3svNRAC&printsec=frontcover&hl=lv&source=gbs_ge_summary_r&cad=0#v=onepage &q&f=true https://www.peterlang.com/view/product/61359

Table of Contents Acknowledgements ..................................................................................................... vi List of Figures........................................................................................................... viii List of Tables ................................................................................................................ x Abbreviations ........................................................................................................... xiii 1.

Introduction ........................................................................................................ 1 1.1. Various forms of entrepreneurship: productive, unproductive and destructive .. 1 1.2. Research questions .............................................................................................. 3 1.3. Definition of a key concept: what is entrepreneurship? ...................................... 4 1.4. Entrepreneurship in transition: context matters ................................................. 10 1.5. Outline of thesis and a preview of the following chapters ................................ 13

2. Theoretical exploration of productive, unproductive and destructive entrepreneurship ........................................................................................................ 17 2.1. Productive, unproductive and destructive entrepreneurship: conceptual review and empirical evidence ............................................................................................. 17 2.2. Activities and value creation: a general framework .......................................... 21 2.2.1. Value creation from SMEs activities by different types of firms ................... 21 2.2.2. Output and activities in the context of the assessment of productive, unproductive and destructive entrepreneurship ........................................................ 24 2.3. Conceptualisation of the term ‘activities’ .......................................................... 28 2.4. Antecedents and influences for activities .......................................................... 31 2.5. Influence of environment................................................................................... 33 2.6. The relationship between motivation, activities, environmental influences and output ..................................................................................................................... 37 2.6.1. Conceptualising the general relationship between activities and value creation ..................................................................................................................... 38 2.6.2. Conceptualising the relationship between motivation, activities and value creation ..................................................................................................................... 40

ii

2.6.3. Addressing the complexity of the relationship between activities and SME value creation ............................................................................................................ 41 2.6.4. Conceptualising the relationship between EO, EB and SME value creation ..... ..................................................................................................................... 47 2.7. Framework for the assessment of (directly and indirectly) productive, unproductive and destructive entrepreneurship ........................................................ 49 3.

Towards a measurement of productive, unproductive and destructive

entrepreneurship ........................................................................................................ 52 3.1. Towards a measurement of EB and EO............................................................. 52 3.2. Towards a measurement of SME output on a venture level .............................. 54 3.3. Towards a measurement of SME output on a societal level.............................. 55 3.4. Towards a measurement of activities’ antecedents and influences: expectations and goals ................................................................................................................... 59 3.5. Towards a measurement of environmental influences and access to capital .... 60 4. Methodology ........................................................................................................... 63 4.1. Description of the survey method ...................................................................... 64 4.2. Questionnaire forms and measures .................................................................... 67 4.2.1. Measurement of EB ..................................................................................... 67 4.2.2. Measurement of EO .................................................................................... 69 4.2.3. Measurement of SME output on a venture level ......................................... 72 4.2.4. Measurement of SME output on a societal level......................................... 73 4.2.5. Measurement of environmental influences and access to capital ............... 74 4.2.6. Measurement of expectations and goals, and controls ................................ 76 4.3. Response rate and item non- response ............................................................... 78 4.4. Survey biases ..................................................................................................... 81 4.5. Characteristics and sample representativeness .................................................. 85 4.5.1. Individual level characteristics and sample representativeness .................. 86 4.5.2. Venture-level characteristics and sample representativeness...................... 88 4.6. Approach for the analysis of results .................................................................. 91

iii

5. Empirical exploration of productive, unproductive and destructive entrepreneurship ........................................................................................................ 95 5.1. SME output: venture and societal level ............................................................. 95 5.2. Determinants and influences of productive, unproductive and destructive entrepreneurship ....................................................................................................... 98 5.3. Determinants of productive, unproductive and destructive entrepreneurship: individual indexes of EO and EB ........................................................................... 108 5.4. Determinants of productive, unproductive and destructive entrepreneurship: an exploratory approach .............................................................................................. 116 5.4.1. The relationship between EB, EO and SME value creation ..................... 117 5.4.2. Relationship between the environment, access to capital and activities ......... ................................................................................................................ 128 5.5. Determinants of productive, unproductive and destructive entrepreneurship: joint influences of EO and EB on SME value creation .......................................... 135 5.6. Assessing the relationship as conceptualised by goals and expectations, EB and EO, and SME value creation .................................................................................. 139 6. Conclusions and implications ............................................................................. 142 References ................................................................................................................. 149 Appendix 1. Questionnaire form for the face to face interviews, 2005 ..................... 162 Appendix 2. Questionnaire form for the follow up study, phone interviews, 2006 .. 175 Appendix 3. Selected individual level characteristics ............................................... 184 Appendix 4. Selected venture level characteristics.................................................... 185 Appendix 5. Rent Seeking versus Conforming Behaviour ........................................ 186 Appendix 6. Unofficial Payments versus Conforming Behaviour ............................ 187 Appendix 7. Unethical Behaviour versus Conforming Behaviour ............................ 188 Appendix 8. Means, S.D's, and correlations for quantitative variables (EO) ............ 189 Appendix 9. Means, S.D’s, and correlations for quantitative variables (EB) ............ 190 Appendix 10. Individual dimensions of EO: results from the analysis ..................... 191 Appendix 11. Types of irregular payments versus conforming activity and resulting SMEs output in venture level short and long term .................................................... 192 Appendix 12. Types of irregular payments versus conforming activity and resulting SMEs output in societal level short and long term .................................................... 194

iv

Appendix 13. Perceived environmental influences and involvement in various dimensions of EO....................................................................................................... 196 Appendix 14. Access to capital and involvement in various dimensions of EO ....... 198

v

Acknowledgements According to my experience, the research process is based on three values. First of all it requires ability, willingness, motivation and patience from the authors themselves. Secondly support from the experts in the particular research field as well as family and friends is of major importance. Third, nothing is possible without proper financial support. My doctoral studies have been characterized with lots of uncertainty and risk taking with regard to all of these three determinants. Here, in the very beginning of this doctoral thesis I would like to express my gratitude to all the individuals and organizations that provided me with support, thus making this thesis possible. First of all, my greatest acknowledgements go to the supervisor of my thesis Prof. Dr. Friederike Welter. Your guidance from almost the very beginning of my doctoral studies, patience, insightful comments combined with the support of my freedom in pursuing the topic of my interest has been a great learning experience for me. Not least, generous financial support from the TeliaSonera Institute at the Stockholm School of Economics in Riga is gratefully acknowledged. My acknowledgements also go to the Scandinavian Institute for Administrative Research for providing financial assistance for a research visit (March- May, 2006) as well the Swedish Institute for a scholarship to conduct PhD research work at Jonkoping International Business School, Sweden (2004 - 2005). Furthermore, the author acknowledges financial support from the School of Slavonic and East European Studies, the Foreign and Commonwealth Office and the Ministry of Education and Science of the Republic of Latvia for the possibility to conduct PhD research work at the School of Slavonic and East European Studies/ University College London, UK (2003-2004). I would also like to express my deepest gratitude to all those people from various parts of the world who were with me during the different stages of writing this PhD thesis. Thanks to Dr. Vija Strazdina and Prof. Ilmars Skards for encouraging me to undertake PhD studies; Dr Ruta Aidis for huge support and guidance on how to conduct a literature review and plan the research work during the first year of my PhD thesis; Dr. Martins Kazaks for providing friendly support in the first weeks of my arrival in London; Prof. Per Davidsson for encouraging me to take up the

vi

challenge of exploring this rather complicated topic within the entrepreneurship research domain as well as for very insightful comments on the early versions of this thesis. My thanks also go to Prof. Mirjam Van Praag, Prof. David Audretch and Dr. Mark Sanders for very valuable discussions during the Summer Institute (2005) at the Max Planck Institute of Economics in Jena with regard to the measurement of societal-level performance and asking sensitive questions, Prof. Johan Wiklund for teaching quantitative methods and offering valuable comments regarding the conceptual part of the thesis as well as to Dr. Anders Paalzow, Dr. Roberts Kilis and Dr. Valdis Avotins for the valuable comments on the final draft of my thesis. Thanks also to all staff of University College London, including Dr. Tomasz Mickiewicz, Dr. Slava Radosevic, Prof. Alan Smith, Dr. Piotr Jaworski Dr. Bohumil Fort, Helena Hannula, Reka Rozsavolgyi; Violetta Krakovska, the staff of the Jonkoping International Business School, especially Borje Boers and Anna Jenkins, the staff of the Stockholm School of Economics in Riga, especially Dr. Diana Pauna, Dr. Karlis Kreslins, Dr. Anders Paalzow and Dr. Girts Dimdins as well as doctoral students from the University of Siegen for very valuable comments and friendly support, which was of major importance for improving this PhD thesis. Many thanks to Benjamin Breggin for proofreading, Maija, Ramona, Aleksandrs and Ilze for the great assistance in conducting fieldwork, as well as to the more than 130 owners of SMEs who agreed to participate in the interviews. Finally and most importantly, my greatest acknowledgement goes to my family, especially Liga Beldava and my mother Bronislava Sauka, as well as friends: Kaspars Galkins, Inga Skuja, Laila Eglite, Viesturs Eglitis, Agnese Korbe, Laura Millere, Inguna Draudina, Mareks Ekis, Aiga Vendelina, Artis Seja and Inese Sondore, those who shared both ups and downs in the often turbulent process of writing this PhD thesis. Thank you very, very much!

Arnis Sauka January, 2008 Stockholm School of Economics in Riga, Latvia

vii

List of Figures Figure 1. The entrepreneurial cycle ............................................................................... 8 Figure 2. The outline of the thesis. .............................................................................. 14 Figure 3. New enterprise output on different levels .................................................... 22 Figure 4. Assessment of SME value creation: productive, unproductive or destructive entrepreneurship .................................................................................................... 26 Figure 5. Directly and indirectly productive, unproductive and destructive entrepreneurship .................................................................................................... 28 Figure 6. Model of motivation ..................................................................................... 32 Figure 7. Framework for analysing environmental influences on entrepreneurship processes within a transition setting ...................................................................... 37 Figure 8. General relationship between activities and value creation ......................... 40 Figure 9. Proposed relationship between goals, expectations and EO and EB, affecting SME value creation ............................................................................................... 41 Figure 10. Conceptualising the complexity of the relationship between EO and SME value creation ........................................................................................................ 44 Figure 11. Conceptualising the relationship between EO, SME value creation and environmental influences. ..................................................................................... 47 Figure 12. Proposed relationship between entrepreneurial behaviour, EO, value creation, environmental influences and access to capital...................................... 48 Figure 13. Proposed relationship between EO and EB, affecting SME value creation ............................................................................................................................... 49 Figure 14. The proposed framework for assessing (directly and indirectly) productive, unproductive and destructive entrepreneurship .................................................... 50 Figure 15. Assessment of (directly and indirectly) productive, unproductive and destructive entrepreneurship ................................................................................. 99 Figure 16. Summary of the findings using summated indexes of EO and EB .......... 107 Figure 17. Assessment of (directly and indirectly) productive, unproductive and destructive entrepreneurship: individual indexes ................................................ 109 Figure 18. Summary of the findings using both summated and individual indexes of EO and EB: venture-level, short and long term .................................................. 115 Figure 19. Summary of the findings using both summated and individual indexes of EO and EB: societal level, short and long term .................................................. 116

viii

Figure 20. Summary of the findings using an exploratory approach: activities and resulting output on a venture level, short and long term ..................................... 127 Figure 21. Summary of the findings using exploratory approach: activities and resulting output on a societal level, short and long term..................................... 128 Figure 22. Summary of the findings using summated indexes of EO and EB .......... 135 Figure 23. Assessment of (directly and indirectly) productive, unproductive and destructive entrepreneurship: the joint influence of EO and EB on SME value creation ................................................................................................................ 136 Figure 24. Assessment of (directly and indirectly) productive, unproductive and destructive entrepreneurship: the influence of goals and expectations ............... 139

ix

List of Tables Table 1. Empirical studies on aspects of productive, unproductive and destructive entrepreneurship .................................................................................................... 19 Table 2. Dimensions of Entrepreneurial Orientation ................................................... 31 Table 3. Operationalisation of ‘conforming’ and ‘deviant’ behaviour for measurement within an advanced transition setting .................................................................... 53 Table 4. Operationalisation of venture-level output for measurement ........................ 55 Table 5. Operationalisation of societal-level output for measurement ........................ 59 Table 6. Operationalisation of expectations and goals for measurement .................... 60 Table 7. Classification of barriers to SME development and growth: formal, informal, economic and other environmental influences:..................................................... 61 Table 8. Major barriers specific to the advanced stage of transition ........................... 62 Table 9. Rotated component matrix: assessment of degree of involvement in deviant behaviour ............................................................................................................... 68 Table 10. Measures for assessing conforming and deviant behaviour in a transition setting .................................................................................................................... 69 Table 11. Rotated component matrix: assessment of entrepreneurial orientation ....... 70 Table 12. Measures for assessing EO .......................................................................... 71 Table 13. Measures for assessing venture-level outcomes: short and long term ......... 73 Table 14. Measures for assessing societal-level outcomes: short and long term......... 74 Table 15. Measures for assessing contextual influences ............................................. 75 Table 16. Measures for assessing SMEs’ expectations regarding growth and expansion............................................................................................................... 77 Table 17. Control variables .......................................................................................... 78 Table 18. Assessment of the overall output of SME owners or owner-managers ....... 83 Table 19. Expectations of SME owners or owner-managers regarding growth as reported in the 2005 survey vs. actual growth as reported in the 2006 survey ..... 84 Table 20. SMEs owner-managers’ perceptions about the fulfilment of their growth expectations ........................................................................................................... 85 Table 21. Age profiles of entrepreneurs according to level of education (column %) 86 Table 22. Surveyed SMEs according to main type of activity by type of customers (row %).................................................................................................................. 89 Table 23. Venture level output of surveyed SMEs ...................................................... 96

x

Table 24. Perceived societal-level output of surveyed SMEs...................................... 97 Table 25. SME output on a venture level, short term: universal, contingency and configuration model. ‘EO approach’................................................................... 101 Table 26. SME output on a venture level, long term: universal, contingency and configuration model. ‘EO approach’................................................................... 101 Table 27. SME output on a societal level, short term: universal, contingency and configuration model. ‘EO approach’................................................................... 102 Table 28. SMEs output on a societal level, long term: universal, contingency and configuration model. ‘EO approach’................................................................... 102 Table 29. SME output on a venture level, short term: universal, contingency and configuration models. ‘EB approach’. ................................................................ 104 Table 30. SME output on a venture level, long term: universal, contingency and configuration models. ‘EB approach’. ................................................................ 104 Table 31. SME output on a societal level, short term: universal, contingency and configuration models. ‘EB approach’. ................................................................ 105 Table 32. SME output on a societal level, long term: universal, contingency and configuration models. ‘EB approach’. ................................................................ 105 Table 33. EO vs. SME output on a venture level, short term. ................................... 110 Table 34. EO vs. SME output on a venture level, long term. .................................... 110 Table 35. EO vs. SME output on a societal level, short term. ................................... 111 Table 36. EO vs. SME output on a societal level, long term. .................................... 111 Table 37. EB vs. SME output on a venture level, short term. ................................... 112 Table 38. EB vs. SME output on venture level, long term. ....................................... 112 Table 39. EB vs. SME output on a societal level, short term. ................................... 113 Table 40. EB vs. SME output on a societal level, long term. .................................... 113 Table 41. Involvement in rent seeking vs. conforming behaviour and resulting output on a venture level, short and long term ............................................................... 118 Table 42. Involvement in rent seeking vs. conforming behaviour and resulting output on a societal level, short and long term ............................................................... 119 Table 43. Involvement in ‘irregular, additional payments’ vs. conforming behaviour and resulting output on a venture and societal level, short and long term .......... 120 Table 44. Involvement in unethical behaviour vs. conforming behaviour and resulting output on a venture level, short and long term .................................................... 123

xi

Table 45. Involvement in unethical behaviour vs. conforming behaviour and resulting output on a societal level, short and long term.................................................... 124 Table 46. Involvement in various dimensions of EO and resulting output on a societal level, short and long term .................................................................................... 125 Table 47. Perceived environmental influences and involvement in rent seeking vs. conforming behaviour ......................................................................................... 130 Table 48. Perceived environmental influences and involvement in unethical vs. conforming behaviour ......................................................................................... 130 Table 49. Perceived environmental influences and involvement in ‘irregular, additional payments’ vs. conforming behaviour ................................................. 131 Table 50. Perceived environmental influences and involvement in specific types of irregular payments vs. conforming behaviour .................................................... 132 Table 51. Access to capital and involvement in ‘irregular, additional payments’ vs. conforming behaviour ......................................................................................... 132 Table 52. Access to capital and involvement in rent seeking vs. conforming behaviour ............................................................................................................. 133 Table 53. Access to capital and involvement in unethical vs. conforming behaviour ............................................................................................................................. 134 Table 54. Direct effect of EO on EB.......................................................................... 137 Table 55. EO and EB vs. short-term, venture-level output ........................................ 138 Table 56. EO and EB vs. short-term, societal-level output ....................................... 138 Table 57. EO and EB vs. long-term, venture-level output......................................... 138 Table 58. EO and EB vs. long-term, societal-level output ........................................ 138 Table 59. Direct effect of Goals and Expectations on EB ........................................ 140 Table 60. Direct effect of Goals and Expectations on EO ........................................ 140 Table 61. Goals, expectations and EB, EO vs. venture-level output in the short term ............................................................................................................................. 141 Table 62. Goals, expectations and EB, EO vs. societal-level output in the short term ............................................................................................................................. 141

xii

Abbreviations

EB: entrepreneurial behaviour EBRD: European Bank of Reconstruction and Development EO: entrepreneurial orientation EU: European Union CEE: Central and Eastern Europe CEEB: Central and Eastern Europe, and the Baltic States FSU: Former Soviet Union GEM: Global Entrepreneurship Monitor SD: standard deviation SME: small and medium sized enterprise

xiii

1. Introduction 1.1. Various forms of entrepreneurship: productive, unproductive and destructive According to entrepreneurship literature the nature and extent of entrepreneurial value creation not only affects the sustainability and success of the venture, it also influences its wider contribution to the development of the economy. On a societal or economy level, especially from an Austrian point of view (Mises, 1949; Kirzner, 1973; Foss and Foss, 2002) entrepreneurship and SME activity are generally viewed as contributing positively to economic growth. Although there are also diverse arguments (Brown, Hamilton and Medoff, 1990; Davidsson and Delmar, 2000; Dunne, Roberts and Samuelson 1996), from a macro-economic perspective several authors highlight the innovativeness and the potential to generate new jobs as a major role and particular strength of SMEs (e.g., Birch, 1979; Storey, 1994). On a venture and individual level, the value creation of the SME sector is commonly reflected in the expansion and growth of an enterprise, which involves both a monetary dimension such as the generation of profits, and a variety of sources of satisfaction, including non-pecuniary sources (Welter and Smallbone, 2004). Entrepreneurship, however, can take various forms, and not everything labelled as ‘entrepreneurial’ is actually desirable (Baumol, 1990, 1993; Dallago, 1997). On the one hand, it is often emphasised that the main engine of entrepreneurial activity is profit as well as various forms of self-fulfilment. On the other hand, Baumol (1993) points out that if we define entrepreneurs simply as persons who are innovative in generating profits or in adding to their power and prestige, it cannot be expected that they will be concerned with how much or little the activities employed to achieve these goals will contribute to the net output of economy. In this context, a key question concerns the activities entrepreneurs pursue in order to create value, pointing to the necessity to distinguish between ‘positive’ and ‘negative’ activities and their output, or as argued by Baumol (1990, 1993), productive, unproductive and destructive entrepreneurship. Several authors (e.g. Baumol, 1990, 1993; Dallago, 1997, 2000; Foss and Foss, 2002) have contributed to these concepts. Rent seeking in the form of litigation, lobbying, takeovers, tax evasion and avoidance efforts as well as ‘use of the legal 1

system’; illegal and shadow activities, including drug dealing, prostitution, racketeering, blackmailing; and various forms of corruption are often mentioned among unproductive or destructive entrepreneurship activities. Job generation and innovativeness, if not used for rent seeking purposes, are mainly associated with a ‘productive value’ on societal and economy levels (e.g., Baumol, 1990, 1993; Foss and Foss, 2002; Dallago, 2000). According to entrepreneurship literature, however, there is no consensus on what determines productive, unproductive and destructive entrepreneurship on a conceptual level. The main reason for this is that in reality a few activities among, for example rent seeking, make absolutely no contributions to economy output (Davidsson and Wiklund, 2001; Davidsson, 2004). In general, it can be argued that on a venture level unproductive entrepreneurship may constrain the longer-term growth prospects of enterprises in situations where the development of new markets is dependent on increased legal compliance and legitimacy (Welter and Smallbone, 2004). On an economy level unproductive entrepreneurship may hamper the contribution of entrepreneurship to economic development (Sauka and Welter, 2007a). On the other hand, referring to the link between economic growth and new venture volatility (Reynolds, 1999), Davidsson (2004) suggests that some rent seeking efforts also may have a significant positive impact both on the growth of a company and the economy. For example, tax evasion in the early stages of business activity can help a company to survive competition, whereas at a more mature stage of development the same company may contribute to the growth of the economy. Further, activities which generate unproductive outputs in the short term can also lead to positive outputs in the long term and vice versa, thus time frame should also be considered when SME value creation is assessed (Sauka and Welter, 2007a). In this context, Foss and Foss (2002) argue that it is an important function of contracts and property rights systems to implement optimal combinations between ‘productive’ and ‘destructive’ entrepreneurial activities. As a cut-off point to distinguish between productive and destructive entrepreneurship, Foss and Foss (2002) indicate either an increase in entrepreneurial activity or a decrease in joint surplus. Further, in the framework of institutional theory (North, 1990), actions entrepreneurs take to create value are linked with the influence of institutional settings, e.g. the ‘formal’ and ‘informal’ constraining and enabling forces of individual behaviour, influencing the nature and extent of value creation (Welter and Smallbone, 2004; Dallago, 2000). The features of an economic system, defined as a 2

coordinated set of institutions and structures, determine not only what entrepreneurs can do and find profitable to do (Baumol, 1993; Dallago, 2000) but also influence, the type, features as well as the entrepreneurs’ use of resources. According to Dallago (2000: 305), “Environmental complexity, the variety and variance of the features, perceptions and goals of individual actors, and the latter computational limitations, e.g. bounded rationality, to understand, process, and utilise information about the environment to predict the behaviour of actors jeopardise productive activity.” In this regard, previous research has demonstrated that most new and small companies are actually involved both in productive and rent seeking activities at the same time (e.g., Glinkina, 2003; Los, 1992; Rehn and Taalas, 2004; Scase, 2003; Smallbone and Welter 2001; Aidis, 2006). Davidsson and Henrekson (2002: 1) emphasise that productive entrepreneurship is an “… essential factor of the economic performance of a country.” The aforementioned challenges in the conceptual framework, however, make empirical assessment of value creation rather ambitious. In the context of the above discussion, it is not surprising that little work has been done to empirically assess productive, unproductive and destructive entrepreneurship. This thesis aims to fill this gap. 1.2. Research questions Recent entrepreneurship literature emphasises the necessity of theoretical underpinnings for entrepreneurship and SME research (Davidsson and Wiklund, 2001; Chandler and Lyon, 2001; Ucbasaran et. al., 2001). The determinants of and potential influences on SMEs involvement in productive, unproductive and destructive entrepreneurship in this thesis are derived both from theory (e.g. Baumol, 1990; 1993; Davidsson, 2004; Warren, 2003; Wiklund and Shepherd, 2005) and existing empirical evidence from the studies exploring some important aspects of this issue (e.g. Aidis and Van Praag, 2007; Fadahunsi and Rosa, 2002; Wiklund and Shepherd, 2005; Sobel, 2006). The first section of the thesis highlights the importance of addressing productive, unproductive and destructive entrepreneurship on a conceptual level as well as the necessity of exploring various types of SME value creation empirically. To fill the gap in entrepreneurship literature, the author uses a novel approach to develop a new

3

theoretical framework which allows operationalising the concepts of productive, unproductive and destructive entrepreneurship for empirical assessment. Drawing on this framework, the thesis furthermore explores which activities entrepreneurs pursue in order to create value and what factors influence small-firm value creation leading to productive, unproductive and destructive entrepreneurship, thus making one of the first attempts to assess these concepts empirically. More specifically, the following main research questions are addressed by the thesis:

•

What is meant by productive, unproductive and destructive entrepreneurship on a conceptual level?

