BAUMOL-LIKE ENTREPRENEURSHIP: AN EMPIRICAL ASSESSMENT AND AN EXTENSION

(1)

BAUMOL-LIKE ENTREPRENEURSHIP: AN EMPIRICAL

ASSESSMENT AND AN EXTENSION

Research Master Thesis

Seçil Hülya Danakol

Research Master Student of International Economics and Business Profile

Supervisor: Arjen van Witteloostuijn

SOM Graduate School University of Groningen Groningen, the Netherlands

(2)

Abstract

In his leading paper, Entrepreneurship: Productive, Unproductive and Destructive Baumol has made a huge contribution to the entrepreneurship literature. He questions taken for granted assumption underlying the theory of entrepreneurship: entrepreneurship is good for society. Contrary to the conventional wisdom, he claims that it is not the level of entrepreneurship but types of it that should be of interest if we want to understand its potential contribution or detriment to the wealth of the countries. The more the amount of productive activities are, the greater the welfare-increasing effects of entrepreneurship for the society as a whole. On the other hand, it is noted that the prevalence of unproductive and destructive entrepreneurship inhibit economic growth. It is this unconventional interpretation of the entrepreneurship articulated in Entrepreneurship: Productive, Unproductive and Destructive that inspired much of this work.

Baumol and his followers explicitly assumed that entrepreneurial talents are not concerned at all about the potential increase or decrease in overall welfare stemming from their activities. In other words, positive or negative externalities are out of interest for entrepreneurial talents, since the focus is only on personal gains, it is assumed. Therefore, the first task of this paper is to investigate whether this assumption holds empirically in reality. In fact, our results provide strong evidences in favor of Baumol’s claim. The findings suggest that entrepreneurially talented people make investment decisions without taking into account the potential productive or destructive outcomes stemming from their activities.

Secondly, and perhaps more importantly, I have pointed out some shortcomings inherently present in his 1990-paper and suggested some viewpoints to remedy them. More clearly, I have proposed an extension to Baumol-like entrepreneurship thinking where entrepreneurs are labeled as talented individuals possessing either altruistic or exploitative entrepreneurial attitudes. Based on this Baumol-extension, an additional hypothesis is constructed and empirically tested. Results suggest that once entrepreneurial talent is classified as altruistic vs.

exploitative in varying degrees, the former type has a tendency to make more investments

(3)

Acknowledgements

First of all, I would like to thank my supervisor Prof. Dr. Arjen van Witteloostuijn for tolerating a lazy student such as me, without his bright ideas and ever-lasting patience I could not be writing these lines. I would also like to thank Asst. Prof. Dr. Utz Weitzel, for his great support and guidance during this thesis.

I would like to send my heartfelt thanks to Assoc. Prof. Dr. Bart Los and Prof. Dr. Jakob De Haan for granting me tuition waiver awards throughout my master studies. Without this scholarship I would have been unable to take part in this master programme.

I would like to express the deepest gratitude to Mr. Osman Özer Özkan who provided me with a two-year scholarship for my living expenses in the Netherlands. Without his help, I could not have started my academic life and could not be as successful as I am now. Besides his financial assistance, he was extremely generous and passionate to share his 60-year life experience with me. Among other valuables, he just told me not to fear of learning new things; and his advices will be with me forever.

Wonderful Selvi, my very best friend in Groningen, deserves a special recognition. Like me, she is also a (alleged) victim of the research master programme but in a different field: Molecular Biology and Genetics. We have shared the very hard times of being a research master student as well as good ones. While I was writing these lines, we have learned about her job offer as a PhD researcher at Netherlands Cancer Institute in Amsterdam. This basically means our wonderful times together in Groningen will mirror in Randstad during the coming four years. Selvi, thank you for your great support for my thesis and your sincere friendship

And the Oscar goes… ups… sorry and my biggest thanks go to my family, especially to my dearest father, Sefer; then to my mother Sebahat and my brother Sedat. I owe everything that I have ever succeeded to them. You are my only reason being who I am now. I will feel forever blessed to have such a great family.

(4)

List of Figures and Tables

In the Main Text:

Figure 1: Baumol-extension model: Altruistic vs. exploitative attitudes………20

Figure 2: The entrepreneur can not be isolated from the context………24

Figure 3: Overview of 36 investment decisions in experiment………30

Figure 4: Marginal effect of ESE on investment, conditional on destructive dummy……….38

Figure 5: Predicted investment with ESE………39

Figure 6: Predicted investments with GRP………..42

Figure 7: Marginal effect of GRP on investment, conditional on destructive dummy……....43

Figure 8: Destructive activities as the causes of both productive and destructive outcomes..51

Table 1: Estimation results with regard to ESE………...36

Table 2: Estimation results with regard to LOC………..40

Table 3: Estimation results with regard to GRP………...41

Table 4: Estimation results with regard to EMAD………...44

Table 5: Estimation results with regard to EI………...46

Table 6: Estimation results with regard to AE……….47

In Appendix: A1: Correlation matrix of the variables………53

A2: Descriptive statistics of the variables……….57

A3a: Marginal effect of LOC on investment, conditional on destructive dummy………...…60

A3b: Predicted investment with LOC………..60

A4a: Marginal effect of EMAD on investment, conditional on destructive dummy………...61

A4b: Predicted investment with EMAD………...61

A5a: Marginal effect of EI on investment, conditional on destructive dummy………...62

A5b: Predicted investment with EI………...62

A6a: Marginal effect of AE on investment, conditional on destructive dummy………..63

(7)

List of Abbreviations

ESE: Entrepreneurial self-efficacy

GRP: General risk propensity LOC: Locus of control

EMAD: Exploitative manipulative amoral dishonesty EI: Entrepreneurial intent

AE: Attitude towards entrepreneurship AA: Altruistic attitudes

(8)

1. Introduction

As developing and underdeveloped parts of our world have been engaging in a constant battle to achieve economic growth thus alleviate poverty, it is of great concern for both economists and policy makers to uncover the root causes of economic growth. The literature to date points out the presence of both economic and non-economic factors which may potentially, though in varying extents, induce economic growth (see Sala-i Martin, 1997; and Bleaney and Nishiyama, 2002). Among these factors, entrepreneurship is also viewed as part of this big picture of development where it can take on various forms to achieve its mission. In this regard, since Cantillon, perhaps even earlier, the potential contributions of the entrepreneurial activities to economic development have frequently been emphasized by many prominent economists including Schumpeter and Kirzner. Although, the key roles that have been assigned to the entrepreneur and entrepreneurship vary according to each, they share a common ground: that is entrepreneurship is good for society.

