• No results found

Entrepreneurship, overconfidence and overoptimism : an overview of the relevant literature

N/A
N/A
Protected

Academic year: 2021

Share "Entrepreneurship, overconfidence and overoptimism : an overview of the relevant literature"

Copied!
50
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Koen Rijper 10188924

Bachelor Thesis Science –Business Studies Word count: 11328

University of Amsterdam

Faculty Economics and Business 01-06-2014

Supervisor: Professor P.D. Koellinger

Entrepreneurship, overconfidence and overoptimism: an overview of the relevant literature.

Abstract:

This article is going to react to some contrasting results presented by studies about overconfident and overoptimistic entrepreneurs. Besides this the literature about overconfidence and overoptimism do not use their definitions in a consistent manner. This study is therefore is going to critically review all the literature concerning overconfident and overoptimistic

entrepreneurs, and will create a good and clear overview. By doing this, this study will try and solve the literary differences between the articles, and will discuss the issue with overconfidence and overoptimism. On the basis of the overview, a light will be shed on the differing results of the overconfidence, overoptimism, and entrepreneurial literature, and there will be found answers what the definitions of overconfidence and overoptism truly are.

(2)

1. Introduction:

‘’There was no doubt about it: I had discovered The Next Big Thing. Like Edison and the lightbulb, like Gates and the pc operating system, I would launch a revolution that would

transform society while bringing me wealth and fame.’’ (Strauss, 1999)

In his quote Robert L. Strauss was talking about a new product, called a condom key chain. As the quote would suggest, he thought to have the best idea ever. After his ‘’amazing’’ discovery he entered the market of souvenirs with an MBA in Stanford and an initial investment of 10.000 dollars. Was this a dumb decision? Well it turned out it was, because his business failed after only two years and left him with 10.000 dollars in debt. (Strauss, 1999).

Robert L. Strauss is not the only entrepreneur who failed. Actually, 75 % of all entrepreneurs starting a new business could have better stayed at their wage job (Hamilton 2000). Entrepreneurs take not only the risk of their financial investments, but also risk career opportunities, family relationships, personal wealth and physical well-being (Brid, 1989). Despite these risks, high failure rates, and low returns (Baldwin, 1995), still too many entrepreneurs are entering markets (Koellinger, 2007).

The following question then arises: Why do these entrepreneurs still enter these markets when the payoffs are low? Researchers as for example Koellinger, Busenitz, Lowe, Ziedonis, Friedmann, Haylard, Shepherd, Griffin, Camerer and Lovalloidentified different kinds of biases in entrepreneurial behavior that could possibly answer this question. Specifically the above mentioned researchers all address the same topic: overconfidence and/or overoptimism as a bias for entrepreneurial activity.

Where Camerer and Lovallo found a causal relationship experimentally between overconfidence and excess entry, and Koellinger (2007) found a relationship in a field study between these two variables. Lowe and Ziedonis didn’t find a relationship between

overoptimism and the entry of entrepreneurs. Why would this be? This is just one example of an issue within the literature about overconfident and overoptimistic entrepreneurs. A possible answer could be that the research strategy differs between these articles. This requires a critical view towards the literature to identify those differences. Another explanation could be the

difference between overoptimism and overconfidence. Moore and Healy (2008) discussed the trouble with overconfidence, and states that the literature about overconfidence and

overoptimism do not use their definitions in a consistent manner. Moore and Healy (2008) shed

(3)

some light on the situation, which could further implicate the research that has been conducted in the past about overconfidence and overoptimism.

This study is therefore going to critically review all the literature concerning overconfident and overoptimistic entrepreneurs, and will create a good and clear overview. By doing this, this study will try and solve the literary differences between the articles, and will discuss the issue with overconfidence and overoptimism.

To do this, this study will first further illustrate the variables entrepreneurship,

overconfidence, and overoptimism. Secondly the method will be explained including the finding and selection criteria of the literature. Thirdly the chosen literature will be critically reviewed. This will be done with the use of table two presented in the appendix, which contains a

summary of every article used in the literary review. After this a conclusion will be given where an overview will be presented. In this the main findings together with the issues that still need to be overcome will be stated. Finally in the discussion the implications of this study will be

discussed and future research will be proposed.

2. The variables

2.1. Definition of Entrepreneurship.

Sharma and Chrisman (1999) define entrepreneurship as encompassing acts of organizational creation, renewal, or innovation that occur within or outside an existing organization. They further define entrepreneurs as ‘’individuals or groups of individuals, acting independently or as part of a corporate system, who create new organizations, or instigate renewal or innovation within an existing organization’’. Gartner (1990) defines entrepreneurs as individuals with unique personality characteristics and abilities (e.g. risk taking, locusof control, autonomy,

perseverance, commitment, vision, and creativity). Gartner (1988)further argues that

entrepreneurship researchers have generated incomplete definitions. The definition: ‘’a person who establishes a new organization’’ is an example of this problem(Shane & Venkataraman, 2000). Shane and Venkataraman (2000) further argue that this definition does not include consideration of the variation in the quality of opportunities that different people identify, it leads researchers to neglect to measure opportunities. Therefore they define the field of

entrepreneurship as ‘’The scholarly examination of how, by whom, and with what effects opportunities to create future goods and services are discovered, evaluated, and exploited’’.