•

How

can

the

concepts

of

productive,

unproductive

and

destructive

entrepreneurship be operationalised for empirical assessment? •

What types of activities do small firms pursue in order to create value on different levels?

•

What are the outputs of the different types of entrepreneurial activities?

•

What are the main factors influencing SMEs involvement in productive, unproductive and destructive entrepreneurship? By exploring different forms of entrepreneurship, the answers to these questions

imply an answer to the underlying composite question: what internal and external factors determine and affect SMEs involvement in productive, unproductive or destructive entrepreneurship. As already highlighted in the first section of the thesis, no consensus on what determines productive, unproductive and destructive entrepreneurship on the conceptual level exists in the literature. Moreover, there is still no agreement on the definition of entrepreneurship as such (e.g. Low, 2001; Shane and Venkataraman, 2000; Davidsson, 2004). Before addressing the research questions, the key concept, e.g. entrepreneurship, should be assessed. Drawing on the entrepreneurship definition as introduced by Kirzner (1973) and some recent trends in the definition of entrepreneurship (e.g. Davidsson, 2004), the aim of the next section is to introduce a working definition of entrepreneurship to be used for the scope of this thesis. 1.3. Definition of a key concept: what is entrepreneurship?

4

As Per Davidsson put it, “Researching entrepreneurship is fun, fascinating, frustrating and- important, if you ask me. One of the fascinations is the richness of the phenomena… .” (2004: 1). As pointed out by Davidsson himself as well as number of other entrepreneurship scholars (e.g., Shane and Venkataraman, 2000; Zahra and Dess, 2001; Low, 2001), however, although the definition of the term entrepreneurship and the identification of an entrepreneur’s functions has been the subject of active debate for the past 200 years, there is still no consensus on what entrepreneurship is. For a field of social science like entrepreneurship to have usefulness, it must have a conceptual framework that explains and predicts a set of empirical phenomena not explained or predicted by conceptual frameworks already in existence in other fields (Shane and Venkataraman, 2001). The multidimensional nature of entrepreneurship seems to be a particular problem in this regard when it comes to the definition of the field. Various components of entrepreneurship pertain to economics, psychology and other areas of research. In this context, it is often argued that entrepreneurship as a term needs restriction. While distinguishing between entrepreneurship as a research domain and entrepreneurship as a phenomenon, Davidsson (2004: 7), for example, suggests limiting the concept to the market and market- like contexts, “… that is when the setting involves customers, suppliers and (potential) competitors or very close equivalents to those.” Further, the start-up of new firms as a ‘restriction’ for the field of entrepreneurship has been actively supported by a number of entrepreneurship scholars (e.g. Sexton and Landstrom, 2000; Davidsson, 2004). According to the author’s opinion, however, if the understanding of entrepreneurship is limited, for example, to the creation of a new company (e.g. Low and MacMillan, 1998), various types of start-ups should be considered. For instance, there are owners who start their business ‘from scratch’, whereas others use capital generated in businesses they already own in order to start a new company. Entrepreneurs can also buy a firm which already exists or become involved in an established business with the aim of restructuring it (re-branding, etc.). Furthermore, some become involved in business as venture capitalists or business angels, and can still be considered entrepreneurs as they own a considerable part of the firms and are most often involved in the management of the companies at the same time. These are only a few examples.

5

Another trend in recent entrepreneurship research, however, emphasises the need to consider the multidimensional nature of the entrepreneurship phenomena (e.g. Zahra, Karutko and Jennings, 1999). In the scope of this discussion, one can argue, that the sun, for example, as a phenomenon is always the sun, nevertheless we can do research on the sunrise, sunset and all processes in between in different environmental settings. By focusing only on a few aspects of the phenomena, arguing that these have never been fully captured by any other disciplines, we run the risk of losing the beauty of the phenomenon itself. This example can be easily attributed to entrepreneurship. We know from ‘real-life’ examples that entrepreneurship as a process can be very complicated and often difficult to explain or measure. A start-up can thus be somewhere in between the ‘sunrise’ and the ‘sunset’. Knowing the multidimensionality of entrepreneurship, in this light, it could be worth considering suggestions in entrepreneurship literature (e.g. Gartner, 2001) that entrepreneurship research might be better served by the creation of several scholarly communitiesdomains, focused on more tightly defined subject areas. The danger with definitions of different entrepreneurship domains in this regard, however, in the author’s opinion, would be the difficulty of finding more or less clear-cut points in between these domains, as most, if not all of them, will overlap. Although differences in emphasis exist, followers of both trends seem to agree on the necessity of a distinctive domain for entrepreneurship research and argue that it should be concerned with the discovery, creation, and exploitation of opportunities in order to generate goods and services, new economic activity, and/or new organisations, resulting in changes in the market place (Busenitz et al, 2003; Gartner, 2001; Shane and Venkataraman, 2000; Venkataraman, 1997; Davidsson, 2004). Despite this emerging common trend, so far researchers have limited their attention mostly to the Kirzenian discovery (e.g. Shane, 2000) and Shumpeterian exploitation of opportunities. Little, however, has been done to explain how or why entrepreneurship leads to economic growth (e.g. Audretch, 2002). In light of this discussion, Kirzner (1973: 19- 20), for instance, proposes that “… a fruitful way to define entrepreneurship is the notion in Austrian economics that entrepreneurship consists of the competitive behaviours that drive the market process.” Referring to this quote, Davidsson (2004) argues that this definition is favoured because of two reasons. First, “It puts entrepreneurship squarely in a market context and makes clear that it is the suppliers who exercise entrepreneurship- not 6

customers, legislators, or natural forces that also affect outcomes in the market.” (Davidsson, 2004: 6). Secondly, it is based on both behaviour and output. In terms of output, recent entrepreneurship literature emphasises the necessity to link both micro levels and aggregate levels when entrepreneurship is analysed (Davidsson and Wiklund, 2001; Smallbone and Welter, 2006). Micro level studies are mostly focused on individuals and/ or firms, whereas aggregate levels could be viewed as analyses of the contribution of entrepreneurship to society, either on industrial, national or regional levels (Davidsson and Wiklund, 2001). In this light, as emphasised already by Low and MacMillan (1998), entrepreneurship researchers should acknowledge the necessity of conducting studies on multiple levels of analysis, since these complement each other. Considering the aforementioned arguments that time frame should be considered when entrepreneurship processes are analysed (e.g. Sauka and Welter, 2007a), this thesis thus focuses on entrepreneurship output on both micro and aggregate levels, short and long term, altogether referring to this as companies’ ‘value creation’1. Following these arguments, in the scope of this thesis, entrepreneurship is observed as multi-conceptual phenomena. Further, drawing on Kirzner (1973), and Davidsson and Wiklund (2001) in order to assess productive, unproductive and destructive entrepreneurship, both activities and output on different levels and time perspectives are considered2. However, recent entrepreneurship literature, especially articles focusing on the determinants of the start-up process, emphasises the role of processes which happen before involvement in entrepreneurial activities (e.g. Delmar and Davidsson, 2000; Carter, Gartner and Reynolds, 1996). In this light, there are rather diverse arguments in entrepreneurship literature on whether entrepreneurs create opportunities or whether opportunities already exist, for instance (e.g. Venkataraman, 1997; Shane and Venkataraman, 2000). In the author’s opinion, the ability to take advantage of an opportunity is the issue deserving more attention. Altogether, in the context of the discussion in entrepreneurship literature, one can refer to processes prior involvement in business, various activities and outcomes created by activities like the ‘entrepreneurial cycle’ (see Figure 1). To simplify, the 1

If not stated otherwise, the term ‘value creation’ is further used in this thesis to refer to firm output on both micro and aggregate levels, short and long term. 2

See section 2.3 for further discussion on the relationship between activities and output in the context of assessing SME value creation.

7

‘entrepreneurial cycle’ usually starts with opportunity creation or recognition and more importantly the willingness and ability to take advantage of it. The next phase involves various types of activities, such as capital and skills accumulation, effective allocation of resources through various strategies, etc. Activities also create output, on both micro and aggregate levels, short and long term, leaving visible changes in the market place (Davidsson, 2004). Figure 1. The entrepreneurial cycle Processes prior to involvement in entrepreneurship activities

Activities

Value creation: output on micro and aggregate levels, short and long term

Source: Author, based on Davidsson (2004) This relationship between processes prior to entering business activity, activities and output, however, can be ‘business-stage’ specific and not always linear by nature. In this context, during the start-up stage, acquiring start-up capital and finding the right target group could be a major challenge. Whereas in case the existing company is bought a bulk of other processes might need to be implemented in order to develop a company further. In this light, when it comes to the generation or identification and implementation of opportunities during various stages of the entrepreneurial cycle, it is important to emphasise that both individuals, including self-employed persons and groups of individuals can be the players in the game. Further, a group of individuals can also consist of a ‘group of set of individuals’, such as two or three companies as starters of a new firm. Regardless of the business stage, however, entrepreneurs face risk, uncertainty and usually also competition and they are involved in often very complicated decision-making processes. Innovativeness most probably can help them to acquire a more competitive edge in such situations, nevertheless, according to the author’s opinion, it should not be regarded as a major factor for entrepreneurship to take place. This notion is especially relevant for the case of SMEs, as the vast majority of small and medium firms do not create innovations (Birch, 1979) but are still considered as a driving force for the economy. In this regard, we should, however, be aware that

8

although SMEs in high-tech and skill-intensive industries are mostly considered to be the main sources of innovations (e.g. Audretsch, 2002), this might be partly due to the definition issue and problems associated with the measurement of the innovations as such. These assumptions mostly refer to the introduction of new products, whereas the small business role in innovations will be more important if innovations refer to processes. Further, and not least important, the influence of context should be considered when entrepreneurship processes are assessed3 (North, 1990; Karlsson and Dahlberg, 2003). Taking into the consideration the above-mentioned assumptions, and building on Kirzner (1973) and Davidsson (2004), the following working definition of entrepreneurship is proposed for the scope of this thesis: Influenced by the context, entrepreneurship is the ability to take advantage of an opportunity in order to achieve monetary and non-monetary goals, resulting in activities that lead to productive, unproductive or destructive value creation. In the context of the discussion, here, ‘the ability to take advantage of an opportunity’ refers to the outcomes of processes prior to involvement in business activities during various business stages. ‘Activities’ or more specifically ‘processes arising from involvement in entrepreneurship’ take place if an individual or group of individuals are both willing and able to take advantage of the business opportunities. In a way, this can be seen as the bridge separating non-entrepreneurs, including potential entrepreneurs, and entrepreneurs who have started their business, regardless of the start- up stage they are in. Activities are driven both by monetary and/or nonmonetary goals. Further, activities create visible changes in the marketplace in the form of productive, unproductive or destructive value creation. It is furthermore of importance to consider that there are number of arguments in entrepreneurship

literature

stating

that,

although

very

overlapping

sets,

entrepreneurship and SMEs are often linked together in ‘a very loose fashion’ (Storey and Sykes, 1996). One of the main explanations in this regard is that SMEs cannot be viewed as downscaled versions of big companies. A different form of involvement in business in terms of management, the availability of resources and the ability to 3

See section 1.4 for further discussion on the influence of context on entrepreneurial processes.

9

influence changes in the marketplace are often mentioned among the most important differences (Burns, 2001). In light of this discussion, since no commonly accepted definition of what entrepreneurship is exists, SMEs altogether or micro, small and medium-sized companies separately as well as big companies could thus be regarded as one of the frameworks, or putting it another way- levels/ units of analysis, in which entrepreneurship takes place. Referring to the working definition of entrepreneurship as introduced previously in this section, this thesis explores entrepreneurship on the level of SMEs, considering its importance in terms of innovativeness, employment generation and fostering economic growth in general. Moreover, Wennekers and Thurik (1999:29) also note that “small firms are the vehicle in which entrepreneurship thrives,” providing support for this choice. 1.4. Entrepreneurship in transition: context matters The context or environment in which firms operate, transforming the personal skills, qualities and ambitions of individuals in outputs, have been explored in literature as the framework within which entrepreneurial processes take place, thus highlighting the importance of environmental influence on the entrepreneurial processes as such (e.g. Smallbone and Welter, 2006). When it comes to conceptually and empirically addressing various stages in the above-mentioned entrepreneurial cycle, contextual influences should thus be considered, as they shape not only the role of entrepreneurship and small firms but also their structure and performance (Karlsson & Dahlberg, 2003). In light of this discussion, according to entrepreneurship literature the transition countries of Central and Eastern Europe (CEE) and the former Soviet Union (FSU) seem particularly suited for a study on the influence of context on various entrepreneurship processes (e.g. Aidis, 2006; Aidis and Sauka, 2005; Smallbone and Welter, 2001). Transition per se denotes the process of change from a centrally planned system to a well functioning market-oriented system (EBRD, 1994)4. According to the

4

Though the answer to this question is beyond the scope of this thesis, it is perhaps of interest to mention that no single indicator or definition currently exists that accurately describes the end of transition. A number of authors have suggested that the end of transition is achieved by reaching the level of an ‘advanced market economy’. Unfortunately no generally acceptable definition for what precisely characterizes an ‘advanced market economy’ exists. For further discussion, see Brown (1999).

10

definition introduced by the European Bank of Reconstruction and Development (EBRD, 1994), transition is about institutional change, involving not only the advance of the private sector but also a fundamental transformation of the role of the state, in particular in the economic, financial and legal institutions underpinning the market economy. It is the institutional arrangements for the allocation and generation of goods and resources, and the ownership incentive and rewards structures that institutions embody, that characterise the differences between a market and a command economy. It may also be regarded as an ultimate objective in itself as well as an end in itself (EBRD, 1994:3). Of specific interest to the thesis is the effect of context on the development of a productive SME sector, which under the central planning system was severely restricted5. According to entrepreneurship literature, although not unique, entrepreneurship in a transition setting has specific features as compared with more advanced market economies (Smallbone and Welter, 2001, 2006; Aidis, 2006; Dallago, 1997; Scase, 2003; Chilosi, 2001). As summarised by Sauka (2005), the following main characteristics of entrepreneurship and entrepreneurs specific to a transition context can be identified from the previous empirical studies on entrepreneurship within a transition setting: the important role of an informal network (blats), short-term ‘get rich’ strategies, a negative general attitude of government and society towards entrepreneurship, previous employment in related sectors before transition, part-time vs. full-time involvement in business, lack of knowledge and previous entrepreneurship experience, a relatively high level of education, and a comparably high use of personal savings to start up a company. Furthermore, the special role of SMEs in a transition context are often highlighted in entrepreneurship literature (e.g. Smallbone and Welter, 2001; Dallago, 2000; Aidis, 2006). The ability to provide economic benefits beyond the boundaries of the individual enterprise in terms of experimentation, learning and adaptability is one of them. Apart from this, research in transition countries also shows that even if SMEs do not generate net new jobs, they reduce the erosion of human capital by providing alternative employment opportunities for relatively skilled yet unemployed workers (EBRD, 1995).

5

As Earle and Sakova have noted: “It is difficult to imagine a regime more hostile towards entrepreneurship than the centrally planned economies of Eastern Europe” (2001: 6).

11

While emphasizing the role of the external environment on entrepreneurial processes as such, existing empirical evidence also suggests that differences in the features of entrepreneurship in transition countries at various stages of development exist (e.g. Kolodko, 1999; Smallbone and Welter, 2001). According to the literature, transition economies cannot be treated as a unified group since various important differences exist among them as a result of different starting points, previous traditions as well as willingness and ability to successfully implement reforms (Smallbone and Welter, 2001; Aidis and Sauka, 2005). It has been highlighted in studies on entrepreneurship in transition countries that while assessing the distinctiveness of entrepreneurship in a transition context, it is therefore important to distinguish between countries at different stages of market reform (Kolodko, 1999; Smallbone and Welter, 2001; Aidis and Sauka, 2005). To date, though there is no consensus on what constitutes ‘transition stages’, researchers have developed different categorisations of the transition process. For example, Campos and Coricelli (2002) created seven stylised facts describing the transition process6. Though insightful, the stylised facts proposed by Campos and Coricelli are not useful from an entrepreneurship development point of view (e.g. Aidis and Sauka, 2005). More suitable for this purpose seem to be the transition indicators developed by the European Bank of Reconstruction and Development (EBRD) that plot the progression of economic transition according to macroeconomic as well as institutional variables. These indicators have been systematically compiled since 1989 for countries undergoing economic transition7. Authors such as Smallbone and Welter (2001) have used selected EBRD indicators in order to distinguish between transition countries where market reforms have been slow or not properly installed and countries where they are more advanced. In this light, research suggests broadly distinguishing between Central and Eastern European and Baltic (CEEB) countries and the countries from the former Soviet Union (excluding the Baltic States). Transition country research mostly refers to the first group of countries as advanced-stage transition economies, whilst the second

6

The Seven stylised facts proposed by Campos and Coricelli (2002) describe the main characteristics of the transition process. They are: 1) Output fell, 2) Capital shrank, 3) Labor moved, 4) Trade reoriented, 5) Structure changed, 6) Institutions collapsed, and 7) Costs were high. 7

For further information regarding EBRD transition indicators see Appendix 7 in Aidis and Sauka (2005).

12

group is mainly considered to be in earlier stages (Kolodko, 1999; Smallbone and Welter, 2001). Such a classification between CEEB and FSU countries is justified based on economic and institutional performance differentials. In addition, the influence of EU pre-accession as well as the accession of CEEB countries has been the main driving force behind institutional reform (Smallbone and Welter, 2006). This resulted in a combination of incentives as well as pressure for reform, which was nonexistent for the FSU countries. It is however, important to note that transition is not a linear process. Even countries in more advanced stages can regress to less advanced transition stages based on their economic and institutional performance. The willingness of the government to implement reform coupled with society’s willingness to accept change continues to play a critical role in all stages of the transition process (Aidis and Sauka, 2005). Empirically this thesis draws on the interviews with SMEs8 conducted in Latvia during 2005 and 2006. As argued by Aidis and Sauka (2006: 3): “Latvia provides an excellent example of a transition country that has successfully transformed its status from a centrally planned Soviet republic to a fast-growing, sovereign, market-oriented and democratic state. Latvia’s ability to implement the institutional and economic requirements for a market-based economy as well as its recent successful accession into the European Union positions it in the advanced stages of transition.” Though it is often argued that SME development is especially crucial for the early phases of transition (EBRD, 1995; Smallbone and Welter, 2001), it is, in fact, just as important for the advanced stages of post-transition. As Porter (1990) has pointed out, invention and entrepreneurship are at the heart of national advantage and country competitiveness. Considering the influence of the specific context, on an empirical level the determinants of and influences on SME involvement in productive, unproductive and destructive entrepreneurship are addressed in an advanced transition setting in the scope of the thesis. 1.5. Outline of thesis and a preview of the following chapters In order to address the determinants of and influences on productive, unproductive and destructive entrepreneurship within an advanced transition setting, 8

Up to 250 employees.

13

the following thesis structure has been developed (see Figure 2). Followed by the introductory chapter, the second chapter deals with the development of a theoretical framework to be used for the scope of this thesis. In general, the second chapter aims to address the following questions: ‘What are the main concepts explored in the thesis?’ and ‘How is the relationship between these concepts conceptualised for the framework of this thesis?’ Figure 2. The outline of the thesis.

Chapter 1. Introduction

Chapter 2. Theoretical exploration of productive, unproductive and destructive entrepreneurship

Chapter 3. Towards a measurement of productive, unproductive and destructive entrepreneurship

Chapter 4. Methodology

Chapter 5. Empirical exploration of productive, unproductive and destructive entrepreneurship

Chapter 6. Conclusions and implications

Source: Author

14

The chapter starts with a review of literature on the concepts of productive, unproductive and destructive entrepreneurship (e.g. Baumol, 1990, 1993; Foss and Foss, 2002) and empirical studies capturing some important aspects of the issue (e.g. Aidis and Van de Mortel, 2006; Earle and Sakova, 2001; Sobel, 2006). A further conceptual framework by Sauka and Welter (2007a) is introduced and extended in order to provide a background for the empirical assessment of productive, unproductive and destructive entrepreneurship. When extended, the framework consists of seven theoretical constructs: SME output on micro and aggregate levels in various time frames, entrepreneurial behaviour and entrepreneurial orientation, goals and expectations as well as environmental influences on entrepreneurship processes. Drawing on this, the Chapter 2 continues with a review of previous empirical findings on the relationship between these concepts. This in turn results in the development of hypotheses and the introduction of a new theoretical framework for the assessment of productive, unproductive and destructive entrepreneurship. Although addressed on a conceptual level, each of these concepts can still be measured using a number of different indicators. This is the issue the third chapter deals with. In general, this chapter can be seen as a link between theory and empirical study. It starts by addressing the following question: ‘How can each of the concepts discussed in the second chapter be operationalised for the measurement?’. This, in turn, is necessary in order to introduce measurement scales, which is the main task of this chapter. In order to address this issue, measures validated by existing empirical studies are extracted from literature. Furthermore, new measurement scales are developed, for example, to empirically capture entrepreneurial behaviour and SME output on an aggregate level. Chapter 4 deals with methodology, revealing most of the empirical choices made for the thesis. The chapter starts with a detailed description of the survey design, sample characteristics and representativeness as well as the data collection process. Additionally non-responses, validity issues and biases are discussed in this chapter. Chapter 5 continues by introducing the main empirical findings, representing one of the first attempts to address productive, unproductive and destructive entrepreneurship empirically. A hierarchical linear regression analysis is used to assess the determinants of and influences on SME involvement in productive, unproductive and destructive entrepreneurship within an advanced transition setting and to test whether 15

a universal, contingency or configurational model best fits the data. In addition, exploratory analysis is implemented to provide a more complete picture of the issue. The thesis concludes with Chapter 6, where among other issues, the main findings, contributions, and shortcomings of the study as well as implications and suggestions for further research on the topic are covered.