On the other hand, over the past decades, there appeared a set of economics people1_who challenge this assertion. They refuse the unreasoning obedience to the claim stating that entrepreneurial activities are usually, if not always, beneficial for economic growth. These scholars argue that the influences of entrepreneurship on economies are not de facto, but the extent of impacts changes based on how the entrepreneurially-talented people make use of their skills by allocating them into productive, unproductive, or destructive opportunities (Baumol, 1990). On the condition that talent is used up in productive activities (e.g., job creation), it is claimed, the realization of the social benefits will necessarily follow which is the utmost desired situation. On the other hand, the prevalence of the unproductive and destructive entrepreneurship (e.g., rent-seeking) inhibits economic growth not only because they are themselves dangerous but also they absorb talent which would have been employed in productive activities and contributes the overall well-being otherwise, they argue.

(9)

conventional wisdom, Baumol describes entrepreneurship broadly and includes rent-seeking and illegal activities, organized crime etc. into his definition of entrepreneurship. According to him, an entrepreneur, regardless of the type of activity he engages in, is:

“…any member of the economy whose activities are in some manner novel, and entail the use of imagination, boldness, ingenuity, leadership, persistence, and determination in the pursuit of wealth, power, and

position, though not necessarily in that order of priority. In other words,

the term is meant to encompass all non-routine activities by those who direct the economic activities of larger or smaller groups or organizations” (Baumol 1993, p: 7-8, first emphasis added, second is in original)

Therefore, he views the general entrepreneurial talent as a combination of exceptional abilities. In his 1990-leading paper and in others’ such as Murphy et al. (1991, p: 506), Acemoglu (1995), explicit the assumption that “talent allocates into activities with the highest

private returns, which need not the highest social return”. Despite stimulating many research

activities and the centrality of this assumption to this body of research, surprisingly no studies till now have empirically tested the validity of Baumol’s original assumption. In this regard, the contribution of this paper to the entrepreneurship literature is threefold. To begin with, the first mission of this study is to assess to what extent Baumol’s assertion holds empirically in reality. Secondly, and perhaps more importantly, I have pointed out some shortcomings inherently present in his 1990-paper and suggested some viewpoints to remedy them. More clearly, I have proposed an extension to Baumol-like entrepreneurship thinking that expands our understanding, I argue. Based on this Baumol-extension, an additional hypothesis is constructed and empirically tested.

In the extension, contrary to Baumol (1990), entrepreneurs are no longer classified as productive vs. unproductive, instead, I have taken a step forward so as to humanize2_the entrepreneurs. The model locates entrepreneurs on a continuum where the poles corresponds either purely altruistic or purely exploitative attitudes. As socially-constructed identities, it is up to entrepreneurs to place themselves on a certain place on this continuum, I claim. This 2_{The word ‘humanize’ means “a behavioral dimension is added to categorize entrepreneurs, of which by}

(10)

indicates a categorization, of which entrepreneurs are labeled as talented individuals possessing either altruistic or exploitative entrepreneurial attitudes.

In short, I find empirical support for the assumption proposed by Baumol. In other words, the results suggest that general entrepreneurial talent allocates into the activities providing the highest private return no matter what kind of externality (productive or destructive) the society has to bear due to the conduct of these activities. Furthermore, results also suggest that once entrepreneurial talent is classified as altruistic vs. exploitative in varying degrees, the former type has a tendency to make more investments when their activities generate positive externality on society. On the other hand, exploitative entrepreneurs act in the opposite direction, as expected, and make higher levels of investments when they know the fact that their operations harm innocent others.

The extension and results based on this view do not contradict Baumol’s claim on the allocation of entrepreneurial talent, but rather completes and expands our understanding of an increasingly vital area of economics. Furthermore, it calls for more advancement in the conceptualization of the terms; productive, unproductive and destructive entrepreneurial activities as well as entrepreneurs engaging in these activities.

The data used in this study is generated by means of experiments under lab conditions with university students as subjects. Therefore, the methodology adopted in this paper is as important as the empirical and theoretical contributions. This is especially because, within entrepreneurship literature, there are just a few studies which make use of experimentally-obtained data. Some other areas of economics research extensively benefit from this methodology, but this is not the case for the entrepreneurship research. However, as suggested by Schade and Burmeister-Lamp (2009), “experiments are especially suited to empirically test hypotheses...” that are relevant to the entrepreneurial decision-making. Therefore, in the context of this study, I have acknowledged their suggestion and made use of experimentally-generated data. In this regard, I firmly believe that one of the important contributions of this study to the entrepreneurship literature resides in the employed methodological approach.

(11)

(12)

2.

Entrepreneurship Defined: A Path from History towards

New Understanding

3

It is well-acknowledged that the position of the entrepreneurship theory within economics thought remains ambiguous despite important advances in the field. There is a fair amount of literature delving into the fundamentals of the entrepreneurship4_{, but this phenomenon still} persists in being “the most emphasized yet least understood by economist” (S. M. Kanbtir; quoted in Hebert and Link, 1988). However, this vagueness does not prevent entrepreneurship from being one of the most intriguing phenomena in economics concerning its central role as an economic development tool (Baumol 1968).

The first use of the term entrepreneurship within economic theory has been attributed to

Richard Cantillon who refers to the general undertaking of a business (Murphy, 1986). In his view, entrepreneurs are economic agents whose major function is to buy goods at certain prices and sell them at unknown prices under uncertainty. Thereby, Cantillon’s entrepreneurship framework already emphasizes the presence of profit motive of individuals, uncertainty and risk bearing aspects of the market transactions (Cantillon, 1755). In the 1800s, attention toward entrepreneurship was still encouragingly alive and its fundamentals have been the focus of other pioneering economists including Jean-Baptiste Say and John Stuart Mill. The contribution of the former one lies in his description of the entrepreneur as an agent who has exceptional ability in creating new economic enterprises as well as combining factors of production (Sobel, 2008). On the other hand, Mill differentiates between entrepreneurs and other business-related people (e.g. managers) and indicates only former ones bear risks (Sobel, 2008)

Frank Knight (1921), another widely influential economist, describes entrepreneurship as dealing with uncertainty. He offers insights on entrepreneurship by making a clear distinction between uncertainty and risk. According to his argumentation, entrepreneurs clearly bear risks, of which they are aware, the magnitude and severity are known due to the computable 3_{Though entrepreneurship is a multidimensional phenomenon whose boundaries cut across many fields, in this}

section it is assessed through the lens of economic theory.