(4)

To define entrepreneurship there has to be recognized that there are two ways to do this. Either someone can use a functional approach, or an Indicative approach (Casson, 1982). The functional approach identifies a function, and deems anyone who performs this function as being, in this case, an entrepreneur. For being defined as an entrepreneur a function could for example be starting a new business (Cooper, et al., 1988). The functional approach does however need to rely implicitly on risk-taking and residual-claimant aspects of entrepreneurship, which also facilitates the analysis of incentives, investments, resource allocation decisions and occupational choices (Parker, 2009).

The indicative approach looks more at the way an entrepreneur can be recognized. This way an indicative approach uses legal status, contractual relations, social status, etc., to

describe an entrepreneur (Casson, 1982). This can be seen in the definitions given by Gartner (1990; 1988) stated above. This indicative approach argues that there is nothing entrepreneurial about being a business owner (Parker, 2009), but that an entrepreneur can be recognized by, for example, their characteristics (Koh, 1996).

Stevenson (2006) believes that neither of these approaches are sound. Both of these approaches focuses only upon some aspects of some entrepreneurs. For example, Thomas Watson of IBM is not a founder of a firm, but is considered an entrepreneur according to stevenson (2006). Besides this Stevenson (2006) argues that a single psychological profile for an entrepreneur is also bound to fail, because for every traditional definition of an

entrepreneurial type, there are counterexamples.

Casson (1982) states that the different approaches should be used by different study fields: ‘’By and large, economic theorists have adopted a functional approach and economic historians an indicative one. This is as it should be. Economic theory offers a set of concepts and techniques for analysing the allocation of scarce resources. Unless entrepreneurship ultimately derives from a scarce resource it is of little economic interest, even though it may be of social importance’’. This research agrees with the latter quote from Casson and a functional approach of entrepreneurship will be accepted in the following literature review.

2.2. Definition of Overconfidence

As stated in the introduction, overconfidence is a variable that is not consistently used within the literature (Moore and Healy, 2008). Moore and Healy (2008) further state that overconfidence is defined in past research in three distinct ways. First of all, literature has defined overconfidence as an overestimation of one's own ability, performance, level of control, or chance of success.

(5)

Cooper, et al. (1988) for example asked their participants ‘’What are the odds of your business succeeding?’’. By comparing this to the actual odds a business will succeed, one could see if someone overestimated their chance of success. Moore and Healy (2008) call this construct overestimation.

Secondly Moore & Healy (2008) argue that overconfidence occurs when people believe themselves to be better than others, also called the better-than-average effect. Cooper, et al. (1988) for example also use this definition by asking the participants ‘’What are the odds of any business like yours succeeding?’’. To get to the better than average effect they measure the participants perceived odds that their business will succeed, and the perceived odds of other businesses succeeding. The better than average effect occurs when people perceive their own odds better than that of others and thereby overplacing themselves. Moore and Healy (2008) call this construct overplacement.

Finally, overconfidence is defined as excessive certainty regarding the accuracy of one’s beliefs. A few examples of this are Busenitz (1999), and Fischhoff et al. (1977), which asked five questions to participants based on series of death rates. The question was ‘’Which cause of death is more frequent in the united states? A. Cancer, B. Heart diseases’’. They were then asked what the certainty of their believe was. Hereby measuring the certainty of the precision of ones answer. This will be measured against the actual performance of their answer thereby seeing if they were unrealistically certain about the precision of their believes. Moore and Healy (2008) call this construct overprecision.

Past researchers argue that the overestimation, overplacement, and overprecision all result from the same underlying psychological cause (e.g., Alba & Hutchinson, 2000; Barber & Odean, 2001) . Moore & Healy (2008) argue that these three different types of overconfidence are conceptually and empirically distinctive. They differ because of the imperfect information that one has about one’s own performance, and even lesser amount of information about the performance of others. Because of this people’s estimates of themselves are regressive, and their estimates of others are even more regressive. This means that when performance is high, people tend to underestimate themselves, but tend to underestimate others even more, causing overplacement (Moore and Healy, 2008). They proved this negative relation between

overestimation and overplacement empirically and conceptually. This result implies that it does matter which construct is used in the literature, and that different usages of these constructs can cause different final results.

Parker (2009) implies that only one of the constructs identified by Moore and Healy (2008) reflect overconfidence. In his book he states the following: ‘’The second moment of the

(6)

subjective probability distribution is excessively small, perhaps reflecting the individual’s misplaced self-confidence in the accuracy of their beliefs’’. With this statement Parker means that overconfidence reflects the underestimated variance of the perceived outcomes by people. He further implies that overconfidence would have only one of the constructs used by past research, named overprecision. There are many others who define overconfidence just as Parker (2009) implies. For example Haran et. al (2010) identified overconfidence as the excessive certainty one knows the truth, Tversky and Kahneman (1974) state that

overconfidence occurs because people do not revise their estimates, because they do not realize that their estimates may be incorrect, and Klayman, et al. (1999) state that

overconfidence occurs when an individual’s certainty that his or her predictions are correct exceeds the accuracy of those predictions.

As can be seen, different authors use different kind of constructs for their overconfidence measures. Because Moore and Healy (2008) state that these constructs are not part of the same psychological cause, the inconsistent use of the constructs forms a problem within the literature. This article will propose a solution for this problem.

