Neoclassical Entrepreneurship Theory: Limits and Insights for a. Heterodox Approach

Transcription

1 Neoclassical Entrepreneurship Theory: Limits and Insights for a Heterodox Approach Gabriel A. Giménez Roche 1 Abstract: Neoclassical approaches explain the distinctiveness of entrepreneurship by watering down perfect rationality to a selective instead of a generalized feature among the population, and complete information to be costly dispersed instead of freely available. But if information remains complete and rationality perfect, the emergence, awareness, and exploitation of profit opportunities are left unexplained as these are exogenously given. This paper aims at a solution by proposing a heterodox entrepreneurship analysis where profit opportunities emerge endogenously as incomplete information, continuously generated by an openended reflexive process of interactions between imperfect but creative agents. Entrepreneurs face the uncertainty thus generated by navigating the institutional dynamics of the market to secure estimated profit opportunities. Keywords: calculation, entrepreneurship, information, institutions, rationality. JEL Codes: B50, D80, L Introduction It has been often pointed out that mainstream economics neglect the fundamental role of entrepreneurship in business creation and market dynamics, concentrating instead on the equilibrium analysis of optimal resource allocation by already existent firms (Baumol 1968, 64; Coase 1992, 714; Coase & Wang 2011, 1 3; Foss & Klein 2012, 15 16; Witt 1999, ). The entrepreneur has no special role beyond that of a mere manager; ingenuity, ruse, innovation, 1 Finance, Economics, and Law Department. Champagne School of Management. 217 avenue Pierre Brossolette, 10002, Troyes France. ggiroche@gmail.com 1

2 charisma, and even the profit incentive become irrelevant for economic analysis (Baumol 1968, 67; Casson 2003, 79; Foss & Klein 2012, 25). Notwithstanding the critiques, there is a corpus of neoclassical theories on entrepreneurship (Baumol 1993, 197). These theories partially drop the assumptions of generalized perfect rationality and costless access to complete information to explain the distinctiveness and necessity of the entrepreneurial function. They assume perfect rationality to be limited to some individuals, and complete information about profit opportunities to be costly dispersed in the market. In this manner, the entrepreneurial function can involve product innovation (Schumpeter 1983), uncertainty-bearing (Kihlstrom & Laffont 1979; Knight 2006), giving completeness to production functions (Leibenstein 1978), alertness to profit opportunities (Kirzner 1973), organization of team production (Alchian & Demsetz 1972; Barzel 1987a; 1987b), market-making (Casson 2003), leadership (Witt 1998), or a combination of these (Baumol 2010). If profit opportunities constitute the object of entrepreneurial action, then a theory aiming at explaining entrepreneurship must concentrate on the emergence, awareness, and exploitation of profit opportunities (Shane & Venkataraman 2000, 218). The problem with neoclassical theories of entrepreneurship is their lingering attachment to perfect rationality and structurally complete information. Considering profit opportunities to be part of structurally complete information must imply that: 1) opportunities do not emerge, but are exogenously given information to be discovered; 2) perfectly rational awareness of opportunities is a matter of objective perception rather than subjective estimation; and 3) exploitation of opportunities is limited to resource reallocation. Despite these implications, neoclassical entrepreneurship analysis is not wrong per se. It complements imperfect competition theories of the firm, plausibly explaining the differentiation process in imperfect competition theory, but not entrepreneurship 2

3 itself. Although this proves to be a limitation of neoclassical entrepreneurship theories, a solution to it could prove a solid foundation for a heterodox theory of entrepreneurship. The essence of entrepreneurship is to continuously negate the pressures of scarcity and entropy by pushing economic boundaries outward to a sort of increasing returns to scale situation (Boulding 1966, 5; Kaldor 1972; 1975; Marshall 1890, , ; Marshall & Marshall 1879, 23ff.). This implies the presence of creative agents participating in a recursively extensive and complex market process, where incomplete information is continuously generated (Giménez Roche 2016, 703ff.). In this open-ended world of unknowable future information, perfect rationality is displaced by imperfect rationality implying that: 1) profit opportunities are imagined, subjective estimations that might objectively emerge out of the interacting actions not only of entrepreneurs, but also of nonentrepreneurial agents; 2) awareness of opportunities is not a matter of objective perception but of subjective estimation of evolving social processes leading to the potential emergence of the price discrepancies constituting these opportunities; and 3) exploitation of opportunities is the speculative selection and deployment of resources directed at entrepreneurial estimations. After briefly defining the entrepreneurial function and the neoclassical nature of mainstream economics, this article examines the emergence, awareness, and exploitation of profit opportunities in the neoclassical theory of entrepreneurship. This allows for a brief exposition of the main neoclassical contributions to entrepreneurship theory, and their limitations. These limitations thus serve as a starting point for determining the foundations of a heterodox entrepreneurship theory capable of explaining the emergence, awareness, and exploitation of opportunities. 2. Entrepreneurship and the mainstream 3