(13)

nature of risky activities (Knight 1921). Therefore, this type of risk, Knight argues, is insurable and can be easily handled through the employment of several mechanisms such as contracts, hedging, diversification etc. On the other hand, what is of more concern for entrepreneurs is Knightian uncertainty rather than calculable risk. In this regard, uncertainty involves immeasurable risk which is not possible to remove through the execution of certain actions. Thus, Knightian uncertainty is unknowable in its real substance (Holcombe, 1998). Therefore, entrepreneurs face with both calculable risk and uncertainty. In this regard, the entrepreneur’s “success or failure depends on the correctness of his anticipation of uncertain events” (Mises, 1949, p: 290).

Of all the theories of entrepreneurship that exist, Schumpeter’s proposition (1911, 1942) is probably the one that has gained most recognition. In this theory, he considers entrepreneurs as economic actors that drive innovation by means of several manifestations of novelty. More clearly, he identifies five different “new combinations” of ideas and resources that take the form of new and superior goods and services, enhanced production techniques, new and expanded markets, better organizational structures, and supply sources. Schumpeterian entrepreneur acts as a “creative destructor”, thus shifts the economy away from equilibrium. It is this creative destruction process that fuels economic growth through more efficient allocation of resources. Ultimately, entrepreneurs are able to receive economic rewards in exchange for their successful introduction of novelty to the system (Schumpeter, 1911, 1942).

(14)

equilibrium state (Holcombe, 1998). Therefore, these two views, Holcombe claims, are complementary rather than opposing perspectives adopted in entrepreneurship theory.

It should be emphasized that the contributions to the entrepreneurship theory discussed above is not a comprehensive synthesis of all literature, but includes the ones which are considered as the most valuable to mention. From the previous discussion, several key properties of an entrepreneur have been suggested by different scholars. Along the spectrum of entrepreneurship, we see individuals are considered as creative destructors and innovators, engaging in market transactions by exploiting profit opportunities in the face of uncertainty and bearing risk. Schumpeterian entrepreneur is responsible for destructing the equilibrium and canalizing the factors of production into better usages. Furthermore, while Knight’s economic agent has an exceptional competence of anticipating uncertainty, the entrepreneurs in Kirzner’s mind excel in spotting unexploited gain opportunities. In this regard, there is no doubt that the notion of entrepreneurship is multifaceted. However, the literature to date was unable to meet the task of integrating all diverse, though not mutually exclusive, features into a single over-arching theory of entrepreneurship.

On the other hand, even though all these economists assign various roles to the entrepreneur and entrepreneurship, they were in search of a unique end, that is the economic growth and development. By this I mean, these scholars have not questioned the darker side of the entrepreneurship assuming that it is generally desirable for society at large. Baumol has taken on this difficult task and reminded us to be more cautious when discussing entrepreneurial activities in relation with economic growth. Now, discussion proceeds with the analysis of Baumol’s ground-breaking work.

2.1. Baumol: Entrepreneurship as Productive, Unproductive and

Destructive

(15)

contribution or detriment to the wealth of the countries. Therefore, he calls for the need of such a categorization of entrepreneurial activities based on their economic consequences, and Baumol himself identifies three of them. Namely; productive, unproductive and destructive entrepreneurial activities

To begin with, Baumol takes the departure point for his theory from Schumpeter’s well-known ‘innovative combinations’ and regards the theory as an extension of it. His analysis is roughly based on two main presumptions. Firstly, he assumes that entrepreneurs are always with us and the supply of entrepreneurial talent is constant, more or less equally distributed across societies. Secondly, he points out that the allocation of these talents to different activities -which may be productive, unproductive or destructive- is significantly determined by the nature of the institutions prevailing in different countries. By institutions he means ‘the rules of the game’ whose role is, among others, to shape the level economic gains attributed to each type of entrepreneurial activity. In this regard, entrepreneurial talent is highly responsive to the relative economic rewards of each type of activity that in turn increase their individual well-beings but not necessarily wealth-creating at the society level. Due to the vital role, of which Baumol ascribes on institutions in the theory of entrepreneurship, I believe, a further -though not very extensive- clarification on the description of institutions is in order. Douglass C. North who is a major contributor to Institutional Economics defines institutions as:

“the rules of the game in society or, more formally... the humanly devised constraints that shape human interactions. In consequence, they structure incentives in human exchange, whether political, social, or

economic. Institutional change shapes the way societies evolve through

time and hence is the key to understanding historical change.... Institutions reduce uncertainty by providing a structure to everyday life" (1990, p: 3, emphasis added).

(16)

regards, among others, job generation, innovation, financial operations aiding goods production and service delivery or other entrepreneurial engagements facilitating production as positive and desired contributions on societal and economy levels (Baumol, 1993).

With regard to unproductive and destructive entrepreneurship, the fundamental point to underline is that these activities either generate no additional value or detrimental in nature from the viewpoint of the economy at large. Entrepreneurs tend to engage in these activities, Baumol argues, when rules of the game in the society rewards and values these types more than productive entrepreneurship. In the presence of institutions that are not functioning properly, unproductive and destructive activities attract talents due to more gain opportunities that augment personal well-being. Rent-seeking, illegal activities, takeovers, tax evasion and avoidance, corruption, organized crime and litigation etc. are all classified under the title of unproductive and destructive entrepreneurship (Baumol, 1990, 1993).

The implication of Baumol’s entrepreneurship theory on economic growth is straightforward. The level of economic development is related significantly to the way, of which entrepreneurial talents are exploited within a particular economic system. The more the amount of productive activities are, the greater the welfare-increasing effects of entrepreneurship for the society as a whole. On the other hand, the prevalence of unproductive and destructive entrepreneurship inhibit economic growth not only because they are themselves dangerous but also they absorb talent which would have been employed in productive activities and contribute overall well-being otherwise. In this regard, Murphy et al. (1993) makes a further distinction within unproductive activities in the sense that rent-seeking is realized both within private agents as well as government officials. Mostly, the latter group of plunderers whose actions, they argue, are responsible for blocking the economic development through the attacks of innovativeness.

2.2. Baumol Followers

(17)

improving the fundamentals of the theory and testing whether Baumol’s original claims hold in reality, though the latter is scant. For example, Desai and Acs (2007) point out that Baumol’s original article clearly lacks in-depth analysis of destructive entrepreneurship although, they argue, the main contribution of the article lies on this concept. Therefore, a theory of destructive entrepreneurship has been proposed to conceptualize the idea more accurately, chiefly in a developing country context. By emphasizing the assertion that destructive activities are context specific5_{, they consider slavery and improper extraction of}

the natural resources in African countries as destructive entrepreneurship.