2.3. Definition of Overoptimism

First of all, overoptimism, or delusional optimism is identified within the literature. It has long been recognized that people have a tendency to be innately over-optimistic (Parker, 2009). Parker (2009) argues that an over-optimistic person overestimates the probability of high-x outcomes and underestimates the probability of low-x outcomes. This definition would suggest that overoptimism refers to the overestimation of one's chance of success (Parker 2009; Van den Steen, 2004; Cooper, et al., 1988), overestimation of one's ability to, for example, manage a firm (Hayward & Hambrick, 1997), and the overestimation of performance measured as, for example, projected sales (Cassar, 2010). Besides these definitions of overoptimism, Hmieleski and Baron (2009) us the term delusional optimism. They define this as overemphasizing projects potential benefits and underestimate their likely costs, which complies with the same definition as Parker (2009). As can be seen, these definitions overlap with the overestimation construct identified by Moore & Healy (2008).

Secondly, this study identifies comparative optimism. Comparative optimism is defined as the tendency of people to report that they are less likely than others to experience negative events, and more likely than others to experience positive events (Ucbasaran, et al., 2010). Comparative optimism can be measured by asking the following questions: What are the odds

(7)

of this business achieving your expectations for in the future? What are the chances of any other business like yours succeeding (Cooper, et al., 1988). As can be seen, this definition overlaps with the overplacement construct identified by Moore and Healy (2008).

As can be seen the definitions and proposed underlying constructs overlap between overconfidence and the overoptimism literature. Because the study by Moore and Healy (2008) argues that the three constructs of overconfidence are not part of the same underlying

psychological cause, this study argues that the definitions need to differ for each construct. Through critically reviewing all the articles in the literature about overconfident and

overoptimistic entrepreneurs, this study is going to propose a solution for this issue. In this proposition it will be identified which of the constructs identified by Moore and Healy (2008) belong to which definition of overconfidence and overoptimism.

3. Method

To create an overview of all the relevant literature and eventually solve literary differences, all the relevant literature needs to be found and analyzed in a strict and repetitive matter, which is necessary to ensure replicability of the research. First of all the data needs to be collected. At first this will be done with the use of google scholar. In google scholar the literature will be searched using the following keywords: Overconfidence, Overoptimism, Hubris, Entrepreneur, Entrepreneurial activity, Startups, Founding a new business, self-employment, new entry, and business owner. This literature search will result in an outcome of a certain amount of articles. These articles will be checked for relevance, through looking at the applicability to the subject overconfidence, overoptimism and entrepreneurship, and within these relevant articles there will be searched for more relevant literature through their reference lists. The found literature will be further selected through the use of ISIWebofknowledge. This website identifies which journals have the most impact, are peer reviewed, and which journals publish good, well written articles. Within this databank the entrepreneurial journals will be shown in a list divided into different pages. This research will only include the articles published within the journals in the first five pages of the list presented by the ISIWebofknowledge, to rule out any insignificant and not often cited articles. After this all the relevant articles accepted by this method will be included in the literature review. The declined literature will be included in the appendix, either because of the journal or relevance specification.

After the data collection and selection the chosen literature needs to be reviewed. This will be done in eight parts. First of all the articles will be summarized. Secondly the way the

(8)

variables entrepreneurship, overconfidence, and overoptimism, are identified. Thirdly the question will arise: is the argument sound? With this is meant the way the argument is logically presented, and literary substantiated. The fourth question will be: how do the authors test their predictions. This question will be answered by the method presented in the paper. About this method the question: does the methodology leave room for alternative explanations will be asked. After this the major weaknesses of the paper will be discussed. Finally the findings will be concluded and the impact of the article will be determined, where after the future research propositions will be discussed. This extensive research will be summarized and presented in table two, which can be found within the appendix.

4. Literature review

Within the literature about overconfident and over optimistic entrepreneurs, there were three types of issues which were identified within the selected literature. First there is the issue

whether overconfidence and overoptimism are factors influencing entrepreneurs, and if so, what factors affects this over optimism and overconfidence. Secondly the identified literature

discussed the issue about what overconfidence and overoptimism mean for the entrepreneurs, their businesses and their markets. The third and final issue that has been identified is whether overconfidence and overoptimism increase social benefit and if there is a solution for the

overconfidence and over optimism bias. With the use of the literature summarized in table 2, the three issues will be examined. The literature will be structured through these three issues and will be further structured by the publish dates of the articles. At the end of the literature review, this study will present an overview of the findings to draw conclusions from the literature.

4.1 .Does overconfidence and overoptimism affect entrepreneurs, and if so what affects this bias?

This study identified literature that provides answers to the question: ‘’Does overconfidence and overoptimism affect entrepreneurs, and if so what factors affect this bias?’’. First of all Cooper, Woo, and Dunkelberg (1988) examined how entrepreneurs perceive their chances of success shortly after they have become business owners. This study has been conducted with a large sample of 2994 participants from the US. The participants were asked: ‘’What are the odds of your business succeeding?’’, and ‘’What are the odds of any business like yours succeeding?’’. Overall entrepreneurs have 34-50% chance that their business will fail. Of the participants 95 %

(9)

perceived their odds of success to be above 50 %. This provides evidence for new

entrepreneurs containing a remarkable degree of optimism. Besides this, participants also perceived their prospects to be better than those of others in similar businesses.