4 2.1 The entrepreneurial function While the economic function of factors of production like labour and capital rests on their transformative powers, the economic function of entrepreneurship is defined by its coordination and direction of factors of production towards solving the scarcity problem. This function consists of the enactment of entrepreneurial judgment about profit opportunities, that is, price discrepancies between inputs and outputs (Foss & Klein 2012, 78; Kaldor 1934, 67 68). Since these price discrepancies occur through time, and are subject to other people s actions that are beyond entrepreneurial control, then entrepreneurship is about action under the context of uncertainty (Foss et al. 2007, 1895; Foss & Klein 2012, 20; Mises 1998, 254). Judgment about the possible existence and exploitability of a price discrepancy by means of the coordination and direction of factors of production necessarily precedes the actual undertaking of the enterprise, and the confirmation of its success. Consequently, entrepreneurship theory cannot be limited to explaining the exploitation of exogenously given opportunities, but must also explain the emergence and awareness of these opportunities. Therefore, entrepreneurship theory must explain the entrepreneurial function relative to the emergence, awareness, and exploitation of profit opportunities. 2.2 The neoclassical character of the mainstream The neoclassical element of mainstream economics is defined by the ontological assumptions of perfect rationality (i.e., agents are always capable of optimal decision-making) and structurally complete information (Jevons 1965, 3 5, 37; Walras 1926, 29 30). The use of mathematical 4

5 methodology directed at optimal equilibrium analysis is just a natural consequence of those ontological assumptions. 2 The influence of less orthodox contributions coming from Friedrich A. Hayek (1945) and Herbert A. Simon (1955) has watered down neoclassical ontological assumptions into the notions of rational expectations, information asymmetry, and efficient-market hypothesis (Akerlof 1970; Fama 1970; Lucas & Prescott 1971; Muth 1961; Stiglitz 1975). The idea is that agents retain their perfect rationality and information is still structurally complete, but their access to this information is costly and imperfect. Error becomes thus possible, but it is the result of a faulty access to information, not of a faulty rationality; it is exogenously provoked. Given time, error is eliminated or neutralized through learning, and equilibrium becomes attainable. Therefore, the neoclassical exposition is essentially a description of the final stage of a run out process in which causal claims to the resulting end-state are irrelevant (Hahn 1973; Kirzner 1973, 90 92; Knight 2006, 17 18). 3. Neoclassical emergence of profit opportunities Neoclassical entrepreneurship theories were born out of the necessity to explain the dynamics of decision-making in firms. Based on the representative firm of Alfred Marshall (1890, 115ff.) and the equilibrium firm of A. C. Pigou (1928), the theory of the firm developed by Edward H. Chamberlin (1933) and Joan Robinson (1933) describes a world where all supply and demand 2 Equilibrium has different meanings and not all these meanings imply perfect rationality nor structurally complete information. For instance, both Frank Hahn (1973) and Fritz Machlup (1958) consider that an economy is in equilibrium when market agents do not change their market routines in reaction to perceived information, their actions thus being mutually consistent. This definition does not preclude change in information nor in behaviour. Moreover, agents can both generate and learn new information that may entail changes in their routines. Finally, since equilibrium routines depend on an agent s perception of information, error is possible within and outside equilibrium. 5

6 information is available to the firm. 3 In static terms, this means that technology and preferences do not change. In dynamic terms, whatever the initial conditions of the economy are, the firm is fully able to know how variables in production and utility functions will eventually change. Thus, if all agents, firms and households, are perfectly rational with complete information, the existence of profit opportunities must be structurally exogenous, as creativity, uncertainty, and error are impossible in such a world. This set up resolves into a situation of zero profits as optimisation resumes. Chamberlin and Robinson introduced differentiation, which is based on a defective access by either firms or households to otherwise structurally complete information, to explain the persistence of competition and profits. The result is monopolistic, or imperfect, competition, which leaves an opening for entrepreneurship. 3.1 The absence of creativity Neoclassical entrepreneurship theory aims at explaining how certain firms are better than others in imperfect competition. If access to complete information is defective, then the distinctive character of neoclassical entrepreneurship rests on the ability to provide the firm with a less defective access to this information. For instance, in Harvey Leibenstein s X-efficiency theory, routine entrepreneurs (i.e., managers), although capable of obtaining individual optimal input performance, cannot optimize output performance due to incompletely specified production functions (Leibenstein 1966, ; 1969, 602 3). As a solution to this X-inefficiency i.e., the difference between optimal and suboptimal output performance Leibenstein proposes the N-entrepreneurship concept. N-entrepreneurs provide the information on how to fill the gap and fully specify production functions, thus enabling input coordination and optimal output 3 For more on the evolution of the neoclassical theory of the firm, see Scott Moss (1984) and Frank M. Machovec (1995). 6