Sauka (2008) adopts an alternative, though interesting, approach to the variants of Baumol’s entrepreneurship. In this regard, he distinguishes between the outputs and the activities, of which these outputs are generated. It is stated that whether the level of output reasonably increase as a function of any kind of entrepreneurial activity, be it legal, illegal, formal, informal, rent-seeking or other types of activities, should be of concern. This view is especially relevant within transition context where stable legislation and rules are underdeveloped, and thus rent-seeking (e.g. tax avoidance) benefit to society by enabling firm survival and growth (Sauka, 2008; Smallbone and Welter, 2006).

Douhan and Henrekson (2008a, 2008b) challenge Baumol’s argument of institutions as a one- way channeling mechanism for entrepreneurial talent. The nature of the association between entrepreneurship and prevailing institutions in a specific society, they argue, is unambiguously bilateral rather than only from institutions to entrepreneurship as assumed by Baumol (1990). In its essence, the latter study claims that non-productive6_engagements mostly emerge as a second-best substitute for suboptimal institutions. Corruption and mafia formation are given as examples of productive reactions to inefficient institutions. Furthermore, Douhan and Henrekson (2008a) propose a political economy framework where entrepreneurs are of the main character whose various operations shape the nature of the institutions.

Sobel (2008) provides a direct test of Baumol’s basic hypothesis that sound institutions stimulate productive entrepreneurship. He finds that in the U.S states where higher quality of

5_{This view indicates that slavery and natural resource extraction are bad for the country, of which these activities}

take place but good for the final destination.

(18)

institutions are in place such as secure property rights, efficient enforcement of contracts, effective judiciary etc. enjoy higher levels of productive entrepreneurship, that in turn induces economic growth. With a similar approach, Bowen and Clercq (2008) analyze to what extent various forms of institutions influence the way entrepreneurial efforts directed towards high-growth activities, of which they simply refer Baumol’s ‘productive entrepreneurship’. They distinguish between four forms of institutions suggested by Whitley (1999). Namely, these are the availability of financial capital (sound financial system), of educational capital targeted to entrepreneurship, regulatory protection (e.g. secure property rights) and the level of corruption). The results are in line with the anticipations of the scholars which suggest that the presence of improved institutions channel entrepreneurial efforts towards high-growth activities.

Acemoglu (1995), no doubt, has provided a seminal contribution to the research that concentrate on entrepreneurial talent acting either productive or unproductive. He proposes a theoretical model of talent allocation where productive and rent-seeking7_{activities coexist but} the latter reduces the relative returns of productive agents. This results in a shift in the reward structure in favor of predatory actions so the economy will be dominated by the rent-seeking population. Consequently, the model suggests, since the relative rewards endogenously determined, depending on the initial allocation of entrepreneurial talent the economy may be locked in underdevelopment trap. Furthermore, Murphy et al. (1991) developed a two-sector model drawing on similar argumentation as Acemoglu (1995). In this theoretical design, talent finds its way into either productive or rent-seeking sector where the attractiveness of each depends on three factors. Namely, the size of the market, returns to scale and the compensation contract. Finally, they find empirical evidence using cross-section data of over 90 countries that rent-seeking as an unproductive entrepreneurship reduces growth.

(19)

behaviours of a group of talented people standing in the forefront of the society. In this regard, Murphy (1991, 1) properly defines what is meant by entrepreneurial talented people. These people, according to him, “…possess great intelligence, energy, or other generally valuable traits. Such people can become one of the best in many occupations…” And, the way they behave, as discussed theoretically, depends only on the rewards of a particular activity relative to others. Therefore, it is explicitly assumed in these papers that entrepreneurial talents are not concerned at all about the potential increase or decrease in overall welfare stemming from their activities. In other words, positive or negative externalities are out of interest for entrepreneurial talents, since the focus is only on personal gains, it is assumed.

The abovementioned assumption was explicitly stated by Baumol (1990, p: 898-899) which is the main departure point of this study.

“…entrepreneurs are defined, simply, to be persons who are ingenious and creative in finding ways that add to their own wealth, power, and prestige, then it is to be expected that not all of them will be overly concerned with whether an activity that achieves these goals adds much or little to the social product…”

Also Murphy (1991, p: 506) nicely adds:

“…talent goes into activities with the highest private returns, which need not have the highest social returns.”

(20)

action. Furthermore, economic system with its fundamental component, that is entrepreneurs would be just a paradox if the economic gain motive is abandoned (Schumpeter 1949, 1954). Finally, Mises (1949, p: 254, 290-91) complements and concludes the discussion by nicely stating “the entrepreneur is the man who dedicates them to special purposes. In doing so he is driven solely by the selfish interest in making profits and in acquiring wealth”.

Based on the discussion above, the following hypothesis is formulated.

Hypothesis 1: Entrepreneurs will allocate their talents into the activities providing the highest

(21)

3. Entrepreneurship as a Social Phenomenon

As economics theory claims to be a social science, there are harsh critics raised against, for it places the rational individual, allegedly ruled by self-interest, right at the centre of the theory. Such a view, from the very beginning, abstracts the agents from their social context in the sense that their behaviours are assumed to be minimally, if not at all, influenced by social relations or the social context generally. However, in this part of the study, the merits of the atomized view of economic behavior with a particular focus on entrepreneurship and entrepreneurs will be challenged. In doing so, it will be argued that entrepreneurship is inherently a social phenomenon that is rooted in a broad set of social and cultural relations. The perspective I employ regarding the social embeddedness of entrepreneurship does not aim at over-socializing the concept, but rather emphasizing the importance of the social context in its realization.

The use of ‘embeddedness’ as referring social complexity is initially suggested by Polanyi (1944, Polanyi et al. 1957) who states “man’s economy, as a rule, is submerged in his social relations. …” (1957, p: 46). Granovetter (1985), in his seminal article, criticizes both extremist standpoints that consider economic behaviors either under-socialized with rational arguments or over-socialized under the influence of social context. Rather, he adopts a stance of embeddedness where economic actions take place in an ongoing system of social relations, but not under their pure control. As a result, I draw on this way of thinking in conceptualizing the entrepreneurship as a social phenomenon where a constant interplay occurs between sides: namely, the self and the social.