Because of this article we now know that entrepreneurs are remarkably optimistic, and place their chance of success above that of others. Besides this, the article proves that the constructs overestimation and overplacement are found within new entrepreneurs. The way this research has been conducted does however raise two issues. The first issue is that this article does not compare the results of the entrepreneurs to non-entrepreneurs. This means that it is not clear if every person is optimistic, or only entrepreneurs are. This provides a gap that needs to be filled by further research. The second issue is about the fact that Cooper, Woo, and Dunkelberg (1988) use entrepreneurs that have already made a commitment to becoming a business owner. Abelson and Levi (1985) argue that when someone makes a decision, the attractiveness of this decision will be exaggerated by the decision maker, also called the theory of cognitive dissonance. Because the new entrepreneurs already made the decision to found a firm, they could very well exaggerate the attractiveness of becoming an entrepreneurs, which puts the results in jeopardy.

Busenitz and Barney (1997), and Busenitz (1999), answer to the research of Cooper, et al. (1988) by researching whether there is a difference between managers and entrepreneurs in their decision making processes, and specifically the difference of the two biases,

representativeness and overconfidence. This study uses 176 manufacturing firms to test their predictions, and entrepreneurs were identified as founders of these firms. Besides being a founder, the participants also had to be involved in a startup in the past two years, or getting involved in a startup within at least five years from when the survey was taken. Participants were defined as managers when they managed in at least two functional areas within a company. Furthermore the participants were seen as overconfident when they were overly optimistic about their initial assessment of a situation. They measured this by asking participants to answer five death rate questions, and consequently asking what their confidence in their answer was, 50% or 100%. The results show that entrepreneurs were more likely to be overly optimistic about their initial assessment situation, providing evidence for the tendency of entrepreneurs to be more susceptible to overconfidence than managers. Busenitz (1999) conducted the same study in the same way, but added risk propensity in his analysis. This eventually showed no extra results.

There are three issues with these two articles. First of all Busenitz and Barney (1997), and Busenitz (1999), use a different kind of measure for overconfidence then Cooper, et al.

(10)

(1988) used for their optimism. This means that they cannot address the issue presented in the study of Cooper et al. (1988), whether new entrepreneurs differ in their optimism to

non-entrepreneurs. Secondly, this research again uses entrepreneurs who are founders of the firms. This means that again new entrepreneurs are used, and still the cognitive dissonance bias (Abelson and Levi, 1985) exists within this research. Finally there is a problem with the

generalizability. They only use manufacturing firms within the US. This leaves the possibility that country and firm characteristics will be partly responsible for the results in this research.

Camerer and Lovallo (1999) state that when subjects post entry payoffs are based on their own abilities, individuals tend to overestimate their chances of relative success and enter more frequently. They tested this using 118 undergraduate and MBA students. They conducted an experiment with these students, which were told they could earn money through their

performance on sports or trivia. This research provides experimental evidence that people who are overconfident about their skill, and place their chances of success above others, enter markets more frequently. Besides this finding, they identified a new phenomenon called

reference group neglect. They state that when someone competes on the basis of skill, they will be insufficiently sensitive to the quality of competition. This is a cause of the inside view, which argues that people see their situation as unique and therefore neglect thinking about others. (Kahneman, 1993)

This article proves that there is excessive entry when entry payoffs are based on someone owns ability. There is however one issue with this research. The study conducted shows interesting results, but only through an experiment. To check if these phenomenon are also applicable to the real world, there first needs to be done a field study that tests these predictions.

Arabsheibani, et al. (2000) develop a model in which they propose that people are optimistic, and that people who are self-employed are even more optimistic. They provide evidence from the british household panel study which asked 22965 employees and self-employed two questions every year. They asked: ‘’Would you say that you, yourself are better off, worse off or about the same financially than you were a year ago?’’, and ‘’Looking ahead, how do you think you yourself will be financially a year from now, better off, worse off, or the same?’’. This implies that over-optimism is defined as an optimistic forecasts of someone's finances in a year, compared to the actual perceived financial state about that year. This article proves that self-employed people are more over-optimistic about their future financial state, than people who work for a company. They argue that this is due to the fact that the self-employed

(11)

perceive their performance under their own control, but do not provide empirical evidence for the latter statement.

This article has two key issues. The first problem with this research is that they do not take into account how many years someone is active as a self-employed person. As experience moderates the overestimation of one's performance, this article should also control for this variable (e.g. Hayward, et al., 2006; Ucbasaran, et al., 2010). Secondly the way over-optimism was measured raised a problem. The real performance about the earlier made predictions are subjective. Because they were subjective, the answer to: ‘’Looking ahead how do you think you yourself will be financially a year from now, better off, worse off, or the same?’’ could be

susceptible to the cognitive dissonance theory. When people predict their financial state will increase in the previous year, and it ended up being even worse, the participant can be very disappointed. The cognitive dissonance theory states that people need to find some way of resolving their inconsistencies (Cooper, 2007), and therefore the participant can for example state that their financial state is not that bad. Therefore cognitive dissonance could, if present, affect the amount of overoptimism.