7 performance (Leibenstein 1968, 73 75). This gap-filling notion of entrepreneurship is also described by Robert E. Lucas (1978), but he emphasizes the role of entrepreneurship in delimiting the size of firms (see section 3.3 below). Other economists consider that entrepreneurship is less about gap-filling deficientlyspecified production functions than about defining whole production functions. For instance, Israel M. Kirzner (1973, 32 36; 2000, 15 17) criticizes the homo oeconomicus concept for being a perfectly rational optimizer of given means-ends frameworks, and instead proposes his entrepreneurial homo agens, someone who defines which means-ends framework to be adopted. For Kirzner (1985a, 16 17), optimization is secondary because it follows the determination of production functions. Similarly, Mark Casson (2003, 14 16) considers that the managerial optimization of the material-transformative activities of the firm is subordinated to entrepreneurial market-making activities i.e., contract-brokering with suppliers and customers determining first which material-transformative activities to optimize. In its turn, the entrepreneur-innovator of Joseph A. Schumpeter disrupts an existing market equilibrium by using information on new combinations (i.e., production functions) about existing although not yet exploited new products, methods, technology, markets, and/or trade organizations (Schumpeter 1983, 66, 74 78). Theodore W. Schultz takes a similar position to Schumpeter s by assuming that entrepreneurship is about being a pioneer in learning and implementing new technical and organizational advances (Schultz 1975, 831). Nevertheless, neoclassical entrepreneurship theory fails to explain the role of creativity in the emergence of profit opportunities. Although Schumpeter and Schultz recognize the necessity of novelty for the emergence of profit opportunities, they fail to link entrepreneurial innovativeness to creativity. The novelties introduced by the entrepreneur are developed by specialized laborers, that is, inventors, responsible for generating new information on new 7

8 products, methods, and organizations (Schumpeter 1983, 88, ). However, this generation of new information is structurally exogenous since neither Schultz nor Schumpeter provide any explanation for it (1975, 832). 3.2 Uncertainty and error as the result of dispersion of complete information Neoclassical entrepreneurship theories often link profit opportunities to uncertainty and error. For Frank H. Knight (2006, 3 21, 225), profit opportunities are events so unique that they cannot be predicted in any a priori manner for lack of knowledge of their objective probability distributions, thus resulting in a situation of pure uncertainty. It is this pure uncertainty that makes the emergence of profit opportunities possible otherwise, all income could be reduced to rents equivalent to the marginal productivities of all involved factors. However, the Knightian entrepreneur remains perfectly rational for the problem is one of ignorance, not of a cognitive competence gap. Knight considers that one could even know all the outcomes possible for a course of action, but ignorance of their actual probabilities prevents one from optimizing production (Knight 2006, 35 38). Therefore, under conditions of pure uncertainty, perfectly rational entrepreneurs must resort to subjective estimates of the future, making their decision-making prone to error and allowing for price discrepancies to emerge (Knight 2006, 198). Kirzner (1973, 85 86) also considers that uncertainty is born out of the uniqueness of future events and that this results in error in the pricing of inputs and outputs. However, profit opportunities also depend on the ignorance by most people of the very existence of these opportunities, because the otherwise widespread knowledge of them would eliminate their uniqueness and hence uncertainty (Kirzner 1973, 78). Casson (2003, 119ff.; ), on the other hand, considers that ignorance derives not from uniqueness, but from the information 8

9 and transaction costs surrounding profit opportunities, thus resulting in divergent beliefs regarding their viability. Nevertheless, these expositions never link uncertainty and error to creativity, or to a competence gap. Perfect rationality remains present, it is just bounded by the ignorance of dispersed information among agents. Moreover, it is ignorance of knowable dispersed information, that is, existent information, so that the future is already implied in the present (Buchanan & Vanberg 2001, 380). Even when uncertainty is associate with changes, these are assumed to be structurally exogenous, never cognitively endogenous to creative agents (Buchanan & Di Pierro 1980, 693). If creativity were involved, future information would be unknowable because non-existent in the present. Dispersion of the eventually emergent information would exacerbate the unknowability problem instead of constituting that problem. Certainly, learning could eliminate the dispersion problem of knowable information, but in the presence of creativity it would also trigger a recursive process generating more unknowable information. 4. Neoclassical awareness of profit opportunities Costless structurally complete information and widespread perfect rationality preclude entrepreneurship in the traditional theory of the firm. Neoclassical entrepreneurship theories reintroduce entrepreneurship by assuming well-informed perfect rationality is not widespread in the economy, so that awareness of profit opportunities is restricted to a few individuals. In X-efficiency theory, for instance, agents are selectively rational (Leibenstein 1968, 75 76; 1975, ). Since pertinent information about the full specification of production functions is costly, routine managers choose to be limitedly rational due to differences in preferences and exposure to institutional constraints (Leibenstein 1968, 73; 1969, 602). Thus, entrepreneurs are constrained by institutions and their preferences drive them toward facing 9

10 those information costs and specialize in specifying production functions and gap-filling if necessary. In Richard E. Kihlstrom and Jean-Jacques Laffont s risk-aversion theory of entrepreneurship, agent rationality is not necessarily constrained by information accessibility. Instead, entrepreneurs are distinguished from nonentrepreneurs by their lower risk-aversion (Kihlstrom & Laffont 1979, 720). Both contractual and residual income possibilities are known to all agents from the outset, however, their estimations of the likelihood of those possibilities differ according to their degree of risk-aversion (Kihlstrom & Laffont 1979, ). Residual income is less risk-weighted by entrepreneurial agents, while being more risk-weighted by nonentrepreneurial ones. Despite this difference, all agents optimize their market behaviour; less risk-averse agents run large enterprises, while more risk-averse agents remain contractual factors or run small enterprises (Kihlstrom & Laffont 1979, 721). William J. Baumol (2010) also assumes differences in risk-aversion to modify the traditional oligopolistic competition model. First, he assumes that large and small firms differ in terms of risk-aversion relative to innovation. Small firms engage in riskier radical innovation, while larger firms prefer safer incremental innovation based on earlier radical innovations (Baumol 2010, 28 30). Larger, established firms are more risk-averse vis-à-vis radical innovation because of their more bureaucratic structure preventing them from adapting their productive and managerial routines to radical innovations (Baumol 2010, 29). Second, Baumol considers that innovative entrepreneurs are so optimistic that they accept a lower remuneration relative to what they could have earned if they remained employees in larger companies. The reason lies on the psychic revenue of the thrill of setting up a start-up akin to a lottery ticket (Baumol 2010, 50 52). Given the low monetary cost of innovation in smaller firms in the face of uncertainty, larger firms outsource radical innovation to them, thus avoiding the risks to their more established 10