In order to make my claims clear, I will begin the discussion by briefly examining the position of the entrepreneurship in economics, particularly in mainstream economics theory9_{. Although} this point has been concisely touched upon in the previous section, I deem it necessary to add some further aspects for a better understanding of the analysis that is to follow. Namely, the analysis of the socially-constructed dimension of entrepreneurship and entrepreneur, which in turn leads to the creation of the second hypothesis of this study

(22)

In essence, mainstream economics can be described as a mixture of the notions optimization, methodological individualism, rationality, the distribution of scare resources and general equilibrium (Colander, 2000). Although it has been to a large extent failed to include the entrepreneur and relevant phenomena into its formalized models10_{, it seems still possible to} make some inferences. In this regard, the theory makes explicit assumptions and inferences with regard to the economic agents, so do for entrepreneurs, I argue.

Within the neoclassical framework, the main role assigned to the entrepreneurs is to shift the economy towards equilibrium. Entrepreneurs are assumed to operate under perfect competition where they make purely rational decisions in the light of perfect knowledge (Wennekers and Thurik, 1999). Profit prospects constitute the main driving force behind the entrepreneurial action. It has been argued that when there is lack of profit potential, entrepreneurs abstain from engaging in any operation. This is to say that the presence as well as the continuity of entrepreneurial activities is based on the extent of economic gains. The greater the expectations of profit are, the higher the ‘natural attraction’ that induces entrepreneurs to take action (Ricardo, 1817, p: 290).

Drakopoulou-Dodd and Anderson (2007) criticize mainstream economics for its consideration of entrepreneurship as a very narrowly defined activity taking its grounds from the atomized individual perspective. Despite the significant evidence to the contrary, it is claimed, the roots of entrepreneurial behaviors are assumed to lie deep in the rational, instrumental and material components. Similarly, according to Zafirovki (1999), neoclassical framework falls into the trap of ‘naturalistic fallacy’ due to the fact that entrepreneurship is considered as a merely ‘innate, universal and technical activity’. He continues his arguments by stating that a ‘psychological fallacy’ is also present in the fundamentals of economics, for it assumes entrepreneurship is achieved through the conduct of individual efforts that are entirely logical, egoistic, and rational. This perspective, he argues, indisputably rules out the social aspect inherently exist in the entrepreneurial action and imposes ‘an asocial or even an anti-social’ character. Banks (2006) interprets this kind of individualism adopted within ‘crude’ economics theory as an extreme instance where it is used, he argues, as synonym with making money.

10_{Theoretical models aiming at including entrepreneurship in neoclassical growth models without disturbing}

(23)

Pure rationality assumption is also challenged by Keynes who makes use of the term ‘animal spirits’. By this notion he means, human actions in general, and the economic activities in particular are partly stimulated by waves of optimism and pessimism. This argument reminds us to reconsider the issue in hand and draws attention to the presence of non-rational motivations of economic actions including entrepreneurship. In this regard, he conveys, if we let the entrepreneurship wholly dependent on mathematical calculations and expectations, it soon “will fade and die” (Keynes, 1960, p: 161).

As articulated initially, the point I want to argue here is that entrepreneurship is evidently a social conduct in the sense that it is involved in, and hence is understood through its ties with the social context. As a result, it is not an atomized activity with only rational and material constituents, I claim. Rather, it involves a set of complementary factors such as rational and non-rational, economic and non-economic etc. founded on broader social environment that in turn specifies the basics of it. The end formation is by no means an outcome generated through one-way connectedness; instead things are drawn from, given back to, and shared with the societal settings (Drakopoulou-Dodd and Anderson, 2007).

Furthermore, it is the interaction between the ingredients of the society; that is individuals and social communities that account for and render entrepreneurship meaningful (Ebner, 2005). It is also noteworthy to point out that there is no room for either under-socialization or over-socialization argumentations in entrepreneurship. This is to say that entrepreneurship is freed from the pressures of both social and individual determinism (Granovetter, 1985). Consequently, as stated by Zafirovski (1999, p: 354) “entrepreneurship, development and related economic activities are primarily complex social processes”. Therefore, viewing entrepreneurship as a socially-constructed phenomenon, but not a socially-determined one, ensures the consideration of both sides without granting any privilege to any of two.

(24)

(25)

4. An Extension to Baumol: Towards Disentangling the Causes

from Consequences

Douhan (2008a, p: 5) states “it is important to realize that different types of entrepreneurship all refer to a function rather than an individual. The same individual could be engaged in both productive and non-productive11_{activities at the same time.” Furthermore,}_{Sauka (2008)}

points exactly to the same argument, and by citing various studies he articulates that productive and rent-seeking activities indisputably coexist within single identities whether it be enterprise or entrepreneur. Implicitly, these studies whose fundamentals draw on Baumol’s 1990 article call for a need to differentiate between the entrepreneurs and the activities they undertake. Therefore, I take this perspective as the departure point of the extension to Baumol’s work that is proposed in the context of this study.

More clearly, in the talent allocation studies discussed above12_{, authors assume that the type of} the activity (either productive or unproductive/destructive) that economic agents engaging in also describes the very nature of these agents. In other words, provided that an entrepreneur contributes towards social and economic well-being of others by means of her activities, she is labeled as productive entrepreneur, or otherwise. Shortly, the assumed situation is like this:

Productive Entrepreneurial Activities = Productive Entrepreneur

Unproductive/Destructive Entrepreneurial Activities= Unproductive/Destructive Entrepreneur

This extension at its very basics loosens or to some extent challenges this presumption and suggests that the state of affairs is more complex than this assumption argues. In this regard, I make a clear distinction between entrepreneurs and activities. Following this differentiation, type of the entrepreneurs will be no longer named as productive vs. unproductive/destructive since a rather different perspective is employed in this extension. They will be categorized as entrepreneurially-talented individuals holding varying degrees of either altruistic attitudes (AA) or exploitative attitudes (EA).

11_{Both unproductive and destructive.}

(26)

In fact, Monroe (1994) locates any human being on a continuum where the extreme points of the continuum indicate purely altruistic and purely selfish tendencies. Between these points, he argues, intermediate forms of behavior are possible with varying degrees of altruism or selfishness. For example, pure altruism is defined as the “intention to benefit another, even when doing so may risk or entail some sacrifice the welfare of the actor” (Monroe, 1994, p: 862). The proposition of Monroe can be well extended to include purely exploitative attitudes, of which I think, a much stronger case than being self-interested once their potential damage to others have been considered. Hence, similar to Monroe, in this extension, I locate entrepreneurs on a continuum where the poles corresponds purely altruistic and purely exploitative attitudes. In this regard, Tan et al. (2005) conceptualize six different types of (social) entrepreneurs depending on the level of altruism they may expose on others. Namely these are:

1) “The person who attempts to innovatively profit society alone, in a way that involves that society, at risk of personal loss.”