16

2. Theoretical exploration of productive, unproductive and destructive entrepreneurship 2.1. Productive, unproductive and destructive entrepreneurship: conceptual review and empirical evidence Analysing the determinants which influence the allocation of entrepreneurial inputs and the flow of entrepreneurial talent, and drawing on examples from economic history, Baumol (1990, 1993) distinguishes between productive, unproductive and destructive entrepreneurship activities. Productive entrepreneurship “… refers, simply, to any activity that contributes directly or indirectly to net output of the economy or to the capacity to produce additional output” (Baumol, 1993, p. 30). For example, innovation can be perceived as a productive contribution from entrepreneurs, financial activities which facilitate production, or any activities which contribute to producing goods and services (Baumol, 1993). Foss and Foss (2002) add to this by introducing the element of new discovery, referring to ‘productive entrepreneurship’ as the discovery of new attributes, opportunities, procedures and the like, where the discovery leads to an increase in joint surplus. In this light, research demonstrates that in a transition context productive entrepreneurship cannot be taken for granted (for example, Aidis, 2006; Kuznetsov et al. 2000; Manolova and Yan, 2002; Peng, 2000; Smallbone and Welter, 2001; Van de Mortel, 2002; Yan and Manolova, 1998). Aidis (2006), for example, distinguishes between two major aspects influencing decision-making: environmental factors (such as micro and macro environment and the role of government) and personal factors (such as norms, values and individual characteristics and skills). All factors together influence an individual’s decision-making and awareness regarding the selection between productive or unproductive entrepreneurship (Aidis, 2006). A key idea in defining unproductive and destructive entrepreneurship is that not everything that is entrepreneurial is necessarily desirable. Often, an entrepreneur makes no productive contribution to the real output of an economy, and in some cases even plays a destructive role (Baumol, 1990). This happens when the structure of payoffs in an economy is such as to render unproductive activities such as rent seeking more profitable than productive activities (Baumol, 1993). In this light, Baumol argues that the allocation of resources to either productive or unproductive 17

use varies across societies. Weak and unstable formal institutions (Baumol, 1990) as well as norms and societal values (Welter and Smallbone, 2004) might foster unproductive entrepreneurship. Dallago (2000), however, emphasises the importance of social capital in order to produce a social basis for trust, reputation and relational contracts, pointing out that this does not suffice to explain the allocation of these factors to productive use, but that it is essential to have a ‘proper economic system’. Only a few studies, however, have attempted to empirically address Baumol’s concepts of productive, unproductive and destructive entrepreneurship from such a broad perspective. Sobel (2006), for example, provides some interesting findings on the relationship between ‘productive entrepreneurship’ (measured as venture capital investments or the growth rate of self employment activity in a country), unproductive entrepreneurship (political and lobbying organizations, legal and liability system fairness) and the economic growth of a country. Further key empirical studies capturing some important aspects of productive, unproductive and destructive entrepreneurship are summarised in Table 1. Unproductive and destructive entrepreneurship can take many forms (Baumol, 1990, 1993; Dallago, 1997, 2000). These include, but are not limited to, rent seeking, illegal activities and shadow activities, and different forms of corruption. Rent seeking is a subgroup of the set of unproductive entrepreneurial activities, a concept contributed by Gordon Tullock (1967). It mainly refers to activities whose objective is the “… acquisition of some of the monopoly profit or the economic rents currently generated or potentially available in the economy” (Dallago, 1997: 104). Baumol (1993: 51) refers to rent seeking as the “expenditure of resources in (deliberate) pursuit of economic rents by means that do not (automatically) contravene the accepted rules of society”. Various forms and types of rent seeking can be distinguished such as litigation, takeovers, tax evasion and avoidance efforts or acquiring a monopoly as well as a different use of the legal system; and rent-seeking seems to constitute the prime threat to productive entrepreneurship (Baumol, 1993). Existing empirical studies addressing this issue (e.g. Smallbone and Welter, 2001; Angeletos and Kollintzas, 2000; Shleifer and Vishny, 1991) mainly link these types of rent seeking behaviour to a country’s economic growth. Moreover, according to entrepreneurship literature, rent seeking activity can also be innovative (Dallago 1997, 2000; Baumol 1990, 1993). Examples might be the discovery of a previously unused legal gambit that is effective in diverting rents to those who are first exploring it or 18

‘smart’ speculative financial transactions. Such activity, however, does not contribute much, if any, value to the productive capacity of an economy (Baumol, 1990).

Table 1. Empirical studies on aspects of productive, unproductive and destructive entrepreneurship

Study Aidis and Van Praag (2007)

Earle and Sakova (2000 and 2001) Fadahunsi and Rosa (2002) Fairlie (2002) Morris, Jones and Nel (1997)

Ritter (1998) Johnson, Kaufmann and Shliefer (1997) Shleifer and Vishny (1991) Angeletos and Kollintzas (2000) Murphy, Shleifer and Vishny (1993) Smallbone and Welter (2001) Sobel (2006)

Construct label for assessment aspects of productive, unproductive Explored relationship or destructive entrepreneurship Legal vs. illegal The influence on business performance of entrepreneurship activities. illegal entrepreneurship experience obtained in a centrally planned economy and motivation in a transition context (same data as Aidis, 2006). Legal vs. illegal The effect of illegal entrepreneurship entrepreneurship activities. experience on legal entrepreneurship outcomes in a transition context. Formal vs. informal and legal Discrepancy between law and informal vs. illegal entrepreneurship entrepreneurship activities in a developing activities. country. Legal vs. illegal The effect of illegal entrepreneurship entrepreneurship activities. experience on legal entrepreneurship outcomes in a transition context. Illegal vs. legal and formal Entrepreneurs within an informal sector, vs. informal entrepreneurship their characteristics, motivation, activities. entrepreneurial capability vs. performance of firms and contribution to economic development in a developing country. Informal (illegal) and formal Shift from informal to formal (illegal) entrepreneurship entrepreneurship and economic growth. Unofficial and official Unofficial entrepreneurship, government entrepreneurship policy and economic growth in a transition setting. Rent seeking behaviour. Explores the relationship between rent seeking and economic growth. Rent seeking behaviour. Strategic interactions among potentially rent- seeking entrepreneurs and the role of institutions affecting aggregate economic outcomes. Rent seeking behaviour. Rent seeking and economic growth. Rent seeking behaviour.

Links various forms of rent seeking and characteristics with the contribution of entrepreneurship to economic development in a transition setting. Productive entrepreneurship Links productive or unproductive and unproductive entrepreneurship with institutional quality entrepreneurship activities. with the growth of the state.

Source: Author 19

Illegal entrepreneurship activities and involvement in and ‘informal’ economy are mostly associated with activities such as the production and distribution of illegal drugs, racketeering and blackmail. In this context, productive activity is seen as a legal, registered business. Although likely to be profitable, illegal or informal types of entrepreneurial behaviour are seen as unproductive because little, if any, value is added to the economy and society (Baumol, 1993). Moreover, these activities contravene the legal and normative framework of a society. According to entrepreneurship literature, illegal entrepreneurial behaviour will have a destructive role in an economy in the cases when these activities attract followers (Baumol, 1990). In light of this discussion, entrepreneurs’ previous illegal experience of and the probability that these entrepreneurs will start a legal business (e.g. Earle and Sakova, 2001; Fairlie, 2002) or the link between institutions and illegal entrepreneurship (e.g. Fadahunsi and Rosa, 2002) are the issues captured by most of the empirical studies in this area. Only a few studies (e.g. Aidis and Van Praag, 2007) have addressed the effect of illegal entrepreneurship on the performance of firms. Further, in the definition of unproductive activity, Baumol (1993) also includes undertakings related to production in all cases of governmental intervention in the productive process- regulation. Examples include the granting of exclusive licences, or the enactment of laws by which the productive process is affected directly or through litigation. Overall, however, although there appears to be no lack of suggestions, no consensus on the question of which activities can actually be regarded as productive, unproductive or destructive exists (Sauka and Welter, 2007a). The key challenge here is that in practice there are only a few genuine ‘unproductive entrepreneurship activities’ (Davidsson and Wiklund 2001; Davidsson 2004). As argued by Baumol himself (1993) activities tainted by, for example, rent seeking, cannot in every case be regarded as unproductive. Moreover, in relation to a transition context, several empirical studies show that legal and illegal activities coexist and most new and small firms are actually involved both in productive and rent seeking activities at the same time (e.g. Glinkina, 2003; Los, 1992; Rehn and Taalas, 2004; Scase, 2003; Smallbone and Welter, 2001). In this light, especially in early transition conditions when the legislation and rules are not yet in place, rent seeking activities such as tax avoidance can well be necessary to ensure the survival and growth of the 20

enterprise, thus making at least some contributions to economic development (Smallbone and Welter, 2006). 2.2. Activities and value creation: a general framework 2.2.1. Value creation from SMEs activities by different types of firms Analysing the link between activities and SME output, Davidsson and Wiklund (2001) draw attention to the output of legal, illegal, informal, formal, rent seeking and other types of activities emphasized by empirical studies, distinguishing between output on venture and societal levels. According to Davidsson and Wiklund (2001), output from these activities results in four types of enterprises (Davidsson and Wiklund, 2001: Davidsson, 2004): ’hero’ or ’success enterprises’, ’robber’ or ’redistributive enterprises’, ’catalyst’ and ’failed enterprises’ (Figure 3). The authors argue that entrepreneurship researchers often seem to oversimplify with the assumption that micro- level outcomes can be directly translated to an aggregate level (Baumol, 1990; Davidsson and Wiklund, 2001). Davidsson and Wiklund (2001: 91) emphasise that ”it is fully conceivable that a successful new enterprise at the micro level translates into economic regress at the societal level and that failed entrepreneurship at the micro level contributes to economic development.” A more relevant measurement as to the relative output of firms in entrepreneurship studies is therefore the obsolete performance of ventures on micro and aggregate levels (Venkataraman, 1997). Davidsson and Wiklund (2001) also emphasize that most existing studies tend to regard a new enterprise either as a ‘hero’ or as the ‘failed’ type. However, as they further argue, “There are reasons to believe that neither robber nor catalyst enterprises are marginal phenomena that could be disregarded.” (Davidsson and Wiklund, 2001: 91), suggesting that, for example, a catalyst enterprise may have a significant impact on the economy. According to Davidsson and Wiklund (2001), hero enterprises (quadrant I, Figure 3) are those creating value for society through the introduction of new combinations while simultaneously creating personal wealth. This type of entrepreneur benefits both from making profits and/or achieving personal goals as well as adding value to society. In the context of Baumol’s concept, they are ’analytically unproblematic’ and basically form the main part of productive entrepreneurship (Davidsson, 2004). As exemplified by cases from ‘real economies’

21

(Davidsson, 2004; Smallbone and Welter, 2001), however, it should be noted that even in such cases, some rent seeking activities, for example, can be present. In some contexts, such activities can be ‘excused’ by an unstable institutional framework (e.g. Aidis and Van Praag, 2007). In general, however, one can argue that it is part of entrepreneurs’ nature to seek to exploit opportunities even if such an activity is sometimes not ethical, e.g. in conformity with the norms, values and ‘unwritten laws’ generally accepted by society. Or, on the other hand, such activities may even be illegal, representing cases when companies use gaps in the legislation or cheating, for example importing products such as T shirts with a ‘sports wear’ customs code in order to avoid paying a higher import percentage rate, to conduct one business activity or another. Lobbying to obtain exclusive rights to sell products or services, using informal networks to receive orders from customers can be mentioned as a couple of other examples in this regard (Sauka and Welter, 2007a). Figure 3. New enterprise output on different levels

Venture level outcome

+

-

Societal level outcome

+

-

Quadrant I

Quadrant II

Hero or success enterprises

Catalyst enterprises

Quadrant III

Quadrant IV

Robber or redistributive enterprises

Failed enterprises

Source: Davidsson and Wiklund (2001: 91) and Davidsson (2004: 13) The robber enterprise (quadrant III, Figure 3) creates personal wealth but little, if any value for society. This means that a company makes profits and/or owners

22

achieve their personal goals, however this type of enterprise does not add anything to economic growth. It can be argued that these companies are most likely to be involved in illegal or at least unethical activities. Still the situation in ’real economies’ in this regard can be well characterized by a quote from Davidsson (2004: 14), who argues that “…if one enjoyed the luxury of dealing with an economy characterized by a perfect and just institutional framework it would be easy to argue that this type of entrepreneur equals illegal ventures. Regrettably, we have to live with the fact that in real economies ’legal, yet re-distributive’ and ’illegal, yet societal beneficial’ are both possible.” One example of this type of entrepreneurship might be the introduction of new innovative pyramid schemes or new distribution methods for drugs (Davidsson and Wiklund, 2001). “A graffiti removal operation whose owners used nighttimes to generate demand for their businesses represents another example of this type of entrepreneurship” (Davidsson, 2004: 14). It is important to mention here that such business activities can also attract followers, e.g. inspiring other companies seeing such an opportunity to make ‘easy money’ by conduct similar activities thus further ‘damaging’ the market (Sauka and Welter, 2007a). Failed enterprises (quadrant IV, Figure 3) represent genuine failures, i.e. enterprise attempts that never made it off the ground, therefore lacking any positive spill-over effect on other actors (Davidsson and Wiklund, 2001). One could argue that this type of business activity can be disregarded from the point of value creation. In the long term, however, some of these entrepreneurs can bring some value to society (e.g. Davidsson, 2004). These are instances, especially in an early transition environment, when entrepreneurs without any or with very limited business experience and management training learn by doing. In this context, even if initial attempts to develop a business fail, entrepreneurs may learn the lesson and then use their experience for further business activities which are potentially more successful (Sauka and Welter, 2007a). Catalyst enterprises (quadrant II, Figure 3) represent cases when, albeit unsuccessful on a firm level, companies add to society and the economy. For example, Baumol (1993) and Dallago (1997) state that all that is necessary for entrepreneurship to be productive is that an activity yields a positive marginal product. In this context there may be situations when ideas and methods developed by one enterprise are imitated or sometimes stolen and successfully exploited by others 23

(Davidsson and Wiklund, 2001). In such a situation the initial inventor of the product or service does not make any profits, but might still contribute to society by having driven the process of discovery and creative destruction. Another case of such entrepreneurship would be when the potential threat of the new enterprise leads competitors to innovative responses that benefit society while keeping the new enterprise out of the market (Davidsson and Wiklund, 2001). In light of this discussion, as noted by Davidsson (2004), one of the shortcomings of Figure 3 is, of course, its oversimplification. The main shortcoming here is the lack of introduction with a time frame and exact context. Further, only two levels of analysis, e.g. venture and societal, out of many possible levels (such as industrial, regional, national, and world) were discussed (Davidsson, 2004). As regards the levels of analysis, a similar limitation is also applied for the scope of this thesis in order to address productive, unproductive and destructive entrepreneurship. More specifically, drawing on the framework of analysing the link between activities and SME output as suggested by Davidsson and Wiklund (2001) and Davidsson (2004), output is further addressed on a venture level and as a company’s contribution to the growth of the economy or society in general (e.g. societal level). Such a limitation is necessary in order to make a research model for the assessment of productive, unproductive and destructive entrepreneurship manageable, as, unfortunately, not all dimensions can be included in the analysis, especially when it comes to the empirical assessment of these concepts (e.g. Sauka and Welter, 2007a) 2.2.2. Output and activities in the context of the assessment of productive, unproductive and destructive entrepreneurship In light of this discussion, several authors acknowledge the importance of considering the output of SME activities when productive, unproductive and destructive entrepreneurship is addressed on a conceptual level (e.g. Davidsson and Wiklund, 2001; Baumol, 1993). As argued by Davidsson (2004), for instance, ‘unproductive entrepreneurship activity’ can also lead to some positive output on both a venture and societal level, whereas ‘productive entrepreneurship activity’ will not necessarily lead to a successful company performance or its contribution to society. Drawing on the conceptualisation of productive, unproductive and destructive entrepreneurship in literature, Sauka and Welter (2007a) introduce a new conceptual

24

framework for the assessment of productive, unproductive and destructive entrepreneurship. According to Sauka and Welter (2007a), ‘negative’ activities, such as, for example, rent seeking or tax avoidance, which create positive output should not be regarded as unproductive. On the other hand, ‘positive’ activities, for instance, innovation, which make no contributions on venture or societal levels, should not be regarded as productive. More specifically, Sauka and Welter (2007a) argue that in general, there is no agreement in entrepreneurship literature (e.g. Baumol, 1990, 1993; Dallago, 1997, 2000; Foss and Foss, 2002) as to whether productive or unproductive entrepreneurship refers to activities only or to output or to both. This ‘mix’ of two different dimensions, activity and output, in combination with a blurred and often inappropriate use of terminology, could be a key to the problems both in developing this concept further and for its empirical assessment (Sauka and Welter, 2007a). This is rather clearly exemplified by existing empirical studies assessing important aspects of productive, unproductive and destructive entrepreneurship, which mainly view productive, unproductive and destructive entrepreneurship in terms of ‘illegal, legal, formal, informal and rent seeing activity or behaviour’ (see table 1). In this light, Sauka and Welter (2007a) suggest a need to distinguish between activities and output (venture and societal level) in order to assess productive, unproductive and destructive entrepreneurship. While emphasizing the importance of shifting the focus from the activities to output, Sauka and Welter (2007a) argue that SME output has to be assessed together with the activities which create this output (see Figure 4). According to the authors, this has not yet been properly addressed in literature. Referring to the working definition of entrepreneurship as introduced in the first chapter of the thesis, here activities include any entrepreneurial efforts aimed at profit-making or personal satisfaction. Furthermore, following the notion that the generation of profits and ‘achievement satisfaction’ are the main driving forces behind entrepreneurial behaviour9, Sauka and Welter (2007a) argue that activities 9

Although referring to Baumol’s quotation of Veblen (1904) and Hobsbawm (1969), the author of this

thesis argues that the main driving force behind entrepreneurship is profit (also in terms of personal fulfilment), we should note that entrepreneurs can be driven to do business for other reasons as well, which is addressed in the recent discussion on social entrepreneurship. Baumol (1990: 10), for instance, describes another example, referring to nobles in the context of the Early Middle Ages: “… Many of

25

usually occur on a venture level. In this light, Sauka and Welter (2007a) conclude that this is output from these activities, not a type of activities as such, which then determines the involvement of SMEs in productive, unproductive or destructive entrepreneurship. Figure 4. Assessment of SME value creation: productive, unproductive or destructive entrepreneurship

Various activities (‘positive’ or ‘negative’) conducted by SMEs

Output on a venture and societal levelsuccessful or unsuccessful‘caused’ by various activities

Entrepreneurship: productive, unproductive or destructive

Source: Sauka and Welter (2007a) In this light, regardless of the type of activity (e.g. positive or negative) entrepreneurs become involved in, positive SME output will be interpreted as productive entrepreneurship,

whereas

negative

SME

output

refers

to

unproductive

entrepreneurship. Furthermore, following the discussion initiated in Section 2.1.10, in the context of this framework, entrepreneurship will be considered as destructive when various types of activities, for instance, different forms of illegal entrepreneurship, which in turn lead to negative output, attract followers. In the context of this framework, it is therefore output on different levels (e.g. a venture level and a societal level), ‘created’ by either ‘positive’ or ‘negative’ activities, which result in productive, unproductive or destructive entrepreneurship. As already highlighted in the introduction to the thesis, however, both short-term and long-term effects should be considered as some ‘negative’ activities may have a positive but short-term impact on either the performance of a company or its contribution to the growth of the economy, while in the long term, the same activity might lead to harmful output. Considering the difficulty of providing a clear-cut distinction between the output of ‘positive’ and ‘negative’ entrepreneurial activities the turbulent barons enjoyed fighting for its own sake, and success in combat was an important avenue to prestige in their society. … “. In a different manner, of course, one could apply this notion to today’s entrepreneurs as well.

26

and in line with Dallago (1997: 104), who states that the entrepreneur’s social role can be productive in the sense that he or she contributes directly or indirectly to the net output of the economy or to the capacity to produce additional output, Sauka and Welter (2007a) suggest supplementing the existing concept (see Figure 3) with two additional categories, namely ‘directly productive’ and ‘indirectly productive’ entrepreneurship. Figure 5 presents the approach introduced by Sauka and Welter (2007a). Quadrant 1 represents directly productive entrepreneurship, where SMEs perform successfully on both venture and societal levels, thus creating value both on economic and corporate levels. ‘Indirectly productive entrepreneurship’ can be represented either by Quadrant 2 or Quadrant 4. In the former case, although SMEs are not successful on a venture level, they can still bring some ‘indirect’ value to society, as discussed above with regard to followers. Learning by doing from an unsuccessful previous business experience may be one example of an indirectly productive contribution from companies as represented by Quadrant 4. The unproductive entrepreneurship in Quadrant 2 is mostly associated with a lack of managerial ability to implement innovative ideas, products or services or to exploit opportunities. Finally, Quadrant 3 represents either unproductive or destructive value creation (Sauka and Welter, 2007a).