11 structures (Baumol 2010, 53, 66). This would explain the role of larger firms as business angels or venture capitalists for smaller start-up firms. If the radical innovation succeeds, it cannot pose a threat to the larger firm backing the start-up, which will develop the innovative product eventually through incremental innovations (Baumol 2010, 65). Moreover, the larger firm has stakes in the start-up that allow it to enjoy radical innovation at a safe distance. For Kirzner (1973, 35; 1985b, 56), it is the display or not of alertness to profit opportunities that distinguish entrepreneurs from non nonentrepreneurs. This alertness refers to an attitude of receptiveness to available (but hitherto overlooked) opportunities [emphasis added]" (Kirzner 2000, 18). This alertness does not constitute an active form of search action, but a passive one because the entrepreneur is assumed to passively discover profit opportunities when alertness is switched on by the presence of previously unnoticed information on price discrepancies (Kirzner 1985a, 29; 1992, 44; 2000, 17 19). On the other hand, Schumpeter and Schultz consider that entrepreneurship is not a matter of awareness of profit opportunities, but a matter of acting upon that awareness. Schumpeter implicitly considers that awareness of profit opportunities is quite generalized, but not the motivation to introduce the market innovations underlying them. Since entrepreneurship is about actually introducing market innovations, anyone in an enterprise be it a firm-owner, a manager, or a simple employee is in the position of becoming an entrepreneur (Schumpeter 1983, 137). 4 Schultz, in his turn, considers that entrepreneurship is about resource reallocation, an activity that is exerted by all agents and not just firm-owners. The problem is not one of detecting profit opportunities, but learning how to exploit them faster than others (Schultz 1975, 831). In time, any knowledge could be mastered by anyone, but the more complex that 4 As such, Schumpeter s view of entrepreneurship is much closer to the modern concepts of intrapreneurship (Pinchot 1985) or corporate entrepreneurship (Burgelman 1983; Schollhammer 1982) than to more traditional views on entrepreneurship. 11

12 knowledge becomes, the harder it is for all agents to master it at the same pace. Therefore, education and experience are crucial for mastering new knowledge and detecting its innovative applications in the market (Schultz 1975, ). Awareness of profit opportunities in neoclassical entrepreneurship theories is always awareness of existing price divergences. Their exogenous nature implies that their existence precedes awareness and hence entrepreneurial action. This explains the emphasis of Casson and Kirzner on profit opportunities as arbitrage opportunities (Casson 2003, 53 65; Kirzner 1973, 48). Arbitrage opportunities are supposedly based on new information consisting either of discoveries or updates of information. Nevertheless, discoveries are simply net additions to one s stock of knowledge derived from previously ignored although existent structural information, while updates refer to information adjustments in reaction to environmental exogenous changes (Casson 2003, 22 24). Even when admitting that profit opportunities concern future market conditions (Kirzner 1985b, 54 59; 1992, 27), these refer to discovery of errors. But error is an ex post concept observable in the present, not the future (Buchanan & Vanberg 2001, 106 7). Neoclassical profit opportunities are objective phenomena. Therefore, profit opportunities are totally independent of entrepreneurial action. They are not generated through market action, but are revealed by some external force. This means that profit opportunities must be limited in number. Consequently, the greater the number of successful entrepreneurs, the lower the number of profit opportunities remaining to be exploited. In the Schumpeterian exposition, the entrepreneur-innovator is followed by a swarm of entrepreneur-imitators once information is revealed by the former on the existence of a profit opportunity (Schumpeter 1983, ). Similarly, the greater the number of N-entrepreneurs, the lower the number of available profit opportunities in X-efficiency theory (Leibenstein 1968, 78 79). Although 12