2) “The person who attempts to innovatively profit society alone, in a way that involves that society, at risk of foregoing personal profit.”

3) “The person who attempts to innovatively profit society by profiting himself, in a way that involves that society, at risk of incurring personal loss.”

4) “The person who attempts to innovatively profit society by profiting himself, in a way that involves that society, at risk of forgoing personal profit.”

5) “The person who attempts to innovatively profit himself by profiting society, in a way that involves that society, at risk of personal loss.”

6) “ The person who attempts to innovatively profit himself by profiting society, in a way that involves that society, at risk of foregoing personal profit.” (Tan et al. 2005, p: 359)

Given that entrepreneurs are placed on altruistic-exploitative continuum, this Baumol-extension also suggests that productive outcomes can be generated both from AA and EA, though the incidence of the former case seems more likely to occur. However, this possibility does not refute the view put forward. Similarly, entrepreneurs possessing AA may give rise to undesirable social outcomes by means of unproductive/destructive activities, I propose.

Figure 1 clearly illustrates the proposed model.

(27)

Consider an enterprise operating, let say in a high-growth industry (e.g. possessing the prospect of significant job creation in the early years of establishment13_{). And further assume}

that this firm, besides creating many job opportunities, gives rise to negative externalities to the society in the form of environmental pollution (be it water, air or even noise pollution). This narrative is not an unrealistic one since, over the last decade, public and governmental concerns have highly increased due to extensive environmental damage stemming from industrial operations. The point I want to make here is that, within the same enterprise established by a particular entrepreneur, we witness both productive and destructive activities. Without any doubt, the same individual causes both good and bad outcomes to come true. As a result, I consider this case as an instance where AA (as an intent for generating job opportunities) produces destructive outcomes (environmental pollution). This corresponds to Case 1a relation in Figure 1.

Secondly, let’s illustrate how EA may engender positive outcomes. As it is well-known, many multinational companies transfer their production activities to countries to exploit abundant cheap labor opportunities. Consider a subsidiary of a multinational enterprise having production facilities and employing a few hundreds Chinese there. The entrepreneurial talent behind this intent is clearly, to my opinion, exploitative. Adult and child labor are employed

(28)

with lower wages, and even they sometimes work under unhealthy conditions14_{. However, I} consider this occurrence as a case producing productive outcomes within the Chinese context. The rationale behind my thinking is that these people will be possibly out of work force if such investments are missing since there are not many alternative jobs that are presently available. This is an argument also suggested by (Wolf, 2004) who indicates that little is always better than nothing. Again, in this case productive (job generation) and destructive activities (unwanted health problems within employees) coexist within the same enterprise, initiated and conducted by the same entrepreneur. However, the order of the effect is reversed in the sense that EA (as an intent exploiting cheap labor) produces productive outcomes (job generation). This corresponds to Case 2b relation in Figure 1 depicted above.

In the above model, arrows demonstrated via Case 1b and Case 2a account for the other two alternatives that can be realized. The occurrence of these two cases compared to others is more likely as indicated before, and the paper by no means argues the opposite. Rather, I emphasize that we should also consider the possibility of other alternatives. This line of argumentation is also backed up by the literature focusing on the attitude-behavior (in) consistency issue.

Individuals’ attitudes as psychological constructs receive a fair amount of interest due to the belief that an attitude may constitute precedents of the behaviours15_{. Thus, attitudes are} viewed as essential constructs to predict and explain overt human actions. However, existing literature fails to provide one-to-one correspondence between the attitudes and observed behaviors in the sense that attitude-behavior consistency is not always present in the real life examples. In a recent meta-analysis of attitude-behavior relation, Glasman and Albarracın (2006) report that correlations between attitudes and actual behaviors range from 0.20 (Leippe and Elkin, 1987) to 0.73 (Fazio and Williams, 1986) The basic inference that can be drawn from this occurrence is that people sometimes behave incoherently with respect to their attitudes. For instance, in our case, the implication is that a person possessing altruistic attitudes may not engage in activities generating beneficial outcomes for other people. This is also valid for exploitative attitudes, therefore indicating the fact that the outcomes of the exploitative individuals’ actions do not always harm others.

14_{See Locke (2003) for detailed information.}

(29)

Given the large variability in attitude-behavior consistency, scholars have attempted to provide further reasonings to what extent attitudes can predict the observed behaviours. In this regard, other features of the individual who carries out the behavior, the situation in which it is realized and the characteristics of the attitude itself have been proposed as additional factors moderating the attitude-behavior relationship. For instance, self-monitoring as a personal trait is considered an important determinant that specifies the degree to which behavior is consistent with the attitudes (Snyder, 1974). Snyder argues that people possessing high monitor trait realize their behaviours on the basis of situational factors. As a result, high self-monitoring causes attitude-behavior consistency to be particularly low. Another personal trait that affects this relation is self-consciousness (Fenigstein et al., 1975). It is suggested that individuals differ from each other in the way that their attention is focused on inner thoughts and feelings. This situation corresponds being privately self-conscious. The opposite is so-called public self-consciousness which aims at creating a good impression in the eyes of others. Therefore, it is the presence of private self-consciousness that increases the level of consistency between attitude and overt behavior.

Concerning the situational determinants of the attitude-behavior consistency, Abelson (1982) propose that there are some circumstances (in his terms: individuation) which shift the self-focus of the individuals towards their inner constructs such as attitudes. Therefore, when people perform behaviours under individuating conditions (e.g. lack of environmental constraint), cohesion between attitudes and behaviours is expected to be greater. Besides, the opposite case is also a possible outcome, of which deindividuating conditions give rise to lower levels of attitude-behavior consistency. Finally, it is also worthwhile to mention that the features of the attitude itself also affect its relation with the overt behavior. For instance, attitudes that are developed directly by means of a past behavioral experience induce greater attitude-behavior consistency (see Kraus, 1995 for a meta-analysis). It can be exemplified in our case as follows; if a person develops his exploitative attitudes by involving in real instances of exploitative activities, then future behaviors will be guided more intensely by attitudes.

(30)

theoretically justified and represent a promising extension to the Baumol-like entrepreneurship.