Simon and Houghton (2003) found that overconfidence was positively related to the degree to which product introductions were pioneering. Furthermore the managers introducing these pioneering products were more apt to express extreme certainty about achieving success. They provided evidence for these statements using a sample of 55 managers from computer companies. They defined overconfidence as an excessive certainty about one’s prediction, which complies with the underlying construct overprecision. They measured this overprecision the same way Busenitz and Barney (1997) measured overconfidence.

There were two issues with this article. First they only used risk propensity as a control variable. This control variable alone is not sufficient, because there are other factors that could influence overconfidence. Examples of these factors are experience (Hayward, et al., 2006; Ucbasaran, et al., 2010), education (Busenitz, 1999; Busenitz and Barney, 1997), and age (Forbes, 2005). Secondly the sample that is used limits the generalizability. Their sample

consists of only managers within the computer industry placed in American state Georgia, which jeopardizes the generalizability of the findings. Finally, one could argue that entrepreneurship is not being researched in this article, because they interview managers. We bet the differ and state that these managers can be seen as entrepreneurs. The computer industry is a particular distinct industry, because of their high risk, short product life cycles, and rapidly changing customer needs (Simon and Houghton, 2003).The computer industry therefore requires

managers to cope with these rapidly changing customer needs, high risk, and short product life

(12)

cycles (Simon and Houghton, 2003). Because managers need to deal with these

characteristics, they are forced to comply to important constructs of entrepreneurship, called risk, and opportunity taking (Venkataraman, 1997), and could be argued being established entrepreneurs.

Forbes (2005) found that entrepreneurs are overconfident because they are affected by individual age, firm decision comprehensiveness, external equity funding, and whether the venture managers are founders or not. They provided evidence by using 108 internet firms based in New York. They measured overconfidence of this sample using the same measure as Simon and Houghton (1999), and Busenitz and Barney (1997) did, and entrepreneurs were new venture managers who are founders. With this this study they provide evidence that

overconfidence is a function of both individual and contextual factors

The first issue with this paper is the sample consisting only internet firms in New York. Because of the limited scope of the sample used, generalizability is put in jeopardy. Secondly they use new entrepreneurs just as Cooper et al. (1988) did. They do not use the same measures as Cooper, but it would be interesting if there is a difference between new,

established entrepreneurs, and people who want to become entrepreneurs. Besides this, the article does not look at the temporal stability of overconfidence. Forbes (2005) states that an understanding of how quickly entrepreneurs susceptibility to overconfidence changes over time is necessary.

Fraser and Greene (2005) argue that entrepreneurs are more optimistic than employees, and optimism and uncertainty are diminished by experience. They provide evidence using the same cross sectional samples of BSAS data as Arabsheibani, et al. (2000), which is an annual survey conducted over adults owning a house in Britain. Again the participants were separated between wage-workers and the self-employed, and were asked whether they felt that their business prospects for the following year were better, the same, or worse. This means that optimism is defined as the prediction of your financial state in the future. Besides this Fraser and Greene (2005) continue on the research of Arabsheibani et al. (2000) and provides evidence that experience diminishes the optimism found within entrepreneurs. The test this by measuring the found optimism against the years of business experience the participant had.

There is however one issue that was detected. Arabsheibani, et al. (2000) used the questions ‘’Would you say that you yourself are better off, worse off or about the same

financially than you were a year ago?’’, and ‘’Looking ahead how do you think you yourself will be financially a year from now, better off, worse off, or the same?’’. The research from Fraser and Greene (2005) only uses the latter question to define optimism. This means that within this

(13)

research optimism cannot be defined as overoptimism. However, combining both studies one could argue that the diminishing experience on optimism also applies to overoptimism, but no direct empirical evidence stands for this argument.

Zhao et al. (2005) States that people who have a higher self-efficacy will be more inclined to be an entrepreneur than a wage worker. Besides this, they state that formal learning in entrepreneurship courses, entrepreneurial experience, and risk propensity positively affect entrepreneurial self-efficacy. Self-efficacy is defined by this paper as an individual’s confidence in his ability to perform entrepreneurial roles. They provide evidence for these result by using 265 MBA students of five US universities. They surveyed them when they started the MBA, and when they graduated. They were asked, how much they have learned throughout their MBA study, whether they had experience, and through a series of questions what their risk propensity was. Next to this their self-efficacy was determined by asking how confidence they were in identifying new opportunities, identifying new products, and thinking creatively.

Forbes (2005) researched if self-efficacy will have an effect on overconfidence. Forbes (2005) used the overprecision construct and argued that there is no effect. This suggests that one's perceived ability to be successful, or self-efficacy, does not comply with overprecision. Self-efficacy might however be related to overestimation of one's ability. The factors proposed would then affect overestimation, but it could also be that they have a higher self-efficacy, because they have a higher chance of succeeding. This is a gap in the literature that needs to be researched.