10

See ‘illegal entrepreneurship activities’

27

Figure 5. Directly and indirectly productive, unproductive and destructive entrepreneurship

+

Venture level outcome

-

Quadrant I

Quadrant II

Directly productive entrepreneurship

Unproductive or indirectly productive entrepreneurship

Quadrant III

Quadrant IV

Unproductive or destructive entrepreneurship

Unproductive, indirectly productive entrepreneurship

-

Societal-level outcome

+

Source: Sauka and Welter (2007a), based on Davidsson and Wiklund (2001: 91), Davidsson (2004: 13) 2.3. Conceptualisation of the term ‘activities’ In the scope of the framework as introduced in the previous section of this chapter, however, various types of entrepreneurial activities should be considered. Although a number of different activities can be distinguished from the various trends in entrepreneurship literature, following the suggestions of Wiklund (1998), in the framework of this thesis, ‘activities’ are conceptualised as individual-level entrepreneurial behaviour and venture-level business strategy. Further, ‘physical activities’, such as, for example, processes in manufacturing or services, can be distinguished. Entrepreneurial behaviour and strategy, as used in the scope of this thesis, however, needs more explanation and operationalisation, and this is the aim of this section of the thesis. Entrepreneurial behaviour (EB) has been operationalised in a number of ways in entrepreneurship literature. Drawing on a review of existing studies, Wiklund (1998: 224), for example, argues that, in general entrepreneurial behaviour “ … points to a number of actions that can be regarded as entrepreneurial, i.e. the development of new products and markets, proactive behaviour, risk-taking, the start- up of new

28

organisations and the growth of an existing organisation.” As highlighted in the introduction to the thesis, however, influenced by external environment, entrepreneurship in transition countries has specific features compared to more advanced market economies. In light of this discussion, inadequate access to external capital, frequent changes in legislation, the general attitude of government and society towards entrepreneurs, frequent tax inspections, government corruption and other similar constraints facing entrepreneurs in transition as found by previous empirical studies (e.g. Smallbone and Welter, 2001; Aidis, 2006; Kolodko, 1999; Aidis and Sauka, 2005) also determine specific types of entrepreneurial behaviour in transition countries. In this context, drawing on an institutional perspective, authors such as Welter and Smallbone (2003) argue that entrepreneurs in transition tend to become involved in ‘evasion behaviour’ in order to overcome these constraints, mostly determined by the external environment. Evasion behaviour can have various forms, but mostly refers to a combination of legal and informal activities, including efforts to lower or avoid taxes, the reduction of social security contributions to employees by hiding actual wages or not offering labour contracts, and various informal payments to government officials to ensure contracts and similar activities (e.g. Welter and Smallbone, 2003; Chepurenko, 1999; Gustafson, 1999). As emphasized by Welter and Smallbone (2003), although ‘evasion behaviour’ such as tax avoidance is not uncommon in more mature economies as well, entrepreneurs in transition countries use to follow it “… with the underlying objective to preserve the capital base of their enterprises” (p. 103). In this context, evasion behaviour is also often regarded as a ‘norm’ in society, especially in early transition settings characterised by an even more hostile environment (ibid). Although various types of ‘evasion behaviour’ are perceived as ‘normal’ experiences in transition countries (e.g. Welter and Smallbone, 2003), in general such behaviour is characterised as a ‘deviance’ or ‘departure from norms’ (e.g. Warren, 2003). In this light Warren (2003) distinguishes between two schools of thought regarding ‘deviant behaviour’. According to Warren (2003) the first school of thought consists of literature focusing on negative forms of deviant behaviour, such as tax avoidance, for example. The other school of thought emphasises the beneficial forms of deviant behaviour for organisations, such as functional disobedience and tempered

29

radicalism11. The main conclusion in this regard is that deviance in organisations may be associated with both desirable and undesirable behaviour (Warren, 2003). Warren’s conception (2003) is especially relevant

to the framework

developed by Sauka and Welter (2007a), who acknowledge both positive (e.g. ‘desirable’) and negative (‘undesirable’) output from various ‘productive’ or ‘unproductive’ types of activities. In the context of this discussion, drawing on the conceptualisation of productive, unproductive and destructive entrepreneurship as introduced by Baumol (1990, 1993) and further elaborated by Sauka and Welter (2007a), the author therefore refers to rent seeking, corruption, and different kinds of illegal, informal and unethical behaviour as ‘deviant behaviour’ in this thesis. The opposite of deviant behaviour is labelled as ‘conforming’ (e.g. Sauka and Welter, 2007a), involving, for example, paying taxes and honest competition. In the context of the framework as introduced by Sauka and Welter (2007a), both conforming and deviant behaviour could thus result in ‘positive’ and ‘negative’ output on both venture and societal levels, short and long term. Some entrepreneurship scholars argue that how entrepreneurs operate is even more important than what they do (Wickham, 2004). In this context, drawing on a resource-based perspective, the role of venture-level strategy is highlighted as an important influence on the performance of companies, determining the appropriate use of resources and capabilities available to the company (Wiklund, 1998). In the context of assessing productive, unproductive and destructive entrepreneurship, the influence of strategy on the output of firms should thus be considered. Increasingly, in recent years entrepreneurial orientation (EO) has been used as a major construct to assess firms’ strategy (e.g. Morris and Kuratko, 2002). In general, the concept of EO is based on the previous research-assessing strategy-making “… in terms of patterns of action or decision-making styles that are generalisible across organisations” (Dess and Lumpkin, 2005: 147), deriving dimensions of EO from both entrepreneurship and strategy research. The most often used dimensions of EO, as originally proposed by Miler (1983), are innovativeness, proactiveness and risktaking. Further, Lumpkin and Dess (1996) have introduced two additional dimensions: competitive aggressiveness and autonomy. It is argued that altogether these five dimensions can reflect managers’ decision- making styles and practices, 11

See Warren (2003) for a review of studies on positive and negative deviance.

30

enhancing the performance of firms (Dess and Lumpkin, 2005). Definitions of the dimensions of EO, as summarised by Dess and Lumpkin (2005) are presented in Table 2.12 In sum, for the scope of the thesis, the term ‘activities’ is thus conceptualised as individual-level entrepreneurial behaviour and venture level entrepreneurial orientation. Table 2. Dimensions of Entrepreneurial Orientation Dimension

Definition

Autonomy

Independent action by an individual or team aimed at bringing forth a business concept or vision and carrying it through to competition.

Innovativeness

A

willingness

to

introduce

newness

and

novelty

through

experimentation and creative processes aimed at developing new products and services, as well as new processes. Proactiveness

A forward- looking perspective characteristic of a marketplace leader that has the foresight to seize opportunities in anticipation of future demand.

Competitive

An intense effort to outperform industry rivals. It is characterised by a

aggressiveness

combative posture or an aggressive response aimed at improving one’s position or overcoming a threat in a competitive marketplace.

Risk-taking

Making decisions and taking action without certain knowledge of probable outcomes; some undertakings may also involve making substantial resource commitments in the process of venturing forward.

Source: Dess and Lumpkin, 2005 2.4. Antecedents and influences for activities Referring to the ‘entrepreneurial cycle’, as introduced in the first chapter of the thesis, when discussing the relationship between processes prior to involvement in business activities, activities and output, regardless of the business stage, as well as consequences from this interaction, another important dimension should be considered. Namely, motivation is often highlighted as the core influence on the output of entrepreneurial activities, determining the types of activities SMEs become involved in as well as shaping the growth of the firms in general. As argued by 12

See Dess and Lumpkin (2005) for a further analysis of each dimension of EO.

31

Wiklund (1998: 40), „The underlying logic in the motivation perspective is that someone’s choice of work tasks and the time and energy devoted to these work-tasks (e.g. growing of firm), is dependent on the individual’s motivation to perform different tasks”. Differences in motivation, however, can further explain differences in small firm performance (e.g. Miner, 1990; Wiklund, 1998). Wiklund (1998) also emphasises the necessity to combine various motivation theories, as opposed to focusing only on one single motivation theory. In this light, the conceptual model of motivation by Locke (1991) and Locke & Henne (1986) can be used as a background to describe and conceptualise the influence of motivation on SME involvement in various activities (Wiklund, 1998) (see Figure 6). Figure 6. Model of motivation Expectancy and selfefficacy

Needs

Distal and general

Goals and intentions

Values

Activities

Proximal and specific

Source: Locke (1991) and Locke & Henne (1986) In this light, the influence of motivation on actions can be determined by goals and intentions (influenced by needs and values), and expectancy and self-efficacy. The influence of goals and expectations on activities and also on a company’s output has been specifically highlighted (e.g. Davidsson, 1991; Wiklund, 2003; Wiklund and Shepherd, 2003). For instance, it has been highlighted that the personal aspirations of managers of companies ambitious to grow entrepreneurially impact positively on business performance (Gray, 1992). In this light, scholars affirm that an orientation towards growth can be seen as desirable objective, as it is often associated with increased profits through a more optimal scale of operations (Barrow, 1988).

32

Maximising profits and personal achievement (e.g. Cooper and Artz, 1995) have been highlighted as the main goals in the main driving forces of business activity. In terms of assessing the role of business goals and expectations, strategic orientation research (e.g. Stratos, 1990; Poutziouris, 2003) distinguishes between growth-oriented firms, survival-oriented firms, exit-oriented firms and controloriented firms. To summarise, growth-oriented firms are likely to be younger, smaller and more willing to take risks than survival-oriented companies. Control-oriented firms are most likely to be family companies operating in traditional sectors, whereas exit-oriented companies tend to be sizeable firms mostly involved in manufacturing (Poutziouris, 2003). When addressing exit-oriented firms, however, it should be considered that some of them can also be owned and managed by so-called serial entrepreneurs, who create, succeed, exit and then re-launch other businesses (Poutziouris, 2003). In light of the general discussion, it is furthermore important to consider that goals and expectations can both emerge and change during different stages of the entrepreneurship process. Building on these arguments, in the scope of the thesis the influence of motivation on activities is limited by focusing on entrepreneurs’ goals and expected consequences regarding their business activities. Although other dimensions, such as, for example, efficacy, have been highlighted as possible influences on the various types of activity, these elements are not included in the research model. The main line of argumentation in this regard is that, although a research problem should be viewed from as many perspectives as possible, it is still necessary to keep the research model manageable. Thus, sidestepping some dimensions is often inevitable. 2.5. Influence of environment As discussed in the introduction to the thesis and highlighted throughout, in order to assess entrepreneurial processes, contextual influences should be considered, as they shape not only the role of entrepreneurship and small firms but also their structure and performance (Karlsson & Dahlberg, 2003). In this regard institutional theories have been recognised as a suitable framework for analysing context-shaping entrepreneurial processes, especially in a transition setting (Aidis, 2006; Van de Mortel, 2002). This section provides an overview of these theories, conceptualising

33

the term ‘environmental influences’ for the assessment of productive, unproductive and destructive entrepreneurship in the scope of this thesis. Institutional theories have been applied by numerous researchers as a useful framework for analysing the influences of the context on entrepreneurship development (e.g. Smallbone and Welter, 2001; Aidis, 2006). Institutional theories treat institutions, uncertainty and processes in very different ways. The most widely used distinction between institutional theories distinguishes among old theories, often divided into the school of Veblen and the school of Commons, and neo- institutional theories. Neo-classical theories, however, are considered to be the result of a split in the ‘old institutional theory’ and can be further separated into followers of the school of Commons and the school of Veblen. Further, among other classifications, such as those introduced by Langlois (1986) and Knudsen (1993), several other schools, for example the post- Keynesian, Schumpeterian and Marxist schools, are also sometimes classified as neo-institutionalist on the one hand and new institutional theories on the other (Hodgson, 1993; Groenewegen et. al., 1995). In this light, neo-classical economics is generally recognised as the initiator of new institutionalism. New institutionalist approaches are considered to be even more diverse than old and neo- institutional approaches. The main reason for this is the tendency to use the term ‘new institutionalist’ when referring to all theories that diverge at some point from neo- classical economics (Hodgson, 1993). New institutional approaches are characterised by the assumption that the formation of institutions is the result of individual actions. In this light, the possibility that individuals may be shaped by institutions is not considered, apart from the influence from the supply of information or constraints (Hodgson, 1993). In this view, the crucial difference between new institutionalism and old/ neo- institutionalism is whether individuals are only constrained by institutions or also influenced by them (Hodgson, 1993; Groenewegen et.al., 1995). Some aspects of old/ neo- and new institutionalism are added to Hodgson’s classification by Rutherford (1994) as well . A possible diversification in the new institutionalist approach is introduced by D. North (1990), who defines institutions as the rules of the game in society, or more formally, constraints devised by humans shaping their interactions. According to entrepreneurship literature, the new institutional theoretical approach by North (1990) is suitable for the analysis of entrepreneurship processes, especially in hostile environments such as a transition setting, since it acknowledges the direct link 34

between individual attitudes and perceptions and their effect on economic performance (e.g. Smallbone and Welter, 2006; Aidis, 2006). Further, an institutional approach also takes into consideration the economic and social influence of both formal (laws and regulations) and informal (customs, culture) rules and constraints. In the context of this framework, as summarised by Aidis and Sauka (2006 : 4): “The behaviour of the involved actors (e.g. what they can and must do and what is advantageous) is formed by the existing institutions and plays a critical role in reducing overall transaction costs by reducing the level of uncertainty and risk involved in individual exchanges. Moreover, these institutions, important for facilitating market exchanges and for shaping entrepreneurial activity, are comprised of two distinct categories: formal and informal institutions”. North (1990) refers to formal institutions as the visible 'rules of the game’, including political and judicial rules and contracts. Formal rules are generally enforced by governments and can be altered quickly to adapt to changing economic circumstances (Welter and Smallbone, 2003). It should be noted, however, that although formal rules are officially codified for society as a whole, they are not always created for the purpose of social well- being. In this regard, private interests may play a key role in creating formal rules and this issue needs to be considered (North, 1994). In contrast, informal institutions are often referred to as the invisible ‘rules of the game’ and are made up of the norms, values, acceptable behaviour and codes of conduct of a given society. Informal rules most often evolve to complement formal rules and have a fundamental influence on economic development (Raiser, 1999). According to North (1990:88) “Informal rules tend not to be legally enforced and change to informal rules occurs more indirectly and usually as a result of accidents, learning, natural selection and most of all the passage of time.” Although formal and informal institutions tend to complement one another, North (1990) has identified the often-conflicting role of formal and informal institutions, possibly leading to inefficient institutional outcomes (North, 2005; Aidis, 2006). More specifically, especially during periods of rapid socio-economic change, informal rules may clash with new formal rules. For instance, informal rules may continue to persist since they have tenacious survival ability; they have become part of habitual behaviour and informal institutions provide a sense of stability. When formal and informal institutions clash as in cases when formal rules are changed but informal rules have not yet been changed, the formation of underground economies is 35

possible as a result of the proliferation of noncompliant behaviour (Feige, 1997). As North (1997b: 16) notes: “…the performance of an economy is an admixture of the formal rules, the informal norms, and their enforcement characteristics. Changing merely the formal rules will produce the desired results only when the informal norms are complementary to the rule change, and enforcement is either perfect or at least consistent with the expectations of those altering the rules.” In light of this discussion, North’s emphasis on the influence and often conflicting role of formal and informal institutions on economic outcomes is especially relevant for a transition context. As North notes, the incentive structures provided through the institutional environment directly affect outcomes: “If the institutional framework rewards piracy then piratical organizations will come into existence; and if the institutional framework rewards productive activities then organizations and firms will come into existence to engage in productive activities” (North 1994:361). In light of this discussion, however, according to entrepreneurship literature, formal and informal rules present a limited picture of the overall economic milieu within which entrepreneurship may develop (e.g. Acs and Karlsson, 2002). Other important variables, including demand and supply conditions, the degree of market competition, the state of the infrastructure, the supply and skill level of the labour force, the entrepreneurial climate and access to knowledge and finance, must also be analysed (e.g. Aidis, 2006; Aidis and Sauka, 2005). In order to capture these influences on entrepreneurial development, North’s institutional theory can be extended by adding an ‘economic’ category of analysis as well as an additional category to capture factors not included in the other three categories (Aidis and Sauka, 2005). The additional ‘other’ category includes issues such as a lack of SME policies, procurement problems and the accessibility of specific consulting advice. As shown in Figure 7, this extended institutional approach includes four categories of influences on business development and growth: formal, informal, economic and other factors

13

13

.

See Aidis and Sauka (2005) for a more detailed description of the extended institutional approach.

36

Figure 7. Framework for analysing environmental influences on entrepreneurship processes within a transition setting Formal institutions

Informal institutions

Economic factors

Other factors

Entrepreneurship processes within an advanced transition setting

Source: Aidis and Sauka (2005) This framework is furthermore used for the scope of the thesis in order to assess environmental

influences

on

productive,

unproductive

and

destructive

entrepreneurship within an advanced transition setting. 2.6. The relationship between motivation, activities, environmental influences and output In the context of the discussion as initiated previously in this chapter, productive, unproductive and destructive entrepreneurship is conceptually determined by the following main elements: processes prior to involvement in business activity, e.g. goals and expectations, activities, e.g. entrepreneurial behaviour and entrepreneurial orientation, and SME value creation, e.g. outcomes on a venture and societal level, short and long term. It was then emphasized that environmental influences, e.g. formal, informal, economic, and other factors should be considered when productive, unproductive and destructive entrepreneurship is addressed. In order to develop a framework for the assessment of productive, unproductive and destructive entrepreneurship, however, it is important to conceptualise not only various elements but also relationships between the concepts themselves. It is thus crucial to explore the relationship between the determinants of and the influences on productive, unproductive and destructive entrepreneurship. The next sub-sections of the thesis deal with this issue, aiming to introduce a conceptual

37

framework allowing for the empirical assessment of productive, unproductive and destructive entrepreneurship. 2.6.1. Conceptualising the general relationship between activities and value creation While attempting to explore the relationship between EO and output, previous studies tend to regard EO as venture-level phenomena (Covin, Green and Slevin, 2006), linking the dimensions of EO with the venture-level performance indicators. The positive role of higher EO on the output of SMEs is usually highlighted in this regard (Wiklund, 1998), e.g. the findings presented by most of the existing empirical studies linking EO and output show that EO positively affects the performance of firms (e.g. Wiklund, 1998; Zahra and Covin, 1995; Wiklund and Shepherd, 2005). It should be noted, however, that these findings are not consistent, since some empirical evidence shows that no statistically significant relationship between these concepts exists (e.g. Smart and Conant, 1994). When assessing the possible impact of EO on SME output, however, various time perspectives are not considered by the previous empirical studies. In the context of the framework introduced by Sauka and Welter (2007a), the author argues that, as in the case of conforming and deviant behaviour, EO’s effect on SME output can vary in the short and long term. To fill the research gap, in the scope of the framework introduced by Sauka and Welter (2007a), this thesis explores the impact of EO on venture and societal level performance, both in the short and long term. Furthermore, as argued by Wiklund and Shepherd (2005), time is necessary for the effects of EO to take place. Thus it is advisable to use a longitudinal study where independent variables such as EO are measured at one point in time and dependent variables, e.g. output, at least one year later. In this thesis, the author uses such a longitudinal design in order to assess the direct effect of EO on SME output. More specifically, drawing on Wiklund and Shepherd (2005), the following hypothesis is tested in order to address the direct relationship between EO and SME value creation (e.g. output on venture and societal levels, short and long term):

38

H1: EO has a universal positive effect on SME value creation. That is, firms with higher EO will outperform firms with a lower level of EO both on a venture and societal level, short and long term. A number of empirical studies have also attempted to explore the relationship between various types of EB and some aspects of productive, unproductive or destructive entrepreneurship. By addressing the effect of forms of deviant behaviour on the output of firms, Aidis and Van Praag (2007), for example, report a statistically significant, positive relationship between illegal entrepreneurship experience and business performance for younger entrepreneurs and entrepreneurs who started a completely new legal business in a transition context. A few studies have attempted to address the effect of illegal entrepreneurship experience and entrepreneurship output in terms of the probability of a business start-up, both in transition and more advanced market economies. Earle and Sakova (2000), for instance, find a positive relationship between having a side- business during a pre- transition year and a business start-up after the start of the transition process in various transition countries. Fairlie (2002), however, reports on the significant relationship between illegal entrepreneurship experience, measured as drug dealing experience in the past, and the probability of legal self-employment. As in the case of EO, the previous empirical studies do not provide evidence on the relationship between involvement in conforming or deviant behaviour and SME output on various levels and in various time frames. As highlighted in the previous chapter of the thesis, however, some deviant activities can have a positive, but only short-term effect on SME performance, whereas in the long term these activities can actually lead to unsuccessful SME output on both venture and societal levels. Considering this notion and drawing on the existing empirical evidence, the following hypothesis is introduced in order to explore the relationship between EB and SME value creation: H2 (a) Those SMEs involved in deviant forms of EB on a larger scale will outperform firms preferring conforming forms of EB, both on a venture and societal level, but only in the short term. (b) In the long term, SMEs following deviant behaviour will be less successful than firms involved in conforming behaviour.

39

In the context of the framework introduced by Sauka and Welter (2007), the general relationship between activities (e.g. entrepreneurial behaviour and entrepreneurial orientation) and SME value creation, shaping (directly and indirectly) productive, unproductive and destructive entrepreneurship is illustrated in Figure 8. Figure 8. General relationship between activities and value creation

Entrepreneurial Behaviour

SME value creation

Directly and indirectly productive, unproductive or destructive entrepreneurship

Entrepreneurial Orientation

Source: Author 2.6.2. Conceptualising the relationship between motivation, activities and value creation Drawing on the discussion initiated in Section 2.4., the influence of motivation, e.g. goals and expectations, on various types of activities and resulting SME value creation should be addressed when entrepreneurship processes are analysed. Empirical studies assessing this relationship highlight that entrepreneurs with strong motivation, e.g. desire to succeed, are more likely to exploit entrepreneurial opportunities and thus perform better (e.g. Shane and Wenkatamaran, 2000). Aidis and Van Praag (2007) further report a significant and positive link between illegal entrepreneurship experience and motivation, whereas Aidis, Mickiewitz and Sauka (2007), find a statistically significant positive relationship between higher goals and expectations and the output of SMEs. Considering the positive relationship between motivation and SME output, as well as motivation and activities found by the existing empirical studies, and attributing this relationship to an advanced transition setting, it could thus be argued that entrepreneurs with higher goals and expectations will be both more EO-oriented

40

and active in terms of (either conforming or deviant) EB. With regard to EB, as argued by H2, active involvement in conforming behaviour will result in positive output in the long term, whereas active involvement in deviant behaviour will be ‘rewarded’ on both venture and societal levels, but only in the short term. Referring to these notions, the following hypothesis has been developed in order to address the relationship between motivation and involvement in EO, EB and resulting SME value creation: H3: Goals and expectations positively influence the degree of EO and EB, leading to better SME value creation. The relationship between goals and expectations and activities and value creation is illustrated in Figure 9. Figure 9. Proposed relationship between goals, expectations and EO and EB, affecting SME value creation

Goals and expectations

Entrepreneurial Behaviour & Entrepreneurial Orientation

SME value creation

Source: Author 2.6.3. Addressing the complexity of the relationship between activities and SME value creation Recent studies exploring the link between EO and SME output on a venture level highlight the necessity to consider various additional influences when this relationship is addressed (e.g. Lumpkin and Dess, 1996). Attempting to assess the complexity of the EO and performance relationship, Wiklund and Shepherd (2005), for example, emphasise that previous studies mostly use the ‘universal effect approach’, assuming that EO has a universally beneficial effect on SME output, or exploit contingency models, capturing the two- way interaction between EO and external or internal environment characteristics. 41

In this light, previous empirical evidence suggests that EO is context-specific and has a more positive influence on output in environments perceived by SMEs as hostile (Covin and Slevin, 1989; Zahra and Covin, 1995 in Wiklund and Shepherd, 2005). Furthermore, access to financial capital is highlighted as an internal characteristic with an important impact within the EO and SME output relationship. In general, previous findings show that more access to financial capital facilitates EO14, as provides possibilities to experiment and thus both create and exploit new business opportunities (e.g. Zahra, 1991; Wiklund and Shepherd, 2005). In light of this discussion, the interaction of environmental influences and access to financial capital should therefore be considered when addressing the relationship between EO and SME output (Wiklund and Shepherd, 2005). Following the aforementioned arguments, it is furthermore of importance to include both venture and societal level dimensions and to consider various effects of time when the EO and output relationship is addressed. Referring to this notion and drawing on the existing empirical evidence, in line with Wiklund and Shepherd (2005), the following hypotheses are addressed in order to capture the two- way relationship between EO, SME value creation and environmental influences as well as EO, output and access to capital: H4: The relationship between EO and SME value creation is moderated by environmental influences. SME value creation increases with EO, but at a higher rate for those SMEs perceiving the environment as hostile. H5: The relationship between EO and SME value creation is moderated by access to capital. SME value creation increases with EO but at a faster rate for SMEs that are less concerned about access to financial capital. Although

widely

used

in

entrepreneurship

research,

main-effects-only

relationship and contingency models do not provide a broad enough picture of the relationship between EO and SME output (e.g. Wiklund and Shepherd, 2005). In this light, Lumpkin and Dess (1996) argue that additional insight into the complexity of the relationship between EO and SME output can be gained by assessing a three-way 14

See Wiklund and Shepherd (2005) for a further review of the literature on the role of environmental influences and access to capital in the EO and performance relationship.

42

interaction, considering the joint influences of both environmental influence and access to capital, e.g. using a configurational approach. Following these suggestions and exploiting hierarchical regression analysis, Wiklund and Shepherd (2005) check for the main-effects-only model (EO vs. performance), then the two-way interaction term (contingency model: EO vs. performance moderated by access to capital and EO vs. performance moderated by environmental influences) and furthermore the threeway interaction term (configuration model: EO vs. performance moderated by access to the capital and environmental influences). Wiklund and Shpeherd (2005) find that a three-way interaction model explains variance in performance “… over and above a contingency model and a main effects only model” (p. 1)15. Furthermore, as summarised by Wiklund and Shepherd (2005 : 79), according to the configurational approach “EO has the strongest positive effect on performance among firms in hostile environments with substantial access to financial capital and the strongest negative effect on performance among firms in stable environments with little access to capital”. Drawing on this conception, the following hypothesis is addressed in order to capture the three-way interaction between EO, SME value creation, environmental influences and access to capital within the scope of this thesis: H6. (a) SME value creation is explained by configurations of EO, access to capital, and environmental influences. (b) SMEs value creation is better amongst firms with a higher degree of EO, less concerned about access to financial capital and in environments perceived as hostile than for other configurations. The general relationship addressing the complexity of the relationship between EO and SME value creation is illustrated in Figure 10.