Accordingly, I deem it important to make a follow-up regarding the main points that have been highlighted after constructing the first hypothesis of the paper. Such an integrated perspective is necessary because these points constitute a sound base for the second hypothesis of the study. Firstly, it has been argued that entrepreneurship is a socially-constructed phenomenon as an individualistic perspective fails to explain its complex societal structure. Secondly, an extension to Baumol (1990) is proposed where entrepreneurs are no longer classified as productive vs. unproductive, instead, I have taken a step forward so as to humanize16_{entrepreneurs. This indicates a categorization where they are labeled as talented} individuals possessing varying levels of altruistic or exploitative attitudes. And finally, it is suggested and additionally illustrated by real life instances that both altruistic and exploitative attitudes may cause good and bad outcome for the society.

Now, a wise reader may question what shapes the personality -altruistic vs. exploitative- of our unit of interest; that is entrepreneur.

Chell et al. (1991) proposes a model of entrepreneurial personality and behavior demonstrated in Figure 2, arguing that “the entrepreneur can not be isolated from the context.” Her suggestions theoretically rooted in social constructionism approach that is a widely utilized paradigm in sociology, social psychology and other humanistic sciences. Basically, social constructionism considers how social phenomena develop in social contexts.

Social constructionism deals with entirety, giving meaning to personality, behavior and related characteristics of the individual in their social context (Reason, 1981). The essence of individuals can be understood, it is assumed, not by means of isolating them from environment, but rather assessing how they create their own reality and make sense of it in relation with the social context (Weick, 1995). Therefore, in the context of the model presented above, Chell (2000, p: 66) states that “… entrepreneur as an active agent who

16_{The word ‘humanize’ is made use of here to mean that “I have used a behavioral dimension (includes both}

(31)

shapes his or her own reality…” with reference to the social environment, of which he belongs to.

Figure 2: “The entrepreneur can not be isolated from the context”. (Bouchikhi (1993), cited in Chell 2000)

When entrepreneur is claimed to be socially-constructed, my intention is in no way to mean that it is socially desirable as well. In contrast, being embedded in the social context, I argue, entrepreneur can be either purely altruistic or purely exploitative or somewhere between and this is a natural consequence of being a social and shared phenomenon. In the view of social constructionism, as discussed, they create their own realities, beliefs and self-being in connection with environments. They are neither absolute slaves of the alleged economic rationality nor the pure devotees of social context. Instead, they present a harmony, of which the end shape was determined by their own understanding, perception and the creation of the reality of the world they belong to. In this regard, they are not any more or less than an ordinary human being endowed with a wide range of human, either desired or undesired, attributes: being generous, unfaithful, independence, interdependence, dishonest and alike. Consequently, as socially-constructed identities, I argue, it is up to the entrepreneurs’ freewill to place themselves on a certain place on the continuum proposed in the Baumol extension. Therefore, based on the following theoretical justifications and Baumol extension, the following hypothesis is formulated.

Hypothesis 2: Entrepreneurs holding exploitative (altruistic) attitudes towards others invest

(32)

5. Methodology

5.1. Proxies for the Explanatory Variables

In order to test the hypotheses constructed in the previous section, we need proxies measuring the general entrepreneurial talent (for hypothesis 1) and altruistic vs. exploitative preferences (for hypothesis 2).

5.1.1. General Entrepreneurial Talent

A fair amount of work in the entrepreneurship literature has been devoted to the exploration of factors why some people and not others engage in entrepreneurial activities (Martinelli, 1994). In this regard, various determinants of entrepreneurial dispositions have been offered. Among these constructs, self-efficacy is considered as a key internal asset for the manifestation of entrepreneurship. In its most basic sense, self-efficacy means a personal feel of power to act. A term coined by Bendura (1986, p:391; 1997), self-efficacy constitutes “people’s judgments of their capabilities to organize and execute courses of action required to produce given attainments” and have the consequence that “people’s level of motivation, affective states, and actions are based more on what they believe than on what is objectively true.” Self-efficacy is of particular importance to entrepreneurship since it influences an individual’s perceptions of his own abilities whether he is competent enough to successfully execute the entrepreneurial tasks (Boyd and Vozikis, 1994; De Noble et al. 1999).

(33)

al. (1999) uncover that entrepreneurial intentions of the individuals are positively influenced by the varying extents of ESE.

Consequently, drawing on the common use of ESE in the literature, it will be adopted as the main entrepreneurial talent in the context of this study. It is measured by a set of six questions developed by Wilson et al. (2007), and based on the expert interviews with business leaders. The respondents answer items by rating their ESE levels on a 5-point Likert scale. Final ESE score is obtained by averaging the responses to each item. In this regard, Cronbach’s alpha that shows the internal reliability of these six items is 0.82 indicating an acceptable level of reliability17_.

Besides ESE, risk-taking propensity is also considered as an important dimension of entrepreneurial activities. In this regard, the way risk is perceived and managed by entrepreneurs is essential to achieve success in their endeavors (Delmar, 1994). Generally, relevant literature is of the opinion that entrepreneurs have a higher risk taking propensity. Individuals who can deal with the presence of higher levels of risk are likely to ‘self-select’ into the entrepreneurial operations, while risk-averse people choose others forms of occupations (e.g., Begley and Boyd, 1987; Stewart et al., 1999). Accordingly, Zhao et al. (2005, p: 1269) state that people “…who report a higher propensity to take risks judge themselves to be more capable of performing tasks related to entrepreneurship” Therefore, risk-taking propensity can be viewed as a distinguishing feature of entrepreneurs compared to the rest of the population.

Based on this discussion, general risk-taking propensity (GRP) will be adopted as a second measure of entrepreneurial talent. The basic rationale to use more than one proxy for the entrepreneurial talent is to ensure that the results are reliable and are not sensitive to the choice of the variable. GRP is measured by means of the questionnaire items developed by Zhao et al. (2005). Respondents rate their risk-taking propensities by indicating the level of agreement or disagreement with each of the six statements on a 5-point Likert scale. The overall level of GRP is generated by averaging the responses to each item (Cronbach’s alpha=0.68).