Hayward, et al. (2006) propose from theory and evidence on behavioral decisions that the greater environmental complexity, and the greater environmental dynamism, the greater overconfidence. Besides this they argue that experience in founding successful ventures cause an entrepreneur to be even more overconfident when the focal company differs from the

previous one. They define overconfidence as an overestimation of the likelihood that their venture will succeed, and that they can ensure such success. These are propositions that are founded by arguments and logical reasoning, but no empirical evidence has been provided. Therefore to generate conclusions from their propositions, empirical evidence is necessary. Puri and Robinson (2006) argue that entrepreneurs are more overly optimistic in their life expectancy than non-entrepreneurs. Entrepreneurs are also more optimistic about personal and economic conditions, perceive more private benefits from working than non-entrepreneurs, and are more willing to take substantial risk. These results are tested through the use of the Survey of Consumer Finance (SCF), which consists of 2578 US participants. Entrepreneurs were identified when they were full-time self-employed, and owners of private equity. They measured

(14)

overoptimism by asking participants what their life expectancy was, which was accordingly measured against their real life expectancy based on age, average health, and smoking.

This is another article that provides evidence for the over optimism of entrepreneurs compared to non-entrepreneurs. In this study however they propose a new way of measuring over optimism within the selected entrepreneurial literature. This type of measuring of

overoptimism could be pooled in the construct: one’s overestimation of performance. One could argue that living longer is a performance indicator of one's health and way of living. The way of measuring however is not related to the business context. This provides new evidence that entrepreneurs also outside of their business context are more overly optimistic about their performance than non-entrepreneurs.

Koellinger, et al. (2007) showed that people who enter markets have biased perceptions about their own skill to start a business. This would suggest that entrepreneurs are

overconfident. They test their predictions using the global entrepreneurship monitor (GEM). This dataset consisted of 74,000 observation within 29 countries. Within this dataset nascent, new and established entrepreneurs are identified. They were asked among other things if they had the skill, and the opportunities available to start a business. The results show that within countries there is no big relation between the skill perception and the education of the participants. This is odd, because education should increase one's skill, and therefore their ability perception. However the real result from which they reason that people enter markets more frequently when overconfident is that established entrepreneurs are less confident about their skill than nascent entrepreneurs. Through experience ones skill should increase and therefore also their perception, instead through experience the established entrepreneurs become less confident. Koellinger, et al. (2007) argue from this that entrepreneurs when entering a market, are overconfident in their skill to establish a new firm.

This research replied to a number of issues proposed by previous studies. First, by using such a big sample across that many countries, they provide the means to generalize their findings. Cooper, et al. (1988) and Busenitz and Barney (1997) for example used either a small sample, or a sample only applied to the US. This means that the research of Cooper, et al. can be generalized by the article of Koellinger, et al. (2007). Busenitz and Barney (1997) however use a different construct for their overconfidence. This means that this article does not provide any generalizability for the research of Busenitz and Barney (1997). A second issue that Koellinger, et al. (2007) solve is the issue of Camerer and Lovallo. Camerer and Lovallo also provided evidence that people who overestimate their ability to start a venture will enter markets more frequently, but did this in an experimental study. Koellinger et al. (2007) solved this

(15)

problem by conducting a field study. There is however one issue with this paper. They do not provide any evidence of causality within their results, It could therefore very well be that entrepreneurs are overconfident because they enter markets. This literature review is going to argue that by combining the research of Camerer and Lovallo (1999) and the study of Koellinger et al (2007) causality will probably be the case, because the experiment does provide evidence for this causality.

Cassar (2010) provides evidence that the expectations of the chance of success of becoming an operating venture are overly-optimistic for people who are in the process of starting a business. They also proved that the overoptimistic sales expectancies of

entrepreneurs are positively related to financial projections. Finally in their results can be seen that by making sales, overoptimism will be dampened, which would argue that experience within a business will moderate the effect of overoptimism on an entrepreneurs expectations. He tested this using the Panel Study of Entrepreneurial Dynamics (PSED), of which they used 386 entrepreneurs.

First of all, it is good that they use a longitudinal study overcoming for example the bias of Arabsheibani and his subjective performance measure over the year where the predictions were made about. Besides this, they also solved a problem of the study conducted by Cooper, et al. (1988). The critique point on Cooper et al. (1988) was that they had participants who were already committed to becoming an entrepreneur, thereby possibly exaggerating the

attractiveness of this decision and biasing the results. This research studied the same construct identified by Moore and Healy (2008), namely the overestimation of one's chance of success. They however used people in the survey that were planning on becoming entrepreneur. This means that entrepreneurs are overly optimistic in their prediction about one's chance of success with entrepreneurs who own a business as well as with entrepreneurs who are planning on becoming a business owner.

Further Cassar discusses the inside view. The inside can be explained as people who makes forecasts by focusing on the case at hand, thereby extrapolating current trends into the future (Lovallo and Kahneman, 2003). Cassar (2010) argues that because overoptimistic sales expectancies of entrepreneurs are positively related to financial projections, the inside view exacerbates overoptimism in financial expectations. There is one issue with the statement Cassar (2010) makes. He only found one relation with financial expectations, the inside view cannot be argued to affect the entire overestimation construct.

Ucbasaran, et al. (2010) researched if it is possible for some entrepreneurs to temper

comparative optimism. They used a sample of 576 entrepreneurs from Britain to research this,

(16)

and define comparative optimism as the tendency of people to report that they are less likely than others to experience negative events, and more likely than others to experience positive events. The evidence state that repeat entrepreneurs without experiencing business failure are more likely to report comparative optimism. They further argue that portfolio entrepreneurs (an entrepreneur with multiple firms) are less likely to report comparative optimism than novice entrepreneurs when they have experienced failure. Finally, they state that sequential entrepreneurs (entrepreneurs which have only one firm at a time) will not report less

comparative optimism after a failure, they argue that this may be due to the fact that sequential entrepreneurs are more committed to the failing business, which can retard learning after a failure experience (Ucbasaran, et al., 2010).