13 entrepreneurs might differ among themselves at first, as information becomes more generally available through routinization of N-entrepreneurship, those differences disappear (Leibenstein 1968, 81). Kihlstrom and Laffont (1979, ) also estimate that the number of profit opportunities is given and inversely correlated to the number of enterprises. In this manner, the same profit opportunities can be simultaneously discoverable and exploited by more than one entrepreneur until they are fully exhausted (Casson 2003, 43 44). In conclusion, neoclassical entrepreneurship extinguishes profit opportunities instead of generating new information allowing for potentially more opportunities. Ironically, in the neoclassical world, entrepreneurship makes markets more stagnated instead of more dynamic. 5. Neoclassical exploitation of profit opportunities In the neoclassical world of objective profit opportunities, successful exploitation of profits by means of resource (re)allocation would be automatic through spot transactions. This would mean that the factors market would not suffer from information or transaction costs, and that firms would not be necessary to secure profit opportunities. However, as Casson (2003, ) indicates, if the emergence of firms and their diversity is to be explained, then asymmetric and dispersed information must also afflict factors markets. In this manner, Lucas although not explaining entrepreneurship per se considered it to be the determining factor of firm size in contrast to the limited free entry/exit argument of Jacob Viner (1932). For Lucas, entrepreneurial competence in processing relevant information explains: 1) a firm selling in different goods markets instead of just one; 2) that changes in product demand can also be met by size changes in existing firms, and not just by free entry/exit of firms; and 3) firm growth is independent of firm size (Lucas 1978, 509). Unfortunately, any 13

14 explanation on why entrepreneurs differ in their abilities is a matter of exogenous endowments in knowledge. 5 The problem linking transaction costs with entrepreneurship can be traced back to Ronald H. Coase. He saw production within the firm as an alternative to production organized through spot market transactions of independent factors of production in the market (Coase 1937). Faced with transaction costs and uncertainty about factor performance, the firm substitutes long-term contracts for spot transactions thus imposing a hierarchical monitoring over factor performance. In other words, the firm internalizes a futures market for otherwise independently organized factors of production. However, Armen A. Alchian and Harold Demsetz (1972) indicate that the firm is not simply a collection of independent factors of production under a hierarchical umbrella. If this were the case, then no firm would be necessary because the marginal productivity of these factors could be easily metered and monitored through decentralized long-term contracts (Alchian & Demsetz 1972, 777). Instead, a firm is a team of complementary factors of production that yields a collectively-priced marginal product (Alchian & Demsetz 1972, 778). The difficulty resides in that some factors cannot produce as well as monitor other factors performances at the same time. Consequently, moral hazard problems arise with incentives to shirk, freeride, or hold-up the whole firm s production (Alchian & Demsetz 1972, 779; Barzel 1987b, 117). This interdependence problem makes it difficult to individually price factors according to the traditional sequence of marginal product metering 5 On the other hand, 44 years before Lucas, Ncholas Kaldor (1934, 65 67) provided an endogenous explanation of how entrepreneurship delimits firm size. As long as factors are divisible, marginal and scale economies can occur. Only when knowledge is fixed does a factor become indivisible. In this manner, the typical convex, inverse bellshaped marginal cost curve obtains, where after a minimum, cost begins to rise. However, if knowledge improves, as it does in reality, no such situation fully obtains. The assumption of indivisibilities of factors cannot really explain the size limits of a firm. Consequently, if an optimum size is to be found, then at least one of the factors involved must be fixed, thus, limiting the firm size. If the fixed factor in question displays a fixed supply for the firm but not for the industry, such a factor cannot be naturally limited but is actually limited by some peculiarity in the firm's production function. Only one such factor is necessary, more than one being superfluous. 14

15 preceding marginal revenue. Hence the need for an alternative production organization where the pricing sequence is inverted, with the reward system defining productivity within the firm (Alchian & Demsetz 1972, 778; Barzel 1987a, 103). This alternative is entrepreneurial coordination of factors of production. In the team, there are members whose productivity does not depend on economic conjuncture. Metering and monitoring their performance is easier, but needs someone specializing in at least defining metering and monitoring activities: the entrepreneur. Since this person s activity is to direct and coordinate others efforts under conjunctural fluctuations, it is hard to meter and monitor her efforts as it can be confused with good or bad luck (Barzel 1987a, 104). However, the entrepreneur will not have an incentive to shirk as long she holds the following bundle of rights: 1) to be the residual claimant to the team s net earnings; 2) to monitor input performance; 3) to be the central nexus to all input contracts; 4) to alter the membership of the team; and 5) to sell these rights defining ownership of the organization encompassing the team (Alchian & Demsetz 1972, ). The firm becomes therefore a surrogate futures market for not just the factors of production but also the entrepreneur. The contractual pricing of the factors of production stabilizes their remuneration at a given price, protecting them and the entrepreneur against market volatility (Barzel 1987a, 105). Meanwhile, the entrepreneur sells her opportunity searching and coordinating abilities (i.e., monitoring and metering) to the factors of production who pay her with control of their services (Alchian & Demsetz 1972, 793). Consequently, the harder it is to supervise entrepreneurial performance, the more the entrepreneur must recur to self-finance in order to undertake business, and vice-versa (Barzel 1987a, 112). As a complement to the above, Ulrich Witt (1998, 167; 2007, 1128) considers that the essence of entrepreneurship is the capability to communicate one s vision of an imagined profit opportunity. Factors of production need a cognitive framework to know how to select, interpret, 15