15

See Wiklund and Shepherd (2005: 73, 82) for further discussion on direct-effect, contingency and configuration models as well as the appropriateness of using hierarchical regression in assessing the EO and performance relationship in this context.

43

Figure 10. Conceptualising the complexity of the relationship between EO and SME value creation

Entrepreneurial Orientation

SME value creation

Environmental influences

Access to capital

Source: Author, drawing on Wiklund and Shepherd (2005) As in the case of EO, additional influences should also be considered when the relationship between EB and SME value creation is addressed (e.g. Smallbone and Welter, 2001). As exemplified by various empirical studies, (e.g. Smallbone and Welter, 2006; Aidis, 2006), both formal and informal institutions shape the environment and are major influences on the involvement of SMEs in different types of entrepreneurial behaviour. In this light, emphasising the impact of environmental influences in the relationship between SME behaviour and output, Smallbone and Welter (2001), for example, suggest that in a transition context, especially an early transition setting, some forms of ‘deviant’ behaviour not only increase SME output but are actually necessary for SMEs in order to develop and expand their companies while overcoming constraints brought on by the uncertainty of the hostile environments in these countries. Morris, et.al (1997) furthermore report that the informal sector in general makes a significant contribution to the growth of developing countries and is also a major potential source of entrepreneurship. Furthermore, Sobel (2006) reports a positive and significant link between productive entrepreneurship and economic growth in various states of the U.S., highlighting that higher institutional quality, e.g. a less hostile environment, is associated with a higher level of productive entrepreneurship. In light of these findings, one can argue that, in a transition context, at least in the short term, a hostile environment can act as a catalyst for the successful performance of SMEs involved in deviant behaviour. A stable environment, as

44

exemplified by Sobel (2006), may in turn contribute to successful SME output in the long term. Drawing on these notions, the author proposes that relationship between conforming and deviant behaviour and SME value creation is moderated by contextual influences – formal, informal, economic and other institutions in the following way: H7: The relationship between EB and SME value creation is moderated by environmental influences. (a) SME output on both venture and societal levels increases in the short term with higher involvement in deviant forms of EB, but at an even higher rate for SMEs perceiving the environment as hostile. (b) SME output on both venture and societal levels increases in the long term with a lower level of involvement in deviant behaviour, but at a higher rate for SMEs perceiving the environment as stable. The author is not aware of any studies linking EB, SME value creation and access to capital. Access to capital, however, has been found to be of positive influence on SME output (e.g. Wiklund and Shepherd, 2005). In this light, while providing better possibilities to experiment and thus creating more business opportunities (e.g. Zahra, 1991; Wiklund and Shepherd, 2005), better access to capital could also enhance the level of EB within SMEs as such. It is, however, difficult to predict whether such behaviour will be directed to conforming or, on the contrary, deviant forms. In general, as in the case of EO, the role of the access to capital should be considered when the relationship between EB and SME value creation is addressed. Referring to the previous discussion regarding EB and SME value creation as well as the influence of a specific transition environment in this relationship, one can argue that in the short term, better access to capital could increase SME output, even if behaviour is directed towards less conforming forms. In the long term, however, access to capital will facilitate involvement in conforming forms of EB, thus leading to higher SME value creation. Referring to these notions, the following hypothesis has been developed in order to address the relationship between EB, access to capital and SME value creation:

45

H8: The relationship between EB and SME value creation is moderated by access to capital. (a) SME output on both venture and societal levels increases in the short term with higher involvement in deviant forms of EB, but at higher rate for SMEs showing less concern for access to capital. (b) SME output on both venture and societal levels increases in the long term with a lower level of involvement in deviant behaviour, but at higher rate for those showing less concern for access to capital. In order to address the complexity of EB and value creation, the author furthermore makes an attempt to test a configuration model, e.g. the three–way interaction considering the joint influence of the environment and access to capital within the EB and value creation relationship. In light of the discussion addressing the relationship between EB, SME value creation and environmental influences as well as EB, SME value creation and access to capital, the following hypothesis has been developed in order to test the configuration model: H9. (a) SME value creation is explained by configurations of EB, access to capital, and environmental influences. (a) SME output on both venture and societal levels is better in the short term for firms with a higher level of involvement in deviant forms of EB behaviour and that are less concerned about access to capital and in environments perceived as hostile than for other configurations. (b) SME output on both venture and societal levels is better in the long term for firms with lower levels of involvement in deviant forms of EB and that less concerned about financial capital and in environments perceived as stable than for other configurations. The proposed relationship between EB, SME value creation, environmental influences and access to capital is presented further in Figure 11.

46

Figure 11. Conceptualising the relationship between EO, SME value creation and environmental influences.

Entrepreneurial Behaviour

SME value creation

Environmental influences

Access to capital

Source: Author, drawing on Wiklund and Shepherd (2005) 2.6.4. Conceptualising the relationship between EO, EB and SME value creation When it comes to conceptualising the relationship between EO, EB and SME value creation in order to capture the determinants of and influences on productive, unproductive and destructive entrepreneurship, adding a behavioural dimension in order to explore the EO and performance relationship is also suggested by Dess and Lumpkin (2005), though in a different context. In the concluding section of their article, the authors state that “Studies could explore to what extent a leader’s behaviour enhances his firm’s propensity to be proactive, innovative and take risks” (p.153), providing support for including the behavioural dimension in our analysis. Drawing on the proposed relationships in the previous paragraphs of this section of the thesis, the joint relationship between EO, EB and SME value creation is illustrated in Figure 12.

47

Figure 12. Proposed relationship between entrepreneurial behaviour, EO, value creation, environmental influences and access to capital

Entrepreneurial Orientation

SME value creation

Entrepreneurial Behaviour

Environmental influences & Access to capital

Source: Author, based on Wiklund and Shepherd (2005) Although in might be interesting to empirically analyse such a ‘joint’ influence of both activities (EO and EB) and their interactions (e.g. contingency and configuration models) with environmental influences and access to capital, this thesis nonetheless focuses on exploring the relationships between EO and EB and SME value creation separately (e.g. as proposed in Hypotheses 1, 2 and 4 to 10). This choice arises from the fact that “empirically configurations can be presented by the simultaneous interaction of three variables” (Baker and Cullen, 1993; Dess et al, 1997; Miller, 1988 in Wiklund and Shepherd, 2005: 77). Thus explaining SME value creation by regarding the relationship as interaction between EO, access to capital and environmental influences, and EB, access to capital and environmental influences separately should be perceived as a prerequisite to exploit the model used by Wiklund and Shepherd (2005) and described in 26.2. and 2.6.3.. Furthermore, including all the interactions of both EO and EB with capital and environment would render the model and hypotheses too complicated and difficult to interpret.

48

In light of this discussion, as regards the interaction of EO and EB and SME value creation, it can, however, be argued instead that strategic choices of firms, e.g. EO, can not only influence SME value creation but also determine the degree of involvement in conforming or deviant behaviour. By further addressing the aforementioned suggestions by Dess and Lumpkin (2005) regarding the possible relationships between entrepreneurial orientation and entrepreneurial behaviour, this thesis explores the possible effect of involvement in deviant behaviour and EO on SME value creation (see Figure 13). The following hypothesis has been developed in order to address this relationship: H10: (a) EO has a positive effect on SME value creation and this relationship is mediated by EB. (b) SME value creation increases with EO but, in the short term, at a faster rate for SME that are more involved in deviant forms of EB. (c) SME value creation increases with EO but, in the long term, at a faster rate for SMEs that are less involved in deviant forms of EB. Figure 13. Proposed relationship between EO and EB, affecting SME value creation

Entrepreneurial Orientation

Entrepreneurial Behaviour

Value creation

Source: Author 2.7. Framework for the assessment of (directly and indirectly) productive, unproductive and destructive entrepreneurship In the context of the discussion in the previous sections, the key question addressed by this thesis is: “What activities do entrepreneurs pursue in order to expand their business as well as gain personal satisfaction and how does the nature and extent of these activities influence the value a company creates in terms of productive, unproductive or destructive entrepreneurship?” As highlighted in the first chapter of this thesis, however, although there is no lack of suggestions, no common agreement

on

what

constitutes

productive,

unproductive

or

destructive

49

entrepreneurship exists. In order to measure the productive, unproductive and destructive relationship, however, a conceptual framework allowing for the operationalisation of the concepts for empirical assessment is therefore necessary. Drawing on the identification of the main elements as well as the conceptualisation of the relationship between these concepts, as addressed by this chapter, the author proposes the following framework for the assessment of factors determining and influencing entrepreneurial value creation, leading to (directly and indirectly) productive, unproductive or destructive entrepreneurship (see Figure 14). Figure 14. The proposed framework for assessing (directly and indirectly) productive, unproductive and destructive entrepreneurship

Entrepreneurial Behaviour

Goals & Expectations

SME value creation

Directly and indirectly productive, unproductive or destructive entrepreneurship

Entrepreneurial Orientation Environmental influences & Access to capital

Source: Author More specifically, in the scope of this framework, (directly and indirectly) productive, unproductive and destructive entrepreneurship is explained by the interaction between activities (e.g. EB and EO), SME value creation (e.g. output on venture and societal levels, short and long term) as well as the moderating effect of environmental influences and access to capital in this relationship. Further, the possible influences of goals and expectations on activities are considered by the model as well as the mediation effect of entrepreneurial behaviour in the relationship

50

between EO and SME value creation, altogether shaping the value creation of firms. Overall, drawing on the framework introduced by Sauka and Welter (2007), the author proposes that activity will be regarded as productive if it results in positive output on either a venture or a societal level, short or long term. In turn, an activity will be regarded as unproductive if no value is added to the company or society. Furthermore, destructive entrepreneurship will represent cases when some activities (for example, illegal activities) will create negative output, at the same time attracting followers (Sauka and Welter, 2007a). In the context of this framework, it should furthermore be noted that a number of empirical studies highlight the importance of the business owner and firm characteristics influencing the performance and growth of companies. In this light, studies highlight the important role of human capital indicators, such as education and the business owner’s age, in the development and growth of SMEs (Cooper et al., 1994). Although the findings regarding the business owner’s age are not consistent (e.g. Burns, 2001), a higher level of education is usually associated with better SME performance (e.g. Watson et al, 2003; Chandler and Hanks, 1993; Watson et al, 2003). Further, as argued by Wiklund and Shepherd (2005 : 82), “Businesses of different size and age may exhibit different organizational and environmental characteristics, which in turn may influence performance. The same is true for firms in different industries”. Gender has also been found to be an important influence on the business development, e.g. female-run businesses tend to be smaller and less likely to grow than male-run businesses (e.g. Brush, 1992; Cooper et al., 1994). Drawing on the existing empirical evidence, the size, age and sector of a firm, as well as education and the gender are thus included as controls for the effect of activities on SME value creation.

51

3. Towards a measurement of productive, unproductive and destructive entrepreneurship

In the previous chapter of the thesis, the main concepts determining and influencing SME involvement in (directly and indirectly) productive, unproductive and destructive entrepreneurship and the relationship between these concepts were discussed. This in turn resulted in the development of a new conceptual framework to be used for the empirical assessment of these concepts. For obvious reasons it is, however, not possible to include all dimensions of, say SME value creation or environmental influences when it comes to addressing productive, unproductive and destructive entrepreneurship empirically. Moreover, as addressed by the previous chapter of the thesis, researchers have different conceptualisations of the same concepts (e.g. Wiklund, 1998) and different measures are used to assess the same concepts empirically. Thus, concepts need to be operationalised further for measurement. By addressing this issue, in order to allow the measurement of productive, unproductive and destructive entrepreneurship, this chapter aims to introduce the measurement scales for the concepts discussed in the previous chapter of the thesis.

3.1. Towards a measurement of EB and EO Although deviance in organisations has been addressed by a number of studies (see Warren, 2003, for a summary), the author is not aware of any previous attempts to empirically capture the degree of conforming and deviant behaviour in the context of assessing productive, unproductive and destructive entrepreneurship. In order to fill this gap, it is thus necessary to develop a measurement scale to address various types of EB within the scope of the framework introduced earlier in the thesis. In light of discussion initiated in the previous two chapters of the thesis, the background for constructing the measurement scale for conforming and deviant behaviour comes from the conceptual assessment of the concepts of productive, unproductive and destructive entrepreneurship as introduced by W. Baumol (1990), the assessment of ‘deviance’ or ‘departure from norms’ (e.g. Warren, 2003) and existing empirical studies addressing evasion or deviance in SME behaviour within a

52

transition setting (e.g. Smallbone and Welter, 2003; Chepurenko, 1999; Gustafson, 1999). Drawing on these contributions three main dimensions of deviant behaviour can be distinguished: ‘rent seeking behaviour’, ‘unofficial payments’, and ‘involvement in unethical behaviour’. It was also argued that in the scope of this thesis behaviour opposite to this is labelled as conforming. In this light, by developing a general framework for the assessment of productive, unproductive and destructive entrepreneurship, Sauka and Welter (2007a) also attempted to operationalise these concepts, providing the scope for introducing the measurement scale to assess conforming and deviant behaviour empirically within a transition setting (see Table 3). As highlighted by Sauka and Welter (2007a), it should be noted, however, that several types of both conforming and deviant behaviour could fall into more than one category. Tax evasion and other rent-seeking behaviour as well as ‘order self-generation’, for example, can also take illegal forms. Rather than offering clear cut-offs, this classification should be understood as a first approximation of the types of conforming and deviant behaviour. In the scope of this approximation, conforming behaviour is the reverse of deviant (Sauka and Welter, 2007a). Table 3. Operationalisation of ‘conforming’ and ‘deviant’ behaviour for measurement within an advanced transition setting Rent seeking Unofficial payment activity vs. conforming activity vs. conforming activity Underreporting business Various types of unofficial income payments Underreporting number of employees Hiding actual salary (envelope wages) Litigation

Unethical behaviour vs. conforming activity Informal networking Order self- generation activity Squeezing (newcomers) out of the market Appropriating someone else’s business idea

Lobbying Takeover Different forms of tax evasion and avoidance efforts Acquiring monopoly profit

Source: Sauka and Welter (2007a) based on Baumol (1990, 1993) No further operationalisation is required for the measurement of EO, since it is a well researched phenomenon. For the empirical assessment of EO, in this thesis the

53

author uses the five dimensions of entrepreneurial orientation as classified by Dess and Lumkin (2005) (see Table 2 in Section 2.3.). Each dimension of EO, e.g. autonomy, innovativeness, proactiveness, competitive aggressiveness and risk-taking is measured using opposite statements, representing high and low involvement in EO accordingly. 3.2. Towards a measurement of SME output on a venture level Even though no consensus as regards the measurement of small business performance exists, the venture-level output of SMEs is usually captured by indicators representing the expansion and growth of a company (Aidis and Mickiewicz, 2006). Venture profitability, turnover and increase in employees are the ways in which output on a venture level is typically measured (Chandler and Hanks, 1993; Robinson, 1999; Vesper, 1996; Watson et al., 2003). Some studies also include measurements such as export share, investments and foreign direct investments as well as the survival of firms (e.g. Audretsch, 2002; Acs and Audretsch, 2003). In addition, actual growth, growth in comparison to competitors as well as the firms’ owners’ or ownermanager’s expectations regarding growth are sometimes used for this purpose (Davidsson 1991; Wiklund et al., 2003; Wiklund and Shepherd, 2003; Aidis and Mickiewicz, 2006). Several shortcomings of these indicators are often highlighted in previous research. Turnover may reflect the size of a firm, however, taken on its own, it may not say much about the actual success of the company’s performance. Reported profit, on the other hand, tends to be either underreported or affected by the age of the firm,16 or it might not appear due to the investments a company may make towards future growth17. It is also typical for many firms to experience a period of losses or low profitability, especially in the initial phase of their existence. In addition, given the tendencies towards underreporting and the unique characteristics of small firms, it is difficult to devise a reliable growth measurement (Aidis and Sauka, 2006). Further, as summarised by Audretsch (2002), referring to the general usage of expansion and growth as a measurement for venture-level performance assessment, empirical studies conducted in both a European and a North American context reflect that growth rates 16

For new start-ups there is not much past history to rely on, which can lead to a sample selection bias.

17

This may often involve borrowed capital.

54

are higher for smaller and younger enterprises, especially in technology-intensive industries. In order to avoid these shortcomings, in the scope of the theoretical framework a set of various venture-level performance indicators is used to measure venture-level performance. These include profits, turnover, employment generation and investments. Additionally, an indicator assessing entrepreneurs’ perceptions about personal fulfilment is included, since venture-level indicators should measure both monetary and non- monetary dimensions. The generation of profits or growth achievement may be important, but they are not the only catalysts of the entrepreneurial process (Davidsson, 1989; Davidsson and Wiklund, 2001). Further, as argued in Chapter 1.1., some efforts, for example rent seeking, can lead to productive outputs in the short term whereas in the long term they can harm the economy (Sauka and Welter, 2007a). In light of assessing venture-level output, both monetary and nonmonetary indicators are therefore used considering both short and long-term perspectives (see table 4). Table 4. Operationalisation of venture-level output for measurement Monetary: Expansion and growth.

Generation of profits

Short and long-term dimensions.

Turnover Generation of employment Investment

Non- monetary. Short and long-term dimensions.

Personal fulfilment

Source: Author 3.3. Towards a measurement of SME output on a societal level Most entrepreneurship textbooks and articles highlight the importance of entrepreneurship, emphasizing its contribution to economic growth. In this light, new theories of industrial evolution allow the framework to link entrepreneurship and economic growth (Ericson and Pakes, 1995; Audretsch, 1995; Klepper, 1996). To summarise, although empirical evidence from different environments is often robust,

55

the findings of existing studies show that entrepreneurship, especially SMEs, play a significant role in the generation of employment (e.g. Birch, 1979; Davis, Haltiwanger and Schuh, 1996a, 1996b; Storey and Johnson, 1987), in innovations in products or services (e.g. Acs and Audretsch, 1990; Harhoff and Licht, 1996; Jovanovic, 2001), and in bringing new technologies to a country (e.g. Wagner, 1994). Entrepreneurs are the ones who attract investments from abroad, investing in infrastructure, including roads, electricity, communications (e.g. Acs and Audretsch, 2003). Some of them are involved in social activities, such as sponsorship for social needs by donations for hospitals, social organisations, sports, etc. After all they educate employees or provide them with a possibility to gain experience (e.g. Smallbone and Welter, 2001) and pay taxes to support the country in general (Aidis, 2006). Further, entrepreneurship activity involves spill-over effects, when knowledge generated by one firm can be used by other companies without significant costs involved in the transfer (Van Stel, 2006). Still, assessing societal-level contribution, e.g. capturing total SME output on a societal level, seems to be one of the most challenging tasks in entrepreneurship literature and no consensus exists in this regard (Sauka and Welter, 2007a). Indicators such as employment generation and innovativeness are regarded as the most appropriate for this purpose (e.g. Johnson et. al., 2000) as these are considered to be the main contribution of SMEs to economic development. The importance of job generation from small firms was first highlighted by Birch (1979), who reported that small firms created 81.5% of net new jobs in the U.S in the years 1969-76. Although there are diverging arguments (e.g. Fritsch, 1993; Wagner, 1995), later studies, also in Europe, confirmed the role of small firms in job generation, e.g. small firms create jobs at a faster rate than large companies (e.g. Storey, 1993; Spilling, 1995). In this light, as highlighted already by Birch (1979), only a small portion of firms, so-called gazelles, actually make the main contribution in the job generation process. Despite the contribution of entrepreneurship to the generation of employment, employment generation as an indicator is problematic when it comes to the assessment of entrepreneurship outcomes on a societal level. One of the main reasons here is that companies do not know how many jobs they actually create (e.g. Audretsch, Keilbach and Lehmann, 2006). That is, apart from ‘direct jobs’, companies also create ‘indirect employment’, for instance in related industries. Further, when 56

assessing job generation, it would be advantageous to know what type of jobs we are talking about. Only a few studies have analysed the quality of jobs provided by SMEs (e.g. Wagner 1997). Using innovativeness as the indicator for assessing entrepreneurship contribution on a societal level has proven to be even more difficult. The main challenge in this regard is the definition of innovative activity. Defining and measuring innovation, especially in the context of SMEs, is often problematic and no common agreement exists in entrepreneurship literature. The definition of the term innovativeness goes back to Schumpeter, who was the first to regard the entrepreneur in the context of creative activity. In his book “The Theory of Economic Development” (1912), Schumpeter identified the entrepreneur as a person who creates ‘new combinations’, such as innovations in economics. By concentrating on entrepreneurial action, Schumpeter (1934:66) listed five categories covering the concept of innovation. These are:

•

Introducing a new good

•

Introducing a new method of production - something as yet untried in industry

•

Opening a new market

•

Utilizing some new source of supply for raw materials or intermediate goods

•

Carrying out some new organizational form of industry In this light, there is still no consensus regarding measures for assessing

innovative activity. For this purpose, inputs into R&D activity, the numbers of patented inventions, and new product innovations are mostly used. However, these measures often neglect change as well as important types of innovative activity (Audretsch, 2002). Considering the difficulties described above, it is necessary to introduce a measurement scale in order to assess entrepreneurship’s contribution on a societal level. As noted already by Storey (1991), this, however, is a rather complicated task. Moreover, only recently have scholars begun to try to find an empirical link between the performance of firms, measured in terms of growth, and the societal level (e.g. Thurik, 1999; Audretsch et al., 2002). In this context, Audretsch and Thurik (2002), for example, have identified that higher entrepreneurial activity leads to subsequent higher rates of growth for the economy as well as a decrease in unemployment. The 57

Global Entrepreneurship Monitor (Reynolds et al., 2000) also established an empirical link between the degree of entrepreneurial activity and economic growth, as measured by employment on a national level. Various other measures have been used to assess the contribution of entrepreneurship on a societal level. These include selfemployment rates, business ownership rates, and new firm start-ups, various measurements of industry demography, including the extent of simultaneous birth and exits as well as net entry. Although useful for various purposes, these measurements have often been criticised for various reasons. Self- employment, for example, has received criticism for its lack of projection of the impact on national and global markets while business ownership rates have been criticised for treating all business, such as high tech and low- tech, the same way (e.g. Audretsch, 2002). Several recent empirical studies provide evidence that even subjective perceptions have a strong influence on the actual behaviour of firm owners in terms of their relation to the firm’s growth. Using an extensive data set collected over a 10year period from 3 different studies, Wiklund et al (2003), for example, conclude that entrepreneurs’ beliefs and attitudes play an important role in understanding why they act in a particular way. The authors find that beliefs concerning the consequences of growth may influence entrepreneurs’ overall attitude toward growth (ibid). Furthermore, a study by Watson et al. (2003) has indicted that perceptions of performance may actually be more insightful indicators than objective measures because perceptions draw on insiders’ knowledge of a firm’s goals, strategy, structure and processes. In the scope of assessing societal-level performance, if perceptions can be linked with venture-level performance, one could argue that they might have the same or a similar effect on the contribution of companies to society. As argued by Davidsson (2005), we need an innovative approach for gathering data when we face problematic issues. This is very relevant to the assessment of ‘total entrepreneurial outcome’ on a societal level. Considering the difficulties in using various ‘objective’ measures, in the context of the theoretical framework, this thesis focuses on the perceptions of small firm owners’ contribution to society in order to develop the measurement scale for assessing societal-level output. For similar reasons as in the case of venture-level indicators, both short and long-term perspectives should be considered when SME output is addressed on a societal level. Drawing on the existing empirical evidence regarding the contribution of entrepreneurship on a societal level described previously in this section, the following classification for the assessment of 58

societal-level output is suggested in the scope of addressing productive, unproductive and destructive entrepreneurship (see Table 5). Table 5. Operationalisation of societal-level output for measurement Perceptions about the contribution to society: short and long-term dimensions Generation of indirect employment Innovations in products or services Bringing new technologies to a country Attracting investments from abroad Educating employees Investing in infrastructure Sponsorship for social needs Paying taxes Knowledge spillovers