The argument provides evidence that it depends what kind of experience an

entrepreneur has acquired. Besides this, it also matters what kind of entrepreneur experiences a failure. Previous research has found that experience reduces overestimation of predictions within entrepreneurs (Fraser and Greene, 2005; Cassar, 2010). Because Ucbasaran, et al. (2010) implicitly uses the overplacement construct, this study adds the effect of experience to the overplacement construct. There is however a questions that arises from this study. Previous experience has been measured as the years in a business (Fraser and Greene, 2005), or the number of sales made with a certain business (Cassar, 2010). These however do not take into account if these experiences also consider failures and the amount of commitment of the entrepreneurs to their previous and current businesses. This is a possible new research necessary to check whether experience on overestimation isn’t also affected by the kind of experience and the commitment of the entrepreneur to his or her businesses.

The studies discussed above will provide a clear overview about whether

overconfidence and overoptimism affect entrepreneurs, and if so what affects this bias. A table with all the findings and research that needs additional research will be proposed in the

overview presented after this paragraph. First there is another issue that is surrounded by literature about overconfident and overoptimistic entrepreneurs.

4.2. What does overconfidence and overoptimism mean for entrepreneurs, their businesses, and their markets?

In this chapter the literature will discuss the question ‘’What does overconfidence and overoptimism mean for entrepreneurs and their markets?’’. The literature in this paragraph can

(17)

be found summarized in table two within the appendix. The literature in this chapter will be further structured by the publish date of the articles.

First of all, Cooper, et al. (1988) identified from other studies some factors that influence performance. These factors need to be controlled when someone wants to measure the effect of overconfidence or overoptimism and performance. Entrepreneurs who, were involved in a founding team, had education, had relevant experience, had previously owned a business, had started a business similar to those they had left, came from large firms, and had more initial capital, were more likely to succeed (Roberts, 1972; Cooper and Bruno, 1977; Dunkelberg and cooper, 1982; Mayer and Goldstein, 1961; Lamont, 1972).These factors need to be taken into account when measuring the effect of overconfidence and overoptimism on performance of a business.

Hayward and Hambrick (1997) studied that the hubris, which is defined as one’s

overestimation of ability to affect performance, will have a positive relation to the premiums paid for acquisitions. They tested their predictions by using 106 transactions with payments of over 100 million dollars. Hayward and Hambrick (1997) measured hubris of the CEO’s as the recent performance of the firm, the appraisal from the media for the CEO, the CEO’s self-importance, and a composite factor of these three. There are however a number of issues with their findings.

The way these four factors are measured are not that sound. The fact that they are researching CEO’s of firms that made big acquisitions in America, forms a problem for getting information. It is not that easy for example, getting the CEO of General Electric to participate in a set of surveys. This means they used indirect measures, and information from secondary data. Besides this the four determinants of hubris are too limited. We argue that it is a little bit far-fetched to state that if a company is doing well, the CEO has a good self-image, and is praised by the media, he overestimates his ability to affect performance of his firm. This is

because the overestimation of one's ability lies in combination of personal and contextual factors (Forbes, 2005). By taking this into account , we do not argue that CEO’s are

overestimating their ability and therefore pay a premium on acquisitions. The factors do affect the premium paid, but do not prove that a CEO overestimates their ability to affect the firms performance. Therefore this study is not used in the overview of this research.

Simon, Houghton and Aquino (1999) test whether overconfidence, illusion of control, and the law of small numbers influences the risk perception of a person, which in turn would affect the decision to start a venture. To test this they had 191 MBA students participating in their experiment. They did not find any empirical evidence for the hypothesis about overconfidence, and so the overconfidence, which was measured as overprecision, did not have any relation to

(18)

starting a business either. However they did find that illusion of control affects risk perception and in turn this risk perception affects the decision to start a venture.

There were two issue’s with this paper. First, there is the fact that this article is an experimental study. Besides this the insignificant results of overconfidence on risk perception and the decision to start a venture is only based on the overprecision construct of Moore and Healy (2008), which doesn't state anything about the other two constructs. There was also an interesting finding in this study. Illusion of control was measured through asking three questions, one of these was ‘’whether they could succeed at making this venture a success, even though many other managers would fail?’’. This is the same question that has been asked by Cooper et al. (1988) on which they identify that entrepreneurs tend to place themselves above others. Camerer and lovallo already provided evidence from their experiment, in which they state that people who overestimate their ability and over place themselves, tend to enter markets faster than people who do not. Combining these results one could argue that the overplacement construct affects risk, and the decision to start a firm. This however still needs some empirical evidence from a field study to check the experimental results.

Keh, Foo and Lim (2002) study the same issue as Simon, Houghton and Aquino (1999). In their study however they do not use students, they use owners of small and medium sized enterprises. The sample consisted of 77 participants of which almost all were Chinese, and between the age of 40 and 60. Besides another sample they also add a fourth determinant of risk perception, the planning fallacy.