(34)

Korunka et al. (2003) considers locus of control (LOC) as a personality characteristic that is strongly attributed to the entrepreneurs. It is argued that individuals who exhibit higher levels of internal LOC have control over the events in their lives. The outcomes are viewed to be the results of their own behaviors and actions. In contrast, people with a high level of external LOC are of the opinion that their lives are under intense control of others, fate, luck or other forces that cannot be resisted (Rotter, 1966). In relation with entrepreneurial literature, Brockhaus (1975) finds empirical evidences indicating that locus of control is a predictor of entrepreneurial intentions. Another work of the same scholar, Brockhaus (1980), also points out the similar direction and suggests that successful entrepreneurs exhibit higher levels of internal LOC. Furthermore, the results of a more recent study, Korunka et al. (2003), indicate that during the startup phase of a new venture creation, the presence of high LOC is functional in completing the entrepreneurial tasks successfully. Consequently, based on the relevant literature, locus of control appears to be an appropriate measure for the general entrepreneurial talent in this study. Related to this, Boone et al (1996) have developed a 39-item instrument based on the widely-used Rotter I-E scale (Rotter, 1966) to assess the level of LOC of the individuals. (Cronbach’s alpha=0.69). As a result, I have used the items generated in Boone et al (1996) to measure the LOC levels of the respondents as a proxy of entrepreneurial talent.

5.1.2. Exploitative Attitudes

In order to measure to what extent people exhibit altruistic vs. exploitative preferences in their actions, exploitative manipulative amoral dishonesty (EMAD) scale is used. People who score high on this scale, it is suggested, are prone to improve their own circumstances by exploiting or disadvantaging others. As indicated before, this study locates individuals on a continuum where the two end points correspond to purely altruistic and exploitative behaviours. Therefore, to the extent that respondents score low on this scale, they are viewed as having altruistic preferences with regard to others.

EMAD is measured by means of a psychological questionnaire constructed by Altmeyer (2004) as discussed in Rogers et al. (2006). It consists of 20 items18_{where individuals rate} their exploitative tendencies on a 9-point Likert scale. (Cronbach’s alpha=0.88)

(35)

5.2. The Experiment

This section discusses how the necessary data is obtained by means of a lab experiment.

5.2.1. Decision task

We confront the participants with an entrepreneurial investment decision with risky payoffs and positive or negative externalities. The basic investment decision is based on one of the earliest notions of entrepreneurial behavior by Richard Cantillon (1755), who identified the willingness to bear the personal financial risk of a venture as the defining characteristic of an entrepreneur. In the experiment, participants receive a fixed budget and can then either invest an amount into a coin toss , with , or do not invest at all

and simply keep the budget as a safe income. If participants decide to invest into the coin toss, they receive a multiple of the invested amount, (with ), as a return R to their investment, if they win, and nothing if they lose.

Hence, their expected payoff from the risky investment decision is

with a probability and . We used the following four multiples in the

experiment: , , and .

(36)

investments in Group A to positive or negative externalities on Group B. Participants know from the start to which group they belong.

In order to increase the number of observations in Group A, , we do not match the participants 1:1 by splitting the participants into two equal groups, , but limit the size of Group B to participants. The participants in Group B equally share the average positive or negative externality of the participants in Group A. In order to avoid bankruptcy, all participants in Group B receive an initial budget . This defines the total payoff of a participant in Group B, , as shown below. Essentially it is a function of the average investment success, or return , of the participants in Group A and an ‘impact factor’

, which determines the sign and intensity of the externality on Group B.

In the experiment the impact factor could take nine different values from to in steps of 0.25: -1.0, 0.75, -0.5, -0.25, 0.0, +0.25, +0.5, +0.75, +1.0 Group A’s investment success exerts negative externalities on Group B if and positive externalities if . If , no externalities exist.

In total, the four investment multiples and the nine impact factors α define 36 entrepreneurial investment decisions with differing returns to investment and externalities. Table 1 provides an overview of all 36 decisions. In the following we will refer to the different combinations of investment multiples and impact factors as ‘scenarios’. Appendix 13 shows an example of a decision screen with one of the 36 scenarios.

(37)

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1 6 4 2 1,75 Risk neutral No Externality Impact / Externality: Mul-tiple 5.2.2. Participants

Based on the theoretical literature, the entrepreneurially talented people, present in any population, will self-select into the diverse set of activities available to them in that society. We therefore conducted the experiments with a diverse group of students for whom it is a priori unclear what activities the existing entrepreneurial talent among them will select into. Despite the potential benefits of focusing on selected professional groups for experiments (Alevy et al. 2007, Drehman et al. 2005, Elston et al. 2006), for a direct test of Baumol’s (1990) talent allocation model, there are at least three reasons why established entrepreneurs are not a suitable population for our experiment. First, established entrepreneurs already made their occupational choice and would therefore bias our sample on one of the possible outcomes of entrepreneurial talent. Second, focusing on current entrepreneurs would suffer from a survival bias, as survival depends on many things different from the individual's talent. Third, it is not the real talent, but the subjectively perceived talent that determines the a priori choice and initial allocation we are interested in. We therefore classify participants according to their subjective perception of their talent to perform activities required to become and be an entrepreneur, i.e. their entrepreneurial self-efficacy.

(38)

5.2.3. Procedures

The sessions were computerized, using a program written with z-tree (Fischbacher, 1999). Participants were seated in a random order at PCs, separated from each other by visual screens. Instructions were then distributed and questions were answered in private.

At the beginning of the experiment, participants were randomly assigned to Group A or Group B and privately informed about their assignment on one of the entry screens. Additionally, on each of the 36 decision screens, there was a reminder of their group assignment. Participants stayed in their group throughout the whole session.

The sequence of the 36 decisions screens was randomly determined for each participant individually. The outcomes of the coin tosses were revealed at the very end of the session. To keep participants of Group B busy, we asked them to go through the same set of 36 investment decision screens as Group A. Group B, however, was informed that their decisions are purely hypothetical, and that their own payoffs are solely determined by Group A. As Group B’s decisions are not incentive compatible, we do not include them in our analysis and all results reported in this paper refer to Group A only.

Earnings in Group A were determined by the average payoff of four individually and randomly selected scenarios. To keep the variation of payoffs manageable, and also to stress that all investments multiples are equally important, we randomly selected one decision per multiple. Earnings in Group B were determined by only one randomly and separately selected scenario. For example, if the scenario with a multiple of 2 and an impact factor of +0.5 was randomly selected, all respective investment decisions and winnings of Group A, irrespective of their position in the individual decision sequences, were used to calculate Group B’s payoff. Selecting only one out of the 36 scenarios prevented potential strategizing by Group A on averaging effects across several positive and negative externalities. Further, it ensured the maximum variance of externalities on Group B’s payoffs.

BAUMOL-LIKE ENTREPRENEURSHIP: AN EMPIRICAL ASSESSMENT AND AN EXTENSION