Source: Author 3.4. Towards a measurement of activities’ antecedents and influences: expectations and goals As highlighted in the previous chapter, in the scope of this thesis the influence of motivation on activities is limited by focusing on entrepreneurs’ attitudes towards goals as well as the expected consequences of their business activities. It was also emphasized that motivation is viewed as venture-level phenomena, influencing the outcomes of entrepreneurship activities and determining types of activities SMEs become involved in (e.g. Wiklund, 1998). In this light, since the venture-level output measurements used in this thesis are mostly based on indicators representing the expansion and growth of a company, including the growth of profits, turnover, employment and investments, it is obvious that expectations regarding output should also be assessed within the same framework. Additionally, drawing on strategic orientation literature, which distinguishes between growth-oriented firms, survivaloriented firms, exit-oriented firms and control-oriented firms (e.g. Stratos, 1990; Poutziouris, 2003), the dimension of SMEs’ attitudes towards goals is addressed. The operationalisation of expectations regarding output and attitudes regarding business goals for measurement is provided in Table 6. 59

Table 6. Operationalisation of expectations and goals for measurement Expectations of SMEs regarding: Generation of profits

SMEs’ goals:

Turnover

To survive

Generation of employment

To enhance control

Investments

To exit

To grow

Source: Author 3.5. Towards a measurement of environmental influences and access to capital As argued in the previous chapter, the influence of context on SME value creation in the scope of this thesis is addressed drawing on the extended institutional approach introduced by Aidis and Sauka (2005) and includes four categories: formal, informal, economic and other factors. In this light, Aidis and Sauka (2005) furthermore include a classification and analysis of environmental influences on entrepreneurship development during different stages of market reform in transition countries. More specifically, by exploring possible environmental influences in the form of constraints to development as perceived by SME owners or owner-managers, Aidis and Sauka (2005) first of all attempted to classify these constraints into formal, informal, economic or other categories. The specific areas and issues classified under each of the four categories as derived from the existing literature on barriers to SME development in transition countries are presented in Table 7. In this light, it is highlighted by Adis and Sauka (2005) that there is, however, no clear distinction between formal, informal, economic and other categories; they overlap, and this classification should be viewed as an approximation of various types of contextual influences in a transition setting. Altogether, however, these constraints can be seen as major contextual influences on the entrepreneurship process, capturing various dimensions expressed as formal, informal, economic and other factors affecting SME value creation. Aidis and Sauka (2005) then proceeded to analyse which formal, informal, economic and other barriers are the main influences on SME development and growth during various stages of transition. Here, transition country data was derived from a systematic analysis of previously available surveys on the barriers to SME

60

development in 23 transition countries. The selected 36 studies on constraints facing SMEs cover various transition stages in 23 different transition countries.18 Table 7. Classification of barriers to SME development and growth: formal, informal, economic and other environmental influences: Formal

Informal

Economic

Other

High taxes Government intervention in the economy Laws and regulations (policy instability/uncertainty, frequent changes, nontransparency, unpredictability, operation, customs and trade regulations, accounting standards, business registration) Business inspections High social security payment Poorly functioning state (implementation of business regulations, problems arising from administration as well as lack of personnel) Lack of state support (in terms of building appropriate infrastructure with arising constraints as information provision to small-firm employees, the need for specific consulting advice (marketing, financial, psychological) and lack of SME policies) Financing structure

Payment behaviour of clients Rent-seeking through inspections Motivation of the workforce/quality ethics in the workforce Bureaucracy Poorly functioning state (business security issues such as inadequate measures against organized crime, mafia) Implementation of business regulations (as a consequence of responsible officials attitudes) Lack of state support (represented by attitude of the government towards business)

Macroeconomic stability (inflation) Physical infrastructure – old unreliable equipment Shortage of qualified workers Shortage of high-quality managers Cost of loans, interest rates, collateral requirements Loan application Unfair competition Strong (domestic and foreign) competition Low product demand High input prices Suppliers not ready to deliver (procurement problems) Business training

Inability to expand into new markets

Source: Aidis and Sauka (2005) Furthermore, Aidis and Sauka (2005) identified the main environmental influences, e.g. constraints to development, at various stages of transition. Since this thesis focuses on SME value creation during the advanced transition stage, in order to classify formal, informal, economic and other environmental influences for measurement, it is of particular interest to understand which barriers influencing SME development and growth are most important during advanced transition. The 18

See Aidis and Sauka (2005) for further description of the studies and methodology.

61

main constraints – formal, informal, economic and others relevant for an advanced transition setting as found by Aidis and Sauka (2005) – are presented in detail in Table 819. These are barriers found to be of major importance by at least one of 36 studies assessing SME constraints during the advanced stage of transition. Table 8. Major barriers specific to the advanced stage of transition Formal

Informal

Too many licenses

Too

Accounting standards

inspections

Information provision to

Implementation

small firm employees

business regulations

Need

for

consulting (marketing,

specific advice financial,

Economic many

tax of

Shortage

Other of

Business

workers/labour

expansion

Strong competition

new markets

Lack

into

of

investment/finance for expansion Inflation

psychological) Lack of state support

Source: Aidis and Sauka (2005) These barriers, found to influence SMEs during advanced transition, are used as a background for constructing a measurement scale for the assessment of environmental influences on the relationship between activities and SME value creation within an advanced transition setting. The measurement scale for the assessment of the access to financial capital, however, was also derived from existing studies (e.g. Wiklund and Shepherd, 2005) referring to whether or not SMEs experience financial constraints in the form of lack of capital for business investments.

19

See Aidis and Sauka (2005) for further information and an analysis of SME constraints during all three transition stages.

62

4. Methodology

Drawing on the identification of the main elements, e.g. EB, EO, SME value creation, environmental influences and access to capital, as well as the conceptualisation of the relationship between them, which in turn resulted in the conceptual framework for the assessment of directly and indirectly productive, unproductive and destructive entrepreneurship in Chapter 2, the previous chapter of the thesis dealt with the operationalisation of these concepts for measurement. However, in order to address the above-mentioned concepts empirically, thus gaining an understanding of the factors influencing directly and indirectly productive, unproductive and destructive entrepreneurship focused, systematic and rather specific data is necessary, in order to link both the determinants of and the influences on SME value creation. In this context, as argued in the previous chapter of the thesis, new measures have to be developed to empirically assess aspects of SME involvement in directly and indirectly productive, unproductive and destructive entrepreneurship. Moreover, referring to previous studies such as Wiklund and Shepherd (2005), in order to address the relationship between SME activities and output, longitudinal data is needed. That is, cross- sectional data do not infer causality, but time is necessary for activities to have the effect on outcomes, thus independent variables should be assessed one at time and dependants at least a year later. Since such data is not included in existing databases it was necessary to design a survey and collect our own data. In this light, this chapter deals with methodology issues related to the empirical assessment of directly and indirectly productive, unproductive and destructive entrepreneurship. More specifically, drawing on the operationalisation for the measurement of the above-mentioned concepts introduced in the previous chapter, this chapter aims to introduce a measurement scale for the empirical assessment of these concepts. Issues such as survey design, the data collection process, sample characteristics and sample representativeness as well as non- responses and biases are also addressed here. Since hypotheses addressing the relationship between activities and SME value creation as developed in the Chapter 2 are based on the design used by Wiklund and Shepherd (2005), the authors’ suggestions for the necessity of longitudinal data are considered in the thesis by addressing most dependent variables and controls with the

63

survey conducted in 2005 and dependent variables as well as some of the independent variables with a follow-up survey a year later, in 2006. A more specific description of both surveys and the data collection process is provided in the next section of the chapter. Section 4.3 deals with the questionnaire form and measurement scales for the empirical assessment of directly and indirectly productive, unproductive and destructive entrepreneurship. The response rate and item non- responses are addressed in Section 4.4. Validity issues, such as controlling for biases, are discussed further in Section 4.5, whereas sample characteristics and representativeness is presented in section 4.6. The chapter concludes with a description of the approach for the analysis of the data in Section 4.7 4.1. Description of the survey method The choice of survey design for the assessment of directly and indirectly productive, unproductive and destructive entrepreneurship was determined by the nature of the study itself. More specifically, one of the biggest challenges in the design stage of the empirical study was assessing the degree of involvement in deviant behaviour. Obviously, the biggest problem in this regard arises from the sensitive nature of the questions asked of the respondents. That is, entrepreneurs are usually reluctant to talk about the ‘dark sides of their businesses’ such as underreporting taxes or envelope wages. Considering the sensitive nature of questions addressing the degree of involvement in deviant behaviour, it is not likely that entrepreneurs will answer these questions if they are asked through a standardised questionnaire form sent by mail, for instance. A qualitative approach, such as in- depth interviews involving face- to- face contact with a respondent, would be more appropriate in this regard. In general, however, qualitative approaches do not provide enough scope to generalise findings, which is not what the empirical model of this thesis intends to communicate. A combination of both ‘quantitative’ and ‘qualitative’ approaches was chosen in order to solve the problem. Namely, in the first round of empirical data collection, 133 face-to-face interviews were conducted in the form of conversation, yet based on the list of standardised questions. The interviewees were randomly sampled and the interviews were carried out during May- June 2005. The sampling frame was defined as

64

including owners and owner-managers of SMEs20 who have the right to sign the business documents of their enterprises and whose firms are registered in the capital city of Latvia, Riga, and were operational at the time of the survey. Data on companies was obtained from the official statistics of the Latvian Company Register, compiled in the Lursoft (www.lursfot.lv) database. Latvia was chosen due to practical reasons but also because it provides an excellent example of a transition country that has successfully transformed its status from a centrally planned Soviet republic to a fast-growing, sovereign, market-oriented and democratic state. According to Aidis and Sauka (2006 : 3): “Latvia’s ability to implement the institutional and economic requirements for a market-based economy as well as its recent successful accession into the European Union positions it in the advanced stages of transition”, thus providing a good scope for measuring environmental influences on the entrepreneurship process. Riga, the capital city of Latvia, was chosen as almost 1/3 of the entire population of Latvia lives there, and nearly 60% of its companies are registered in Riga. In this context, however, one should be aware that although the results may reflect major trends of factors and influences determining productive, unproductive and destructive entrepreneurship in Latvia, some caution should be taken in generalisations as certain differences may exist in different regions of Latvia. Further, considering the large proportion of Russian-speakers in Riga and Latvia in general, interviews were carried out in either Latvian or Russian, according to the preference of the respondents. Face-to-face interviews involved two phases:

telephone contact, then the

interview. In the first step, respondents were introduced to the aim and nature of the survey. A summary of results was offered as a non- monetary incentive. In addition, interviewers also received brief training. The main emphasis here was on the strictly confidential nature of the survey as well as the method and sequence of asking questions. As a result 550 contacts were made out of which 142 agreed to interviews, thus resulting in a response rate of over 25 percent. Four of those did not appear because of a lack of time. Another five were excluded as they were obviously neither owners nor owner/managers. Lack of time and unwillingness to participate in any surveys as well as disbelief that ‘this will change anything’ were the main reasons why SME owners or owner- managers refused to participate in the interviews, 20

SMEs are defined according the EU definition: less than 250 employees.

65

although it was explained to them how important their participation in the survey would be during the first phone call. On average, interviews lasted 100 minutes, with a minimum of one and a maximum of four hours. Interviews started with simple, easy-to-answer questions, such as individual and venture-level characteristics. During this phase, interviewers were instructed to gain some level of trust. The conversation form of the interviews also allowed them to choose ‘the right time’ to move on to the sensitive part of the questionnaire form. The experience from the interviews was that sometimes this ‘gaining of trust’ process could take more than an hour, so it was very necessary to keep the respondent interested in the conversation. In order not to lead respondents, most questions were asked openly and coded afterwards by the author. Face-to-face interviews were carried out in order to collect data on most of the independent variables, such as EB, EO and environmental influences. Considering the necessity for longitudinal study, data for the empirical assessment of determinants of and influences on directly and indirectly productive, unproductive and destructive entrepreneurship after one year, the same respondents were contacted again in August 2006. This was done with the intention to mostly to address the dependent variables, including the performance of firms on venture and societal levels, both short and long term. Since we were already in contact with the respondents and the nature of the questions included in the follow-up study was, comparably, far less sensitive, phone interviews were chosen as the method for data collection. Phone interviews were implemented by the author of the thesis himself. Over the period of one year, 3 companies had gone out of business and for 6 of the companies we had contacted a year ago, the owners or owner- managers had changed. The author, however, decided to keep these 6 firms in the sample. As a result, the final sample consists of 130 SME owners and owner managers. Prior to each of the rounds, e.g. face- to- face and phone interviews, questionnaire forms were tested in pilot studies (5 interviews and 5 phone interviews), as a result of which a few changes were made to the questionnaire forms. Building on the classification as introduced in the previous chapter of the thesis, considerable time and effort was also spent on developing appropriate measures and designing the questionnaire forms to assess both independent and dependent variables before the data collection process was initiated. The next section of this thesis deals with this issue further. 66

4.2. Questionnaire forms and measures The questionnaire form for the face-to-face interviews is made up of 63 questions, whereas the follow-up study questionnaire form consists of 23 questions (see Appendixes 1 and 2 accordingly). The main question clusters include individual and venture-level characteristics, SME value creation, e.g. output on venture and societal levels in the long and short term, EB, EO, environmental influences expressed in the form of formal, informal, economic and other constraints to business development, as well as various control questions to increase the validity of the often subjective and indirect methods used in the study. 4.2.1. Measurement of EB The construct for the measurement of the degree of involvement in deviant behaviour consists of 13 questions. These questions were included in the first round of data collection, e.g. face- to- face interviews, and capture various aspects of each of the three dimensions operationalised in the previous chapter of the thesis (see Table 3). More specifically, in order to address rent seeking, respondents were asked to estimate the degree of underreporting business income, salary, the number of employment as well as involvement in informal payments, e.g. bribes. A list of various types of unofficial payments was then presented to the respondents, who were asked to estimate the degree of involvement in each type of unofficial payment, such as, for instance, involvement in unofficial payment for getting licences, permits or dealing with tax inspection. Finally, the respondents were asked to address the degree of involvement in various forms of unethical behaviour, including crowding-out efforts and exploiting business ideas introduced by other firms without any compensation. Similar 5 or 6-item scales were used, where 1 represents a ‘low level of conforming behaviour’ and 5 or 6 a ‘high level of conforming behaviour’. Factor analysis was used to extract sets of variables for the empirical assessment of the degree of involvement in conforming or deviant behaviour. The final factor analysis was run with 11 variables; thus, a few items were dropped because of an unclear loading pattern with side-loadings above 0.30. The author used a principal component analysis and there was no need to extract factors, since the

67

number of factors with eigenvalues over 1 seemed to be optimal. Varimax rotation was also used and missing variables excluded pair-wise. The rotated component matrix is displayed in Table 9. Table 9. Rotated component matrix: assessment of degree of involvement in deviant behaviour Component 1

Underreporting business income Hiding actual salary Lobbying and litigation, including bribes Underreporting number of employees Payments to influence the content of new laws/ regulations Payments to ensure connection to public services Payments to obtain licences and permits Payments to deal with customs/ imports Exploiting others’ business ideas Crowding-out efforts Using old networks (blats)

2

3

,808 ,761 ,666 ,595 ,763 ,695 ,586 ,565

,342

,742 ,632 ,628

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalisation. Rotation converged in 4 iterations.

Source: Author’s research To determine the empirical validity of extracted multi- item variables, Cronbach’s alpha test for multi- item variables, their stability in different samples and relative importance was used. It is argued that Cronbach’s alpha values should usually exceed 0.5, although no common agreement exists in this regard (Davidsson, 2004). As a result, the following measures were derived for assessing the degree of involvement in deviant behaviour (see Table 10). Since Cronbach’s alpha level for the multi- item variables ‘unethical behaviour’ is very low (e.g. .386), these variables will subsequently be included in the analysis only as separate items, not as a summated index.

68

Table 10. Measures for assessing conforming and deviant behaviour in a transition setting Name of the construct

Variables included in the questionnaire form21

1. Summated index of ‘Rent-seeking’

Underreporting business income Underreporting number of employees

Cronbach’s alpha value for multi- item variables .682

Hiding actual salary Lobbying and litigation, including bribes 2. Summated index of ‘Unofficial payments’

Payments to influence the content of new laws/ regulations Payments to ensure connection to public services

.534

Payments to obtain licences and permits Payments to deal with customs/ imports 3. Summated index of ‘Unethical behaviour’

Exploiting others’ business ideas Crowding-out efforts Using old networks (blats)

.386

Source: Author’s research, based on Sauka and Welter (2007a) 4.2.2. Measurement of EO As with the assessment of the degree of conforming and deviant behaviour, questions addressing EO were included in the first round of data collection. As already highlighted in the previous chapter of the thesis, however, EO is a well researched concept and thus it is possible to use already validated measurement scales for the empirical assessment of various EO dimensions. Therefore, the author used the original 7-point scale as developed by Miller (1983) and further developed by Lumpkin and Dess (1996), including the five dimensions of EO: innovativeness, proactiveness, competitive aggressiveness, risk-taking and autonomy. Here 1 represents the lowest degree of EO, whereas 7 mean the highest level of involvement in EO. To avoid response-set contamination, the questions were arranged so that the entrepreneurial and non-entrepreneurial statements appeared on both the right and left sides. Again, as with EB, a factor analysis was run and Cronbach’s alpha test was used to extract and validate these measures. The final factor analysis was run with 13 variables representing entrepreneurial orientation. A few items were dropped because 21

Variables were extracted from the operationalisation of EB for the measurement in Table 3.

69

of an unclear loading pattern with side loadings well above 0.30. The author used a principal component analysis and extracted 5 factors. Varimax rotation was also used and missing variables excluded pair-wise. The rotated component matrix is displayed in Table 11. Table 11. Rotated component matrix: assessment of entrepreneurial orientation Component 1 The firm is very seldom the first business to introduce vs. very often the first business to introduce new products/services, administrative techniques, operating technologies, etc. The firm typically responds to action which competitors initiate vs. initiates actions. The firm has a strong tendency to `follow the leader' vs. a strong tendency to be ahead of other competitors in introducing new products or ideas. In the past 3 years, changes in product or service lines marketed by the firm have been mostly of a minor nature vs. have usually been quite dramatic. In the past 3 years my firm marketed no vs. very many new lines of products or services. The firm favours a strong emphasis on the marketing of tried and true products or services vs. emphasis on R&D, technological leadership, and innovations. Top managers of my firm believe that owing to the nature of the environment, it is best to explore it gradually via careful, incremental behaviour vs. bold, wide-ranging acts are necessary to achieve the firm’s objectives. The firm prefers to study a problem thoroughly before deploying resources to solve it vs. is quick to spend money on potential solutions if problems are holding us back. The firm typically adopts a cautious posture to minimize the probability of making costly decisions vs. bold, aggressive posture in order to maximize the probability of exploiting potential opportunities. The firm ensures adequate coordination to minimize ineffectiveness and duplication of efforts vs. typically adopts a ‘let things go their own way’ strategy (reverse). The firm is quick to implement necessary structural changes in order to stimulate new ideas vs. adopts a ‘the structure of the company will adapt to the new ideas itself’ posture (reverse). In dealing with its competitors, the firm is very aggressive and intensely competitive vs. makes no special effort (reverse). In dealing with its competitors, the firm typically seeks to avoid competitive clashes, preferring a `live-and-let-live' posture vs. adopts a very competitive ‘undo-the-competitors' posture.

2

3

4

5

,853 ,812 ,785 ,860 ,791 ,739

,827

,821

,664

,886

,875

,874

.744

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Rotation converged in 6 iterations.

Source: Author’s research 70

As a result the following measures were derived for the empirical assessment of the degree of involvement in deviant behaviour (see Table 12): Table 12. Measures for assessing EO Name of the construct

Variables included in the questionnaire form

Cronbach’s alpha value for multi- item variables

1. Summated index of In the past 3 years, changes in product or service lines marketed by the firm have been mostly of a minor ‘Innovativeness’

nature vs. have usually been quite dramatic. In the past 3 years my firm marketed no vs. very many new lines of products or services. The firm favours a strong emphasis on the marketing of tried and true products or services vs. emphasis on R&D, technological leadership, and innovations. 2. Summated index of The firm is very seldom the first business to introduce vs. very often the first business to introduce new ‘Proactiveness’ products/services, administrative techniques, operating technologies, etc. The firm typically responds to action which competitors initiate vs. initiates actions which competitors then respond to. The firm has a strong tendency to `follow the leader' vs. a strong tendency to be ahead of other competitors in introducing new products or ideas. 3. Summated index of Top managers of my firm believe that owing to the nature of the environment, it is best to explore it ‘Risk-taking’ gradually via careful, incremental behaviour vs. bold, wide-ranging acts are necessary to achieve the firm’s objectives. The firm prefers to study a problem thoroughly before deploying resources to solve it vs. is quick to spend money on potential solutions if problems are holding us back. The firm typically adopts a cautious posture to minimize the probability of making costly decisions vs. bold, aggressive posture in order to maximize the probability of exploiting potential opportunities. 4. Summated index of Firm ensures adequate coordination to minimize ineffectiveness and duplication of efforts vs. typically ‘Autonomy’ adopts a ‘let the things go their own way’ strategy (reverse). The firm is quick to implement necessary structural changes in order to stimulate new ideas vs. adopts a ‘the structure of the company will adapt to the new ideas itself’ posture (reverse). 5. Summated index of In dealing with its competitors, the firm is very aggressive and intensely competitive vs. makes no ‘Competitive special effort (reverse). aggressiveness’ In dealing with its competitors, the firm typically seeks to avoid competitive clashes, preferring a `live-and-letlive' posture vs. adopts a very competitive ‘undo-thecompetitors' posture.