16 and apply information concerning production (Witt 2007, 1127). As long as they fully assimilate such a framework, they are fully able to optimize their productive activities. The problem is coordinating different factors under a common cognitive framework (Witt 2007, ). Although firms possess formal communication channels specifying each factor s productive activities, these channels do not necessarily provide them with the motivation to harmonize their personal goals to avoid moral hazard. The entrepreneur provides factors with harmonization through leadership. By means of informal communication channels (e.g., charisma, rhetoric, galvanization, etc.), the entrepreneur provides a cognitive framework congruent to their own personal frameworks while harmonizing diverging personal goals toward the productive end of the firm, avoiding moral hazard (Witt 1998, ; 1999, 104 7; 2007, ). Witt goes as far as determining who is or is not an entrepreneur by describing a market selection process where only those displaying strong leadership qualities become entrepreneurs, while those with weaker leadership qualities remain contractual factors (Witt 1998, 169; 1999, 105). 6. Toward a heterodox theory of entrepreneurship Neoclassical entrepreneurship theories have the merit of departing from traditional neoclassicism somewhat watering down the extreme assumption of generalized perfect rationality and costless access to a structurally complete information. Nevertheless, the merit of these theories constitutes their major limitation as proper entrepreneurship theories. They do not explain why profit opportunities emerge out of information dispersion, and why these opportunities are differently exploited by entrepreneurs. The main contribution of neoclassical entrepreneurship theories is improving the theory of the firm by assuming entrepreneurship as a special factor, but not by explaining entrepreneurship itself. 16

17 Instead of just assuming profit opportunities to be exogenously given, a proper entrepreneurship theory should explain how these opportunities endogenously emerge in the first place. Neoclassical theories are correct in linking profit opportunities to information dispersion, and rationality differences. However, this link is done to the dispersion of structurally complete information, and rationality differences refer to the demographic distribution of perfect rationality. This implies that profit opportunities are still exogenous and limited in number. In other words, the emergence, awareness, and exploitation of profit opportunities take place in a closed-ended world, where no further endogenous generation of opportunities is possible. If the emergence, awareness, and exploitation of profit opportunities are to take place in an open-ended world, where their endogenous and continuous generation is possible, then it is necessary to assume continuous incomplete information, and imperfectly rational individuals. This makes a proper entrepreneurship theory fundamentally heterodox in nature as the orthodox assumptions of neoclassical economics are completely dropped off. 6.1 Emergence of opportunities: Imperfect rationality and reflexivity In all currents in economics, rationality purely stated refers to consistency between means and ends in decision-making (Choi 1993, 20). Perfect rationality is just an extreme case where this consistency is assumed to be optimal (Hodgson 1997b, 668). In this manner, all purposeful actions are rational for they all imply a means-ends framework in decision-making of varying consistency (Mises 1944, ) only non-purposive actions could thus be considered irrational. In the real world, agents are not perfectly rational. They cannot substantively process all available information each time they are confronted with a problem, but only procedurally 17

18 (Simon 1976). Processing all pertinent available information in search of an optimal solution demands too much effort and is extremely time-consuming (Tversky & Kahneman 1974). Furthermore, it would imply agents know perfectly how to look for information, as if they actually knew what they ignore (Kirzner 2000, 8). Instead, agents use their experience, preferences, and estimations about their unravelling environment to create a set of procedures on how to process information (Hayek 1976; Shackle 1974; 1992; Simon 1976). In this manner, they use more modest, although much less time-consuming, subjective procedures to process whatever information they believe is specifically pertinent for a particular problem (Choi 1993, 47 67; Kahneman & Tversky 1979; Tversky & Kahneman 1974; Holland et al. 1989, 29ff.). Agents must be creative in how they select information to be processed because of their competence gaps. This might involve the deliberate use of their existing cognitive frameworks (Finke et al. 1992; Ward 1995; Weisberg 1993), the subconscious blending of existing concepts (Fauconnier & Turner 2002), or the continuous honing of their cognitive frameworks (Gabora 2017). Whatever the procedure, agents can attribute not only existing but also new meanings to the information they process, thus potentially creating new information (Searle 1995). Therefore, valuation of present and future goods depends on the singular set of preferences, experience, and estimations of each market participant. This is the true origin of price discrepancies, and hence of profit opportunities: agents build their market valuations subjectively and thus variably. Of course, agents can learn to adjust their market valuations to each other. But learning is always lagged; it is about making an estimation at t based on t 1 about what is going to happen in t + 1 (Hahn 1973). Unfortunately, there is always something happening between t and t +1 that cannot be integrated into learning (Buchanan & Vanberg 2001, ). This is so because price formation is a recursive feedback market process where the output of one s 18

19 evaluation procedure is the input for another s. The process can result in both positive i.e., reinforcing and negative i.e., disruptive feedbacks, which are the object of further interpretation by market agents, and hence of new information. A reflexive process sets in, where information feedback loops continuously generate previously inexistent information (Soros 2003, 42). Consequently, new profit opportunities can emerge out of previous ones, because individual agents will always find themselves in different parts of the space-time continuum of the market process; always lagging behind new information generated by others, while also generating new information ahead of others. Market evaluation is thus a matter of choice regarding future action (Shackle 1979, 22 27). Once a course of action is chosen, new information is generated through action, which is subject to new information generated by others. Therefore, entrepreneurial action is always crucial to the market process (Shackle 1955, 6). It profoundly changes the market process by both generating new information affecting others actions and decision procedures, and by being affected by others actions in unpredictable ways. In part because of imperfect rationality, in part because one cannot integrate information that is unknowable, that is, future information (Buchanan & Di Pierro 1980, 699). 6.2 Awareness of profit opportunities: Extensiveness, complexity, and uncertainty The entrepreneurial market process can be assimilated to a Boolean network. The greater the number of market participants integrating the market, the greater the potential for new information, and hence for emerging profit opportunities (Beinhocker 2006, 149). This explains why smaller and more isolated markets have lower profit opportunity potential than larger and less isolated markets. The factors contributing to the emergence of profit opportunities are easier to identify in the former, making such markets a decomposable system that is, profit 19

20 opportunities can be reduced to their individual component parts (Simon 1962, 474). If the situation persists, then profit opportunities become a simple matter of quasi-rent appropriation, and the neoclassical view of awareness of objective profit opportunities resumes. However, creativity and market growth entail an increase in the extensiveness (i.e., amount of information) and the complexity (i.e., intricacy between pieces of information) of new information (Hodgson 1997b, ). This exacerbates the competence gap problem of imperfect rationality as reflexivity accelerates due to the feedback loops of more extensive and complex information (Hodgson 1997a, 405; Rosser 1999, ). Consequently, profit opportunities cannot be traced back to its individual constituents independently of their interrelationships (Elsner 2010, 450). These opportunities are phenomena resulting from the specific interdependence between its component parts, not from their mere individual presence (Holland 1997, 32; Humphreys 1997, S341-2). They are thus emergent factors of the market process. Inasmuch as profit opportunities are emergent, the future is not implied in the present, that is, it is not only ignored, it is unknowable and implies therefore the problem of pure uncertainty (Shackle 1983, 33). This pure uncertainty is substantive, that is, it is not just cognitively limited vis-à-vis currently available complex information (Dosi & Egidi 1991, 145). Since information about the future is unknowable at any given moment, the entrepreneur faces the fundamental impossibility of a structurally predetermined state of events (Dequech 2011, 624ff.) that is, no probability distribution is objectively available to be known. The point is, a profit opportunity is a price discrepancy between present factors of production and future goods. It is not a synchronic discrepancy, but a diachronic one (Giménez Roche 2011, ). This diachronicity implies that if the evaluation of goods remains in the future, then it is unknowable information. As such, profit opportunities do not exist objectively. Because of reflexivity, they remain subjective phenomena prior and during entrepreneurial 20

21 action. Their objectivity depends on the potential confluence of actions coming from entrepreneurs, competitors, suppliers, customers, legislators, among others. The problem is that this confluence can only culminate in the future, not in the present of decision-making. Profit opportunities become objective only if the confluence of market actions corroborate a positive price discrepancy accruing to the entrepreneur. As long as this confluence does not culminate, they remain subjective in nature; the fruit of entrepreneurial imagination. Furthermore, profit opportunities are singular events, since they are emergent phenomena resulting from the confluence of a myriad of market actions, and of the entrepreneur s imagination. If this were not the case, then they would result independently of this confluence. Knowledge of which are the component parts of the market process resulting in a specific profit opportunity, would allow to decompose and reaggregate that process in any way possible so that the same opportunity would unravel (Wimsatt 1997, S375 76). However, the reflexivity of imperfect creative rationality makes a profit opportunity specific to a singular organization of market factors (Wimsatt 1997, S373). An agent s market action is a consequence of one s knowledge of the past and estimation of the future, a consequence that impacts others market actions, and so on. A profit opportunity is the result of this unravelling chain of events. If this were reorganized, agents would reflexively act differently, and another profit opportunity would result, if at all. Since this unravelling of the future is not available until agents act, profit opportunities at the time of decision-making are not available to be objectively discovered. Instead, entrepreneurs imagine their current or upcoming existence (Foss & Klein 2012, 75; Klein 2008, 176). In this manner, awareness of profit opportunities in a heterodox theory of entrepreneurship is not a question of perception, but one of imagination. The entrepreneurial cognitive framework coordinating factors of production is based on the entrepreneur s business 21

22 conception. This consists in subjective and highly idiosyncratic imaginings in the mind of potential entrepreneurs of what and how business should be undertaken. It is the basis for the entrepreneur's interpretation of incoming information relative to its relevance and meaning for the imagined business venture (Witt 1998, 165; 1999, 104; 2007, ). In this manner, entrepreneurial imagination does not only determine the uniqueness of the entrepreneurial function, it is the factor defining the cognitive framework giving its identity to the entrepreneurial firm. The entrepreneurial firm cannot be dissociated from the profit opportunity as imagined by the entrepreneur. Consequently, the entrepreneur is not just a function within the firm, it is the function giving it its very identity. This explains why there is no spot market for entrepreneurship. The only way for an entrepreneur to materialize her vision is to create a firm from her imagination by directing and coordinating factors of production through her cognitive framework (Witt 1999, 105). 6.3 Exploitation of profit opportunities: institutions Entrepreneurial success or failure in seizing opportunities is not a matter of noticing and collecting information, but of one s judgment about the future that is, about how inputs might contribute to the future market success of output (Casson 2003, 20ff.; Foss & Klein 2012; Knight 2006, 281ff.). It is about enacting one s judgment under the burden of uncertainty about that future. This uncertainty does not arise from information per se, but from how it is generated. The entrepreneurial problem is not ignorance of available information; it is ignorance of other people s judgment abilities (Knight 2006, 285). Entrepreneurial proficiency is thus a question of one s success in judging the future and dealing with its pure uncertainty relative to one s competitors (Knight 2006, 281). 22