.767

.835

.745

.763

.569

Source: Author’s research, based on Miller (1983) and Lumpkin and Dess (1996),

71

4.2.3. Measurement of SME output on a venture level In the previous chapter of the thesis, short and long-term venture-level output measures were operationalised as capturing both monetary dimensions (e.g. expansion and growth of SMEs in terms of profits, turnover, employment generation and investments) as well as non- monetary incentives, including self-fulfilment from the involvement in entrepreneurship as such (see Table 4). Drawing on this operationalisation, in order to derive a measurement scale for the assessment of venture-level output indicators, the following questions were included in the questionnaire form. First, to address the monetary dimension, respondents were asked whether net sales profits, sales turnover, employment, investments (including investments in production, training, etc.) and export share increased, decreased or remained stable in comparison to the situation 12 months ago (short term) or 3 years ago (long term)? Each of these variables was then measured on a five- item scale: decreased a lot, decreased, remained stable, increased or increased a lot. Additionally, one question was included to assess the non- monetary aspect of venture-level performance: “Regardless of the success of your company’s performance or your future plans regarding growth, has your last 12 months (short term) or 3 years (long term) of business activity brought personal fulfilment (satisfaction) to you?” Personal fulfilment was also measured on a five- item scale, where 1 represented ‘not at all’ and 5 represented ‘yes, to a very great extent’. The measurement scale for the assessment of short and long-term venturelevel output indicators was derived by conducting a reliability analysis, e.g. testing the validity of these measures using the Cronbach alpha test for multi- item variables. As a result, the following measures were developed (see Table 13). The growth of export share was not included in the scale in the short or long term since most of the respondents are not involved in exporting activities.

72

Table 13. Measures for assessing venture-level outcomes: short and long term Name of the construct

Variables included in the questionnaire form

Increase or decrease in profits compared to 12 months ago Increase or decrease in turnover compared to 12 months ago Venture-level output in Increase or decrease in employment the short term compared to 12 months ago Increase or decrease in investments compared to 12 months ago Self-fulfilment from involvement in business compared to 12 months ago Increase or decrease in profits compared to 3 years ago Increase or decrease in turnover compared to 3 years ago Venture-level output in Increase or decrease in employment the long term compared to 3 years ago Increase or decrease in investments compared to 3 years ago Self-fulfilment from involvement in business compared to 3 years ago

Cronbach’s alpha value for multi- item variables

.885

.927

Source: Author’s research 4.2.4. Measurement of SME output on a societal level Entrepreneurship literature suggests using measurement scales which have already been validated by previous research rather than developing original ones (e.g. Wiklund, 1998). The main reasons for this are that these have already been tested and it is possible to compare the results with results from previous surveys. Nevertheless, as argued in the previous chapter of the thesis, capturing the ‘total’ societal-level output by SMEs has proven to be a very challenging task and no consensus exists in this regard in entrepreneurship literature. As in the case of EB, the author thus chose to develop an original measurement scale for the empirical assessment of firms’ societal-level output. As argued in the previous chapter of the thesis, the measurement of societallevel output was operationalised by drawing on the subjective perceptions of entrepreneurs of their contribution to society. Various dimensions, often highlighted as the specific contribution of SMEs to economic growth, such as innovation in products or services and the generation of indirect employment, were included in this

73

classification scale (see Table 5). Drawing on this operationalisation, the following question was included in the questionnaire form: “In the last 12 months (short term) and 3 years (long term), do you believe that your business has contributed to the growth of the Latvian economy with regard to the following”. Further, dimensions as classified in Table 5. (e.g. indirect generation of employment, education of employees according to economic requirements, innovations in products or services, bringing new technologies to the country, attracting investments from abroad, investing in the infrastructure (roads, electricity, communications), sponsorship for social needs (by paying taxes, generating knowledge spill-overs), were included as possible answers. Entrepreneurs’ perceptions of their contribution to various dimensions of society were measured on a five- item scale, where 1 represents no ‘contribution at all’ and 5 means contribution ‘to a very great extent’. As in the case of venture-level output, Cronbach’s alpha test was used to test for the validity of using multi- item variables. As a result, the following measures were developed (see Table 14). Table 14. Measures for assessing societal-level outcomes: short and long term Name of the construct

Societal-level output by SMEs in the short term (12 months) and long term (3 years)

Variables included in the questionnaire form. Perceptions about contribution regarding: Indirect generation of employment Educating of employment according to economic requirements Innovations in products or services Bringing new technologies to the country

Cronbach’s alpha value for multi- item variables

.768

Attracting investments from abroad Investing in the infrastructure Sponsorship for social needs

.772

By paying taxes Generating knowledge spill-overs

Source: Author’s research 4.2.5. Measurement of environmental influences and access to capital

74

The measurement scale for the empirical assessment of environmental influences on SME value creation is based on the operationalisation of formal, informal, economic and other constraints to SME development found to be of the greatest importance within an advanced transition setting (e.g. Aidis and Sauka, 2005, see Table 8). In this light it can be argued that if it were possible to define more exact cut- offs among various institutions, e.g. formal, informal, economic and other types, the author would prefer to run a factor analysis and attempt to distinguish between sets of variables conforming to each of these categories (see also q46 in Appendix 1). Since such cut- offs cannot be clearly determined (Aidis and Sauka, 2005), it is more appropriate to use a multi- item variable capturing all four dimensions as represented in Table 8. Drawing on this idea, the measurement scale for addressing contextual influences is displayed in Table 15. Table 15. Measures for assessing contextual influences Name of the construct

Variables included in the questionnaire form

Taxes are too high Inconsequent business legislation Implementation of business legislation by the government as a consequence of poor administration and lack of personnel Lack of specific consulting advice on Contextual influences on marketing, financial, and psychological directly and indirectly issues productive, unproductive Too many tax inspections and destructive entrepreneurship within Attitude of Latvian government towards business in general an advanced transition Governmental corruption setting Heavy competition from legal business Low purchasing power Shortage of qualified workers Shortage of qualified managers Inflation

Cronbach’s alpha value for multi- item variables

.647

Source: Author’s research, based on Aidis and Sauka (2005) Influences of context, e.g. formal, informal, economic and other barriers, on the involvement in directly and indirectly productive, unproductive and destructive entrepreneurship are thus empirically determined by the summated index of 12 barriers found to be of influence during the advanced stage across transition countries. By asking ‘what are the major barriers to your business development?’, environmental 75

influences are measured using a five- item measurement scale, where 1 represents a dynamic (hostile) and 5 a stable environment. Further, drawing on Wiklund and Shepherd (2005), considering the importance of addressing access to capital in the relationship between EO and performance, one more question was included in the questionnaire form to empirically address the ‘access to capital’ issue. More specifically, respondents were asked whether or not they experience financial constraints, e.g. lack of funds for investments. The five- item measurement scale was also used to measure the constraints as regards the access to capital, where 1 represents low access to capital and 5 represents high access to capital. The question addressing access to capital was included in the first round of data collection (see q46 v135 in Appendix 1). 4.2.6. Measurement of expectations and goals, and controls The measurement scale for addressing SMEs’ expectations from involvement in business activity were based on the operationalisation introduced in Table 6. More specifically, the following retrospective question was included in the follow- up survey: ‘What were your expectations regarding your business growth 12 months ago (short term) and 3 years ago (long term)?’ The measurement scale for expectations is similar to the one used for the assessment of venture-level output, e.g. the ‘expansion and growth’ dimension. That is, entrepreneurs were asked to rank expectations regarding growth for profits, turnover, employment generation, investments and export share. A five-item measurement scale was used, 1 representing expectations regarding ‘decrease on a large scale’ and 5 representing ‘increase on a large scale’. Since export share was excluded in the case of venture-level output assessment, the same choice was also made here. It should furthermore be highlighted, that although the question addressing SME expectations captured both short and long-term dimensions, in the end only short-term dimensions are addressed in the thesis. This is due to the reason that both the pilot study and the main data collection proved that respondents could not answer this question for the long term. The measurement scale for the assessment of SMEs’ expectations regarding growth and expansion is displayed in Table 16.

76

Table 16. Measures for assessing SMEs’ expectations regarding growth and expansion Name of the construct

Variables included in the questionnaire Cronbach’s alpha form. Expectations regarding increase or value for multi- item decrease in the following indicators 12 variables months ago: Expectations regarding Profits growth and expansion Turnover .919 Employment Investments

Source: Author’s research The following retrospective question was included in the follow-up study in order to capture SMEs’ attitudes towards goals: “What have been your main business goals for the past 12 months?” This time respondents could choose only one of the following alternatives: to grow, to survive, to provide a living for my family, to improve personal household income, and to prepare the firm for sale (see q20 v57 in Appendix 2). Furthermore, numerous questions were included in the survey in order to address the additional influences determining directly, indirectly productive, unproductive and destructive entrepreneurship (e.g. controls). More specifically, to determine the age of a firm, respondents were asked in which year their companies started operation, which was used to calculate the firm’s age. The size of the firm (number of employees, full-time equivalent) was calculated for the response to the question, ‘What is the total number of employees including yourself employed in your enterprise at the present time?’. Respondents were then asked about their education level. Finally, to determine the industry, respondents were then asked whether the company’s main activity is manufacturing, service or retailing. All controls (see Table 17) were included in the first-round questionnaire form, however the author also checked for possible changes in these indicators in the follow-up survey. A list of control questions is displayed in Appendix 1 (v1 (age of firm), v56A (size of firm), v162 (gender), v156 (education), v20 (sector)).

77

Table 17. Control variables Variable Age of firm Size of firm Education Gender Industry

Measurement Continuous variable (year from the start of operation) Continuous variable (number of employees, full-time equivalent) Dummy variable. 1= higher education, 0 = otherwise. Dummy variable. 0 = Female, 1 = Male. Dummy variables. 1= Manufacturing, 0 = otherwise. 1= Service, 0 = otherwise. 1 = Retail, 0 = otherwise.

Source: Author’s research

4.3. Response rate and item non- response As already mentioned in Section 4.1 of this chapter, the first round of data collection achieved a response rate of just above 25 percent. Further, in the follow-up study only 3 out of 133 companies contacted in 2005 were dropped, which does not significantly influence the overall response rate of the study. In the context of the empirical assessment of directly and indirectly productive, unproductive and destructive entrepreneurship, however, exploring single-item non-response can be of more importance than overall response rate. This is due to the fact that the questionnaire method, mainly face-to-face interviews, involves a number of very sensitive questions and thus, even if they agreed to the interview, respondents might not be willing to answer some particular questions. Considering the relatively small sample, this can influence the validity of the results and should thus be addressed before the characteristics of the sample, representativeness and the results themselves are discussed. Certain steps were thus taken during both the preparation stage of data collection and the data collection process to increase responses to each of the questions. As already mentioned in Section 4.1, interviewers were instructed to start the interview with ‘easy-to-answer’ questions and establish trust prior to moving on to more sensitive questions, including involvement in deviant behaviour. In this light, and mostly due to the sensitive nature of the questions, interviewers themselves also had to meet some formal criteria to be involved in the data collection process. That is, 78

persons from the Stockholm School of Economics in Riga (SSE Riga) with previous business management experience in Latvia were employed (e.g., the author of the thesis and two students in their final year at SSE Riga). These two criteria are important since SSE Riga has a very good reputation in Latvian society, which increased respondents’ trust regarding the confidential nature of the interviews. Business management experience in Latvia or another transition country, on the other hand, was necessary in order to be able to ‘speak in the respondents’ language’ as well as share experience as regards various forms of entrepreneurial behaviour within a similar environment. The author believes that such a selection can contribute to the reliability of obtained data, if not to the overall ability to obtain data, and this was actually confirmed by the survey. Arguably the choice of interviewers, the sequence of asking questions and the survey design as such could all ensure a higher response rate. Apart from this, in order to reduce item non- response, it is also important how questions are asked. For example, even if experienced interviewers are hired, following the above-mentioned ‘rules’, respondents might still not be willing to answer questions such as, ‘How often are you involved in tax evasion?’ or ‘How often have you been involved in unethical behaviour during the last year?’ One way to overcome this problem is by asking the same questions indirectly, referring to the industry or a ‘company such as yours’ instead of the respondent – for example, ‘How often are companies in your industry involved in tax evasion?’ In such a way some previous studies (such as Aidis, 2006) have used such indirect methods to capture various forms of ‘deviance’ in organisations. Still, when it comes to the empirical assessment of SMEs’ involvement in deviant behaviour, one cannot be confident that the general situation ‘in my industry’ or ‘firms such as mine’ can be attributed to the particular respondent. In order to link behaviour and outcomes, however, for this study it is of major importance that questions asked indirectly be attributed to the respondent or respondent’s company, not the industry in general, for instance. This issue is of particular importance, since most of the questions asked in order to address SMEs’ involvement in deviant behaviour during face-to-face interviews in 2005 (see Appendix 1: q 28, q 29, q29a, q 31, q32, q38, q39, q41), were ‘indirect questions’. Namely, a pilot study and common sense suggested that if we want to get answers at all, some questions simply could not be asked directly, especially during the first contact with the respondent. 79

Although most sensitive questions were asked indirectly, a number of subsequent steps were taken to increase the probability that answers received could be attributed to the firms in question. In such a way item non- response was reduced and the validity of the study increased at the same time. Namely several ‘control’ questions addressing ‘direct’ vs. ‘indirect’ assessment of involvement in deviant behaviour were included in the follow-up survey. More specifically, at the end of the phone interview (e.g. the follow-up study in 2006), each respondent was reminded that the year before we had talked about issues such as the underreporting of employment and income or unofficial payments. In this light, respondents were asked to assess the general situation regarding these issues now in comparison to the situation after one year. A five-item measurement scale was used, where 1 represents ‘in general, involvement in these types of activities has increased considerably’, 3 represents ‘the same as a year ago’ and 5 represents ‘in general, involvement in these types of activities has decreased considerably’. This question not only initiated the subsequent discussion about involvement in deviant behaviour but also shed some light on whether the situation in Latvia has changed over the period of a year. The results show that the majority, e.g. 82.3% of respondents state that the situation is the same as a year ago while 14.6% report that involvement in deviant behaviour has decreased and only 2.3% believe that involvement in deviant behaviour has decreased considerably. Further, in order to turn ‘indirect answers’ from face-to-face interviews in 2005 into direct answers, the following method was applied. In general, the same questions as one year ago were asked, only in a different manner (See Appendix 2: list of control questions for deviant behaviour). That is, respondents were ‘reminded’ that last year they stated that, for example, the degree of involvement in underreporting business income by their companies (not by ‘firms such as yours’ or ‘firms in your industry’) was, for example, ‘on a large scale’. Each respondent was then asked whether the situation is still the same this year and if not, to what extent it has changed, again offering a 5-item measurement scale (1: increased considerably; 3: the same as a year ago; 5: decreased considerably). In order not to make the phone conversation too long, the author addressed each respondent with only three (out of eleven) variables measuring the degree of involvement in deviant behaviour (e.g. first five respondents were addressed with variables 1-3, the second five respondents with variables 4-6, etc.). Although this approach does not provide a complete picture, it can 80

be argued that the number of responses is still sufficient to capture the general trend and thus decide whether or not ‘indirect’ answers from the 2005 survey can be treated as ‘direct’ answers. The main findings from these control questions suggest that although asked indirectly, in 2005 respondents reported on the situation in their company and the majority stated that the situation had not changed during this year either. In general, although the application of such a method could be seen as manipulation of the respondents, the author believes that the manipulation may be justified since it does not cause any harm and contributes to a decrease in item non- response as well as the general validity of the study. Additionally, such a methodology allows one to obtain a deeper understanding of the situation as regards the involvement of SMEs in deviant behaviour, providing a further scope to link entrepreneurial behaviour with SME value creation. 4.4. Survey biases In the case of this particular study, the responses from SME owners may be biased not only due to the sensitive nature of the questions, as addressed in the previous section of the thesis, but also for the reason that subjective methods were used to assess a number of variables. In this light, as already argued in the previous chapters of the thesis, subjective methods are often used in entrepreneurship research. A number of studies emphasise that the perceptions of SME owners often reflect reality, e.g. what they do (e.g. Davidsson, 1991). Moreover, subjective methods are sometimes preferred over objective methods since provide better insight into the research phenomena (e.g. Watson et al, 2003). Still, due to the relatively small sample, in order to increase the validity of this study, the author attempted to control for most of the subjective methods used in the thesis. These include subjective methods for the assessment of SMEs’ output and expectations. In this context, one should be aware, for instance, that scale used for the assessment of the monetary dimension of venture-level22 output can generate misleading results when two or more companies are compared. Namely, for one SME owner sales that have ‘increased a lot’ could mean, for example, a 15 percent 22

As discussed in Section 4.2.3., the monetary dimension (e.g. increase or decrease in net sales, turnover, employment generation and investments ) of venture level output was measured using a fiveitem scale: increased a lot, increased, remained stable, decreased or decreased a lot

81

increase, whereas another might perceive 15% as sales having ‘increased’ only. In order to control for these possible differences in perceptions of SME owners or owner-managers, additional questions were included with the aim of specifying the exact percentage of increase or decrease with regard to all of the mentioned venturelevel indicators. More specifically, respondents were asked to report the approximate increase or decrease for each of the mentioned venture-level indicators. An increase or decrease of more than 40 percent was then considered as ‘a considerable increase or decrease’, but less than 40 percent as simply an ‘increase’ or ‘decrease’. The second set of control questions was again used to control for the subjective methods referring to SME output, specifically addressing output on a venture level, short term. As already discussed, questions addressing SME outputs were included in the follow-up study, phone interviews during August 2006. Some questions measuring venture-level performance, however, were also included in the 2005 survey to control for the subjective methods regarding SME output. More specifically, in the 2005 survey respondents were asked to assess the overall performance of their business in 2004 (see q17 in Appendix 1) as successful, satisfactory, relatively satisfactory, relatively unsatisfactory or very unsatisfactory. Using the same measurement scale, respondents were then asked the same question with regard to the overall performance in 2004 after a year (see q8 Appendix 2). In such a way, it is possible to compare how the same respondent judges the overall performance of his or her company when asked repeatedly (e.g. in 2005 and after one year in 2006) and to see whether there is consistency in this evaluation. Arguably, consistency could mean higher validity for the subjective perceptions regarding the performance of SMEs. In this light, it should be noted, however, that although such control questions can be used for this purpose, the results should be interpreted with some precaution. That is, SME owners may evaluate the level of performance in 2004 differently in 2005 and 2006, e.g. in 2006 study respondents’ judgements are most likely to be influenced by the performance in 2005; thus, answers to the same question, asked a year later, can actually differ. Results, however, show that SME owners or owner-managers are consistent when evaluating the overall performance of their firms (see Table 18).

82

Table 18. Assessment of the overall output of SME owners or owner-managers

Successful Satisfactory Relatively satisfactory Relatively unsatisfactory Very unsatisfactory

Responses from the 2005 survey Assessment of overall SME output for 2004 Count 49 49 22 7 2

Responses from the 2006 survey Assessment of overall SME output for 2004 Count 48 52 23 5 2

Source: Author’s research Further, a number of questions were also included to control for validity with regard to certain discrete venture-level output indicators, such as turnover growth and employment growth. In this light, in the follow-up study in 2006 respondents were asked whether their venture-level output indicators, including turnover and the number of employees , had increased, remained the same or decreased compared to a year ago (q9 f60 and f61 in Appendix 2). In the 2005 survey, however, respondents were asked to provide the actual turnover in 2004 and the actual number of employees in 2004 (q 19 v52 and q20 v56a respectively in Appendix 1). Questions about the actual turnover and the number of employees with regard to the year 2005 in the 2006 survey were also asked (q25 f2006_v52 and q26 f2006_v56A respectively in Appendix 2). Here, one can compare the results from the 2005 and 2006 surveys with regard to the actual performance of SMEs (turnover and employment) for 2004 and 2005, that is, whether in 2004 turnover or the number of employees was the same, increased or decreased compared to the situation in 2005. Then, by looking at q9 f60 and f61 in Appendix 1, one can see whether the same pattern can also be observed here. In general, this approach again allows one to check whether SME owners or owner-managers are consistent when evaluating the performance and growth patterns of their firms not only in general, but also over the period of one year with regard to particular performance indicators. Again, the findings show considerable consistency, suggesting that subjective performance measures sufficiently depict the actual performance of SMEs. Further, since the conceptual framework developed in Chapter 2 also includes the impact of growth expectations on activities and output, the author also attempted to examine the link between expectations and growth reality. To address this issue a

83

number of questions were included both in the 2005 survey and in the follow-up survey in 2006. In 2005, survey respondents were asked what SME owners or owner managers’ expectations are with regard to profits, sales turnover, employment generation and investments for the next 12 months (q26 v80 –v83 in Appendix 1). These answers can then be compared with the reported growth of the same indicators after one year in the 2006 survey (q9 f2006_v60 – f2006 v63 in Appendix 2). The following results were acquired suggesting that, in general, growth expectations confirm to the real (reported) growth of firms (see Table 19). Table 19. Expectations of SME owners or owner-managers regarding growth as

Profits Turnover Employment Investments

2 5 1 6

count 6 6 1 2

81 86 53 60

count 76 78 49 53

34 28 67 47

count 34 31 61 64

9 7 5 6

count 14 15 18 5

0 0 0 1

Decreased a lot

Will decrease a lot

Decreased

Will decrease

Remained stable

Will remain stable

Increased

Will increase

Increased a lot

Will increase a lot

reported in the 2005 survey vs. actual growth as reported in the 2006 survey

count 0 0 0 0

Source: Author’s research Finally, in the 2006 survey respondents were asked what their expectations had been regarding profits, sales turnover, employment generation, investments and export share the year before (q14 f2006_v80C – f2006_v84C in Appendix 2) and whether they thought these expectations had been fulfilled (q15 f2006_new1 in Appendix 2). As mentioned previously, it was found that expectations in general reflect actual short-term venture-level output (see Table 19 above). The findings concerning SME owners or owner-managers perceptions of whether they have fulfilled these expectations are, however, slightly different, showing that, SME owners are rather pessimistic regarding the fulfilment of their expectations (see Table 20).

84

Table 20. SMEs owner-managers’ perceptions about the fulfilment of their growth expectations Completely More yes than no 50/50 More no than yes No Total Missing

Frequency 46 31 33 14 5 129 1

Percent 35.4 23.8 25.4 10.8 3.8 99.2 0.8

Source: Author’s research In general, both this section and the previous section of the thesis provide some evidence that although sensitive questions and subjective measures

were

included in the study, a number of attempts were made to reduce possible biases and thus increase the validity of the gathered data. When it comes to the analysis of the data, however, apart from these important issues, representativeness of the sample as such should be addressed to allow for the generability of findings. The next section of the thesis deals with this issue.