23 Although it can be hard to distinguish successful entrepreneurship from pure luck (Knight 2006, 311), entrepreneurs can reduce uncertainty to better secure success. In fact, entrepreneurs are not faced with a single pure uncertainty factor, but with different purely uncertain events. Uncertainty can be reduced by limiting the number of the uncertain events confronting the entrepreneur. Essentially this amounts to consolidation of events into categories or classes of events, which allow for the derivation of some objective probability distribution (Knight 2006, 239). The problem is that profit opportunities cannot be consolidated into categories because of their futurity. Nevertheless, the events constituting them can. Instead of consolidating profit opportunities, entrepreneurs reduce uncertainty by consolidating specific classes of action, i.e., institutionalized behaviour. To avoid fastidious collection of information each time it is made available, agents develop procedures to interpret situations as these emerge. These procedures are adopted because of their tried-and-true character with dealing with past similar situations. Eventually, the procedure-action set becomes unconsciously routinized as rules of individual action, that is, as customs or habits (Hayek 1969, 56). Although cognitive procedures cannot be objectively observed by others, habits as actions can, enabling reproduction by others. Moreover, whenever those habits are attributed a symbolically-defined meaning by more than one agent, they gain a status-function, a social meaning associated to a given context or situation: an X habit counts as a Y status-function in a context C (Searle 2005, 6 7). Institutions emerge whenever habits become a system of constitutive rules of the form X counts as Y in C (Searle 2005, 10). As constitutive rules, institutions involve codification of interpretation and behaviour, widespread social reproduction, and their embeddedness in procedures and habits (Hodgson 2002, 174). Institutions influence habits, which in their turn reinforce and are reinforced by institutions. In this manner, institutions stabilize socio-economic systems by being perpetuated by agents while 23

24 also buffering and constraining their actions (Elsner 2010, 452; Hodgson 2000, 116). Since institutions involve rules, these can evolve into enforcement systems that reinforce certain behaviours over others in either a formal (i.e., authority-enforced) or informal (i.e., peerenforced) way. The identification of institutionalized behaviour allows entrepreneurs to categorize the actions of others and calculate probability distributions for a considerable number of the events constituting emerging profit opportunities. This reduces the uncertainty facing an entrepreneur, but does not eliminate it. This occurs because socio-economic change is not always institutionalized. First, not all actions are institutionalized behaviour, thus retaining their unpredictable character. Second, not all institutionalized behaviour involves enforcement mechanisms, making their categorization more haphazard. Third, socio-economic change can result in the disintegration of old institutions, and the emergence of new ones. Although the entrepreneur cannot do much in any objective way concerning the first two cases, something can be done relative to the last one. Entrepreneurs can direct production to accelerate either institutional disintegration or emergence with their products. In this manner, the entrepreneur purposefully participates in the process of institutional cohesion of new habits (Giménez Roche 2016, 710). The point is to make profits by identifying these institutional changes not as unpredictable behaviour, but as emerging habits before other entrepreneurs do. If institutional change is not coming, entrepreneurs can also direct production towards disruption of the institutional status quo, thus triggering new habits that will result in either institutional disintegration or emergence (Giménez Roche 2016, ). This heterodox view of entrepreneurship has the advantage of integrating all neoclassical theories of entrepreneurship, without assuming perfect rationality and structurally complete information. Instead, it accounts for uncertainty while assuming imperfect creative rationality, 24

25 and incomplete though open-ended information. The definition of a production function, fundamental in X-efficiency theory, becomes a matter of defining production relative to institutional changes in the market process. Risk-bearing, as in Khilstrom and Laffont s risk entrepreneurship theory, becomes a question of entrepreneurial proficiency in identifying market habits and institutional change in such a way that uncertainty can be reduced. Entrepreneurial alertness is not limited to objective arbitrage opportunities, but is now alertness to institutional changes that can be categorized to reduce uncertainty and better secure profits. The idea of entrepreneurial cohesion and disruption of market habits and institutions gives an endogenous explanation to Schumpeterian market innovation, and Schultzian human capital formation. The specialisation between start-ups and consolidated large businesses described by Baumol becomes just one institutional arrangement among others. Market-making is not just about looking for existent information about transaction costs, but about being alert to institutional changes and preparing for transaction costs better than others. Entrepreneurial coordination of the firm s activities and leadership are defined by entrepreneurial imagination of institutional dynamics. 7. Conclusion This article shows that there is a neoclassical theory of entrepreneurship, which integrates the entrepreneurial role despite the assumptions of perfect rationality and structurally complete information. For this, neoclassical theories of entrepreneurship assume limited distribution of perfect rationality in the population, and a costly access to structurally complete information. Nevertheless, this watering down of neoclassical assumptions does not provide any endogenous explanations of the emergence, awareness, and exploitation of profit opportunities. Entrepreneurship is simply assumed to be limited to those few individuals displaying perfect 25