4.5. Characteristics and sample representativeness Small and medium-sized enterprises (SME), defined as having less than 250 employees, an annual turnover not exceeding 50 million EUR and a total balance sheet value of less than 43 million EUR, make up a large part of the Latvian economy. According to provisional data provided by the Latvian Company Register, there were 48 252 economically active enterprises in Latvia in the first quarter of 2005. As much as 99% of them fall into the category of SMEs (measured by the number of employees), employing almost 70% of private sector employees and creating 63% of the GDP. Referring to the official data of the Latvian Company Register as well as existing studies in Latvia (e.g. Dombrovsky, Chandler and Kreslins, 2005; Kuzmina, 1999) and empirical evidence from other transition countries (e.g. Smallbone and Welter, 2001, 2004; Aidis, 2006), this section of the thesis attempts to address the issue of sample representativeness. In order to do this, selected characteristics of respondents and their companies (see Appendix 3 and 4 accordingly) as found by data collected for the thesis (hereinafter referred to as the ‘SMEs2006 study’) are analysed in the context of these studies. 85

4.5.1. Individual level characteristics and sample representativeness Regarding individual-level characteristics, level of education and age are often highlighted as specific features characterising an entrepreneur in a transition context. More specifically, existing empirical studies point out that entrepreneurs in transition countries are relatively younger (e.g. Žižek and von Liechtenstein, 1994; Aidis, 2006; Kuzmina, 1999) and have a higher level of education compared to advanced western entrepreneurs (e.g. Smallbone and Welter, 2001; Blawat and Dominiak, 1994; Glas and Petrin, 1998). In this light, a high level of education is also reported by the SMEs2006 survey: almost 50% of interviewed owners and owner- managers in Latvia have a university-level education, and 10% hold graduate-level degree. The SMEs2006 survey reports that most interviewed entrepreneurs are in the age groups of 35-44 years (more than 36%) and 45-60 years (33%). Only a bit more than 2% are under 25 years old, however almost 17% are comprised of 25-34-year-old entrepreneurs. The average age of entrepreneurs is just above 44. Very similar findings regarding age distribution are also reported by recent studies in Latvia (e.g. Dombrovsky and Ubele, 2005; Kuzmina, 1999). Level of education by age groups as found by the SMEs2006 survey is presented in Table 21. Table 21. Age profiles of entrepreneurs according to level of education (column %) Under 25 Primary school education Professional (nonsecondary) education Secondary school professional education University-level education (undergraduate) University-level education (graduateengineer) University-level education (graduatemaster’s degree) PhD degree

25-34

35-44

45-60

2.1

66.7

6,.7

22.7

31.9

14.0

36.4

38.3

60.5

8.5

4.7

4.3

2.3

13.6

60-75 6.7

53.3

6.7

2.3

Source: Author’s research

86

Regarding gender distribution, just under 65 percent of respondents interviewed during the SMEs 2006 survey are male, thus female entrepreneurs make up about 35 percent. A very similar gender distribution for Latvian entrepreneurs is given by the Latvian Enterprise Register as well as by Welter and Kolb (2006). Further, the SMEs2006 survey data shows that female entrepreneurs are more represented in the age group of 35-44 (47% vs. 29% male entrepreneurs), whereas men dominate the group of over 60. No significant differences, however, exist in other age groups. Further, there are no significant differences in the education levels of male and female entrepreneurs in Latvia. Similarly, according to the SMEs2006 survey, no significant differences can be observed in the education levels of Latvian and Russian entrepreneurs, except that Latvians are a little more likely to hold graduate-level degree. The findings of the SME survey also reveal the general situation regarding ethnicity in Latvia: 57.7% of interviewed respondents are Latvians, whereas 31.5% are Russians. In general, the ethnic mix of the population of Latvia is believed to be a result of post-war immigration, resulting in a decline of ethnic Latvians from 77% in 1935 to 58% in 2005. According to data provided by the Latvian Institute (www.li.lv), 30% of the population of Latvia consists of Russians and about 10% of other nationalities (the largest of them are Belorussians, Ukrainians, Poles and Lithuanians). Interestingly, as much as 18.5% of interviewed owners and owner- managers in Latvia are classified as non- citizens (e.g. aliens of the Republic of Latvia). There are no major differences, however, in the education levels of citizens and non- citizens, apart from the fact that more citizens seem to hold graduate-level degree. The findings of the SMEs2006 survey also show that more than 70 percent of respondents had business management experience before starting their business. Of these, as much as 16 percent reported previous business experience of more than 20 years. This experience not only refers to private business ownership, but can also include holding a business management position in a state company during Soviet times (more than 30 percent) as well as working for a state or private company in the same sector as their existing business or another sector (30 percent). In this context, it is not surprising that about 20% of respondents have more than one business, of which almost ¼ have at least three businesses. 17% of interviewed business owners or owner-managers also have additional employment in another company not owned by them. 87

In general the above-mentioned characteristics of the SMEs2006 survey sample reflect the general specific characteristics of a transition entrepreneur as found by previous studies. Multiple ownership, for example, is also found to be a common characteristic of entrepreneurship in a transition context (Roberts and Tholen, 1998; Smallbone and Welter, 2001). Further, according to various surveys, most transition entrepreneurs were employed in related sectors before transition (e.g., Lane, 1995). Eurostat (1996), however, indicates that this finding cannot be generalised- a significant portion of entrepreneurs indicated that they benefited from specific technical know- how acquired through university, specialised education and/or onthe-job experience in their previous state-sector job. Lack of knowledge and previous entrepreneurship or management experience (Kusnezova, 1999), including in state-owned enterprises (Žižek and von Liechtenstein, 1994), during and after transition, are furthermore often pointed out as specific features of transition entrepreneurs. Realisation of potentially profitable opportunities in new markets (Grancelli, 1992; Webster, 1993), lack of managerial talent (Brenner, 1992), lack of know- how and experience in the areas of realising production and profit opportunities in a market-economy environment (Webster, 1993) are only some examples in this regard, indicating that finding and realising new profit opportunities is a major challenge in a transition setting. According to recent studies on SME development in Latvia (e.g. Dombrovsky and Ubele, 2005; Dombrovsky, Chandler and Kreslins, 2005), as well as findings from the SMEs2006 survey as reported above, suggest, however, that these can be attributed to early, rather than advanced-transition stages. Moreover, as also found by the SMEs2006 survey, as much as 10% of respondents admitted that they had another business before starting the current one which had gone bankrupt, but had learned a lesson from this experience and were using this knowledge for their current activities. Approximately the same percentage had their own business, which still exists, before starting their current company. Further, apart from education acquired in the school, we also asked whether entrepreneurs had attended any business-management-related courses or seminars before starting their business. As much as 38.5% answered affirmatively. 4.5.2. Venture-level characteristics and sample representativeness

88

Referring to venture-level characteristics, the average age (years in business) of the sample addressed by SMEs2006 survey is about 8 years. Further, both the number of employees and annual turnover in Latvian Lats (1 LVL equals approximately 0.75 EUR) have been used as measurements of the size of firms. The findings of the SMEs2006 survey also suggest that most enterprises fall into the category of micro-companies (1-9 employees), reflecting the general situation of SMEs in Latvia. As furthermore emphasized by most studies on entrepreneurship, during different transition stages in Latvia (e.g., Kuzmina, 1999; Dombrovsky, Chandler and Kreslins, 2005), with regard to legal status, the majority of SMEs in Latvia are limited liability companies (Ltds). Further, most Latvian SMEs are involved in the service sector as their main activity, followed by retail trade, wholesale trade and manufacturing. SMEs in Latvia also seem to be oriented towards serving local private firms and individuals (non- juridical persons). The results from our SMEs2006 survey confirm the findings of these studies. The distribution of surveyed SMEs according to main type of activity by type of customers is presented in Table 22. Table 22. Surveyed SMEs according to main type of activity by type of customers (row %)

Local private firms Foreign private firms State enterprises Municipal enterprises Individuals (nonjuridical persons) from Latvia Individuals (nonjuridical persons) from Latvia

Manufacturing

Wholesale trade

Retail trade

Service

Construction

19.4

23.9

22.4

32.8

1.5

40.0

60.0 14.3

14.3

71.4 100.0

6.7 20.0

4.4

28.9

60.0

20.0

40.0

20.0

Source: Author’s research Further, previous findings from transition countries also indicate that especially in the early stages of transition, start- ups are mostly financed either by personal savings or loans from informal sources, e.g. friends and family (e.g.,

89

Webster, 1993; Žižek and von Liechtenstein, 1994). Some argue stating that although this situation can be seen as a constraint, it also reveals a transition entrepreneur’s willingness to take the great risk of starting up a business (e.g. Aidis and Sauka, 2005). In this light, a similar patter can also be observed in the SMEs2006 survey sample. That is, most business owners and owner- managers have started their companies ‘from scratch’ and 75% of entrepreneurs used 100% of their own savings to start their business. Additionally, borrowing from informal sources (e.g. family, friends) can be observed as a major source of acquiring start- up capital for respondents addressed by the SMEs2006 survey. Empirical evidence also shows that private business owners in transition countries tend to be focused on short-term get- rich- quick business strategies instead of long- term strategies (Roberts and Tholen, 1998) and many private businesses in transition economies function in both the official and informal economy (Roberts and Tholen, 1998; Kontorovich, 1999; Smallbone and Welter, 2001, 2004). In this light, as in the case of previous business management, a short- term focus can be attributed to entrepreneurs in early stages of transition. High growth intentions, as presented in the previous sections of the thesis, found by the SMEs2006 survey as well as other recent surveys in Latvia (e.g. Dombrovsky, Chandler and Kreslins, 2005; Dombrovsky and Ubele, 2006) point to a shift towards long-term strategies instead. SME involvement in the ‘official’ and ‘unofficial’ economy is addressed in detail in Chapter 5. Previous findings on entrepreneurship in transition countries further highlights the importance of various types of informal networks, which are often used to mobilise resources, obtain orders and cope with bureaucracy. In this light Smallbone and Welter (2001) argue that previous experience during the socialist period may often be seen as a factor contributing to private-sector development in a transition period. The authors further highlight that entrepreneurs may use their old networks and connections, often referred to as a blat, in order to develop their businesses. Further, the dominance of old Russian-language networks may also continue to link newly independent FSU countries to Russia (Roberts and Tholen, 1998).

90

In this light, although ‘blat’ and ‘informal networking’ in general are perceived differently in different stages of transition23, the findings of the SMEs2006 survey partly confirm the above-mentioned arguments as exemplified by previous empirical studies in transition countries. More specifically, 41 percent of respondents confirmed that they are involved in different types of networking. Only 7 percent, however, are involved in networking activities with SMEs abroad. In this context, as highlighted by Kuzmina (1998), one of the main problems for the development of Latvian entrepreneurs is the small local market; thus, it can be argued that networking with companies from abroad can provide knowledge and financial resources for a more successful performance. As argued by Burns (2001), however, lack of resources, both financial and human, to establish contacts and enter new markets abroad is often seen as one of the important features characterising SMEs in general. In this light, at least for those interviewed, the situation with exporting also seems to be rather negative, since as much as 80% of SMEs reported that they are not involved in any kind of exporting., reflecting general SME trends in Latvia (e.g. Dombrovsky, Chandler and Kreslins, 2005). To summarise, it can be argued that the SMEs2006 survey reflects major characteristics of SMEs in Latvia on both individual and venture levels. Most sample characteristics were also found to be similar to the findings of previous empirical studies focusing on entrepreneurship in other transition countries. Thus, with some precaution, it is possible to regard the data as representative of the general target population (e.g. SMEs in Latvia), which is important for the generability of results. Before presenting the main results of the survey, however, the data analysis method should be briefly explained. This is what the final section of Chapter 4 deals with. 4.6. Approach for the analysis of results Analysis of the results to address determinants of and influences on productive, unproductive and destructive entrepreneurship can be broadly divided into three main parts. In the first part, SME value creation, e.g. output on venture and societal levels, short and long, term is addressed. This is important in light of the conceptual 23

See Ledeneva, 1998 for an overview of what “blat” meant in Soviet times and what it means in

transition, and Sauka and Welter (2007) for a definition of informal networking.

91

framework developed previously in the thesis, highlighting the importance of shifting the focus from activities to SME output when productive, unproductive and destructive entrepreneurship is assessed on both conceptual and empirical levels. Exploratory analysis is used in this step and the results are reported in Section 5.1. of Chapter 5. In the second part, the author proceeds to explore which activities EB and EO entrepreneurs pursue in order to create this output, resulting in productive, unproductive or destructive entrepreneurship. This again involves several steps: Section 5.2. Determinants and influences of productive, unproductive and destructive entrepreneurship’ aim to address Hypotheses 1, 2 and 4 to 9, that is, to explore the relationship between activities (e.g. EB and EO) and SME value creation as well as the moderating effects of access to capital and environmental influences in this relationship. Drawing on the research design as suggested by Wiklund and Shepherd (2005), the complexity of this relationship can be assessed by exploiting main-effects-only, contingency and configuration models. This design is used in the analysis. More specifically, as suggested by Wiklund and Shepherd (2005), hierarchical linear regression analysis is used to assess the involvement in directly and indirectly productive, unproductive and destructive entrepreneurship and test whether a universal, contingency, or configuration model best suits the data and whether variables have a significant relationship with SME output on various levels and time frame. Hierarchical linear regression analysis has been recognized as a valid procedure both mathematically and in computer simulations and is applied by numerous authors when assessing issues structurally similar to the ones addressed in this thesis24. The general approach of hierarchical linear regression analysis used to test the hypotheses is to add the next higher order of interaction (e.g. two-way and three-way interactions respectively) in each of the steps of the analysis. In other words, first control variables are added, then the independent variables (main-effectsonly model), the two- way interaction terms (contingency model), and finally the three way- interaction term (configuration model). Incremental R² is evaluated in order to determine whether the main-effect, contingency or configuration model best suits the data. According to Cohen and Cohen (1983), “An interaction effect exists if, 24

See Wiklund and Shepherd, 2005: 82 for a more detailed description of this procedure, including validity issues.

92

and only if, the interaction term gives a significant contribution over and above the direct effects of the independent variables” (in Wiklund and Shepherd, 2005:82). F tests of statistical significance are also evaluated from lower- order terms to determine the statistical significance of the dependent variables, e.g. EO, EB, access to capital, environmental influences and interaction terms, in relation to SME value creation. Summated indexes of EO and EB are used throughout the analysis in Section 5.2. This choice was made due to the fact that summated EO is also used by Wiklund and Shepherd (2005), e.g. the design this thesis is based on. Cronbach’s alpha values for these indexes are .787 and .530 respectively. However, in Section 5.3., ‘Determinants of productive, unproductive and destructive entrepreneurship: individual indexes of EO and EB’, analysis continues by including separate dimensions of EB and EO, as distinguished in Sections 4.2.1 and 4.2.2. This is important in light of the discussion of exploring the role of EO on the performance of firms. In this context, although one should be aware that various dimensions of EO work together to enhance a firm’s entrepreneurial performance” (Dess and Lumpkin, 2005), summarising the findings from various empirical studies addressing the relationship between EO and performance. However, Dess and Lumpkin (2005: 148) conclude that in general, “… exploring the relationship among individual dimensions of EO and performance is superior to considering EO as unidimentional construct”. Namely, EO dimensions can vary independently in their relationships with SME output, thus apart from using the summated index of EO, it is more insightful to use separate indices representing each of the five dimensions of EO (e.g. innovativeness, autonomy, proactiveness, competitive aggressiveness and risk- taking). In the context of the empirical model, the author argues that the same notion can also be applied to various dimensions of EB (e.g. rent seeking, unofficial behaviour and unethical behaviour). This is of even greater importance since Cronbach’s alpha levels for EB dimensions are relatively low and in case of ‘unethical behaviour’ do not even reach the level of 0.5. Section 5.4. addresses the question as to which activities SMEs pursue in order to create value on various levels and time perspectives. This time, however, an exploratory approach is used in order to get a more complete picture. In all sections, e.g. 5.2., 5.3. and 5.4., separate analysis is carried out with dependent variables (a) venture-level output in the short term, (b) venture-level output in the long term, (c) societal-level output in the short term (d) societal-level output in the long term. This is 93

in turn necessary since SME output is observed on a venture and societal level, short and long term, in both of the above-described cases. Finally, in the third part of the analysis additional influences on SME value creation, such as the mediating effect of EB in the EO and SME value creation relationship and (Hypothesis 10) the influence of goals and expectations (Hypothesis 3) are addressed. Altogether, the results of the analysis covered by the next chapter of the thesis provide insight into the directly and indirectly productive, unproductive and destructive entrepreneurship.

94

5. Empirical exploration of productive, unproductive and destructive entrepreneurship In this chapter of the thesis the results of the SME survey are summarised, representing one of the first attempts to address productive, unproductive and destructive entrepreneurship empirically. Since the conceptual framework for the assessment of these concepts, as developed in Chapter two, highlight the importance of shifting the focus from activities to output in order to assess SME value creation, the first subsection briefly deals with this issue. In the following sections the author then proceeds to explore the relationships between activities and output as well as the factors influencing these relationships. The analysis draws on the conceptual framework as well as the working definition of entrepreneurship as introduced in the previous chapters of the thesis and aims to address the hypotheses developed in Chapter 2. The answers to the hypotheses provide a scope for some conclusions, implications and policy suggestions regarding SME involvement in productive, unproductive or destructive entrepreneurship, which is one of the tasks of Chapter 6. 5.1. SME output: venture and societal level From both short and long-term perspectives, the surveyed SMEs show a positive tendency in terms of expansion and company growth as well as nonmonetary dimensions of venture-level output (see Table 23). From a short-term perspective, a large share of SMEs (approximately 60%) reported an increase in profits and turnover, whereas only a few were able to increase profits and turnover to a considerable extent (4.6%). Similar results were also reported for the same indicators in longer term, with the difference that considerable increases were more pronounced in the long-term perspective both for profits and turnover. An increase in investments in the short and long term shows a similar pattern. Interestingly, that respondents scored highly on personal fulfilment, with nearly one quarter being very satisfied and 50 percent being satisfied to a great extent, both in the short and long term. Quite a different picture from other venture-level indicators, however, was reported for the generation of employment. In the short term, almost one half of surveyed SMEs reported that the number of employees has remained the same, whereas in the long term the picture is lightly more positive.

95

Table 23. Venture level output of surveyed SMEs Venture level output (%) Short-

Short-

Short-term

Short-term

Long-

Long-

Long-

Long-

term

term

employment

investment

term

term

term

term

profits

turnover

profits

turnover

employment

investment

4.6

4.6

0.8

1.5

20.0

21.5

12.3

15.4

Increased

58.5

60.0

37.7

40.8

52.3

50.8

40.0

31.5

Stable

26.2

23.8

46.9

49.2

12.3

13.1

31.5

37.7

Decreased

10.8

11.5

13.8

3.8

10.8

10.0

11.5

3.8

Decreased

0

0

0

0

1.5

1.5

1.5

0

Increased considerably

considerably Personal satisfaction short term (%)

Personal satisfaction long term (%)

41.5

36.2

46.9

48.5

10.8

13.1

0.8

0.8

0

1.5

Yes, to a very great extent Yes, to a great extent Yes, to some extent Yes, to a small extent No

Source: Author’s research

Although overall SMEs show a rather positive attitude regarding their perception of their contribution to society, different results are reported with regard to individual indicators (see Table 24). According to the survey, the smallest contribution of firms on a societal level, as perceived by the surveyed owners and owner managers, are in the fields of attracting investments from abroad and bringing new technologies to the country. This finding however is not surprising, since SMEs are usually too small to have a significant impact in these fields (e.g. Burns, 2001). On the other hand, SMEs seem to perceive their greatest contribution to society as tax-payers, e.g. 52.3 percent ‘to a great extent’ vs. none answering ‘no’ in the short term. Similar findings in this regard are reported in the long term as well. Comparably high results, however, are reported with regard to SMEs’ contribution to the education of employees and generation of indirect employment. At least to some extent, both in the short and long term, SMEs are also involved in the generation of innovations, investing into infrastructure, generation of knowledge spillovers as well as sponsorship of society’s needs (see Table 24).

96

Table 24. Perceived societal-level output of surveyed SMEs

Yes, to a very great extent Yes, to a great extent Yes, to some extent Yes, to a small extent No

Yes, to a very great extent Yes, to a great extent Yes, to some extent Yes, to a small extent No

Societal level output short term (%) Bringing Attracting Investing in new technologies investments from infrastructure to the country abroad

Sponsorship for social needs

Paying taxes

Generating knowledge spillovers

0

1.5

7.7

0.8

1.5

1.5

5.4

52.3

9.2

9.2

5.4

36.2

50.0

36.2

32.3

20.8

24.6

40.0

32.3

1.5

43.1

8.5

0

12.3

Sponsorship for social needs

Paying taxes

Generating knowledge spillovers

0

1.5

12.3

0.8

1.5

3.1

8.5

50.8

12.3

12.3

10.0

41.5

57.7

30.8

34.6

26.9

23.8

25.4

37.7

23.1

3.8

39.2

8.5

56.9

60.8

15.4

6.9

0

10.8

Indirect employment

Educating employees

Innovations products/ services

0

1.5

2.3

0.8

0

12.3

20.0

9.2

1.5

29.2

51.5

41.5

36.2

23.8

34.6

20.0

0.8

10.0

Indirect employment

Educating employees

Innovations products/ services

0.8

3.1

2.3

0.8

0

12.3

25.4

13.1

3.8

36.9

50.0

46.9

33.1

19.2

14.6

0

Source: Author’s research

65.4 66.2 20.0 Societal-level output long term (%) Bringing Attracting Investing into new technologies investments from infrastructure to the country abroad

5.2. Determinants and influences of productive, unproductive and destructive entrepreneurship This section of the thesis continues to analyse the activities SMEs pursue in order to achieve output as reported in Section 5.1. More specifically, drawing on the conceptual framework introduced in Chapter 2, the author first aims to explore the influence of control variables (e.g. age and size of the firm, sector, education level and gender of the owner as well as the direct relationship between EO and SME value creation (Hypothesis 1), and EB and SME value creation (Hypothesis 2). Further, influences of the environment and access to capital are addressed by using both a contingency model and a configuration model25. It should be noted, however, that the tendencies towards positive output on both venture and society levels in different time frames, as reported in the previous section of the thesis, influence the results and should therefore be considered throughout the subsequent analysis. The relationship between determinants of and influences on (directly and indirectly) productive, unproductive and destructive entrepreneurship as addressed by this section are illustrated in Figure 15. The results of hierarchical regression analysis using summated EO are displayed in Tables 25 and 26 (venture level short and long term) and Tables 27 and 28 (societal level short and longer term). Results representing EB on SME value creation relationship are displayed in Tables 29 and 30 and Tables 31 and 32 accordingly26 Here, in order to test the hypotheses, control variables were first added (results are reported in Column 2), the direct effects (e.g. universal model – results reported in Column 3), the two- way interaction terms (e.g. contingency model – 25

As described in Chapter 2, Subsection 2.6.3., in the scope of the conceptual framework, contingency models address (a) the two-way relationship between EO, SMEs value creation and environmental influences (Hypothesis 4) as well as EO, SME value creation and access to capital (Hypothesis 5); (b) the two-way relationship between EB, SME value creation and environmental influences (Hypothesis 7) as well as EB, SME value creation and access to capital (Hypothesis 8). The configuration model, however, addresses the joint relationship between EO, SME value, access to capital and environmental influences (Hypothesis 6) as well as EO, SME value, access to capital and environmental influences (Hypothesis 9). 26

The means, S.D’s and correlations of the variables determining and influencing SME value creation are displayed in Appendix 8 and Appendix 9 for SME value creation in the short and long term, for the EO and SME value creation relationship and the EB and SME value creation relationship respectively. Although individual tables and figures for each case are not included in order to save space, the author checked for violation of assumptions throughout the analysis. In most of the cases neither collinearity, nor violation of the normality assumption and heteroskedasticity were an issue. Several outliers were also identified and disregarded, resulting in slightly higher R2.

Entrepreneurial Behaviour (summated index)

Directly and results reported in Column 4), followed by the three way- interaction term (e.g. indirectly

Goals & Expectations

configuration model – results reported in Columnproductive, 5). SME value

unproductive or destructive entrepreneurship Figure 15. Assessment of (directly and indirectly) productive, unproductive and creation

destructive entrepreneurship Entrepreneurial Orientation (summated index)

Environmental influences & Access to capital

Source: Author The control variables of business age, size, sector and owners’ or ownermanagers’ education and gender explain 7.9% (short term: Table 25 and 29) and 14.3% (long term: Tables 26 and 30) of the variation in venture-level output (P