Their findings complied with that of Simon, Houghton and Aquino (1999). It redid their study by using the same measures for overconfidence, but provide evidence from the field. This would mean that the results of Simon, Houghton and Aquino (1999) are also applicable to the real world. However It does raise some concern that their sample consists of almost only Chinese between 40-60 years. This reduces the generalizability of this research and leaves the results limited by its sample.

Lovallo and Kahneman (2003) conduct a literature review in which they state that we are subject to cognitive biases. The biases that they identify and fit within this literature study are competitor neglect and exaggerating our abilities and control. on competitor neglect they

comply with Camerer and Lovallo (1999) and state that this is due to the fact that people will use an inside view. As stated earlier, Fraser and Greene (2005) argued that financial forecasts affect this inside view, which provides empirical evidence for entrepreneurs using this inside view. Their findings however only touched the service, which is why this is an interesting further research point.

(19)

Lowe and Ziedonis (2006) researched if overoptimism influences entrepreneurial

venture performance. They found that overoptimism is not the determining factor in the decision to found a firm. They do find that entrepreneurs continue unsuccessful development efforts longer, which is consistent with entrepreneurial overoptimism. They defined overconfidence as the misplaced predictions of the entrepreneurs own chance of success, and the actual data of start-up survivals. They test their prediction by using a sample of 734 inventions from the university of California.

There are a number of issues concerning this article. First of is the generalizability that is in jeopardy. This article only says something about new technologies provided by a university, which is a distinct market (Simon and Houghton, 2003). Besides this Lowe and Ziedonis (2006) purely look at performance differences between established and entrepreneurial firms in

exploiting new technologies. In their study they do not measure any factors concerning the entrepreneurs and managers regarding overoptimism or overconfidence. We therefore argue that this research does not say anything about these two biases. Furthermore the lack of difference in performance can be explained by the fact that the entrepreneurs as well as the established firms are looking at the performance of pioneering technologies. As stated in the article by Simon and Houghton (2003), this is a distinct market in which one could argue that managers are considered to be entrepreneurs, because they need to grasp opportunities as they go, in combination with high risks (Venkataraman, 1997). Because of these critiques, this article will not be used within the overview of the overconfidence and overoptimism literature concerning entrepreneurs.

Hayward, Shepherd, and Griffin (2006) based their hubris theory on theory and evidence on behavioral decisions. Besides the earlier mentioned proposition about if and why

entrepreneurs are overconfident or overoptimistic. They also make some proposition about the effects that over optimism and overconfidence have on entrepreneurs and their markets. They propose that more overconfident founders start their ventures with smaller resource

endowments, commit greater resources to focal opportunities, and reduce liquidity of their ventures, which increases the likelihood that that their ventures will fail. These propositions need empirical evidence, but open up new possible research opportunities.

Puri and Robinson (2007) argue that optimism is related to a numerous of work and live choices. People who are optimistic work harder, expect to retire later, are more likely to remarry, invest more in individual stocks, and save more. Besides this, Puri and Robinson (2007) also find that normal amounts of optimism leads to reasonable financial behavior, and extreme optimism leads to behavior that is not generally prudent. To measure this they used life

(20)

expectancy miscalibration on the survey of consumer finances (SCF). That extreme optimists do not perform reasonable financial behavior, and nascent entrepreneurs are considered to be overoptimistic, which could also affect the performance of the business of the entrepreneur (Puri and Robinson, 2007).

Hmieleski and Baron (2009) provide evidence through a random sample of 207 new ventures from the US. They revealed that entrepreneurs level of dispositional optimism is negatively related to performance. Furthermore they provided evidence that this relation is moderated by entrepreneurial experience and environmental dynamism. In this research dispositional optimism is defined as generalized expectancies for experiencing positive

outcomes and was measured by asking the participants whether they in uncertain times expect the best or the worst. Entrepreneurial experience was measured by the number of previous ventures founded, environmental dynamism was measured by the level rate of unpredicted change, and firm performance as revenue and employee growth.

This study proves that dispositional optimism within entrepreneurs has a negative influence on a firms performance. However there were three issues. First of all the sample they used only consists of firms within the US, which jeopardizes the generalizability. Besides this these moderators explain only a small amount of the variance in new venture performance (Adjusted R-Squared between 0.00 and 0.23). The last issue with this paper is that dispositional optimism does not apply to the overoptimism or overconfidence bias. The dispositional optimism is not measured against an either subjective or objective performance indicator that could explain the ‘’over’’ in the overoptimism. This article does however gives an insight in the fact that the individual level of optimism and contextual variables do matter. This, however, still needs to be applied on overoptimism and or overconfidence.

Townsend, Busenitz and Arthurs (2010) used the same sample as Cassar (2010) and studied 316 entrepreneurs who are planning on starting a firm in a longitudinal study. Just as Koellinger, et al. (2007) they provide evidence that there is a positive relationship between ability expectancies of entrepreneurs and the decision to start a new venture. Furthermore they state that time-lags have a negative effect on entrepreneurs who want to start a firm.

An issue with this article is that the measures do not provide evidence that

entrepreneurs hold inflated expectations of their entrepreneurial abilities (Townsend et al., 2010), because they do not have a measure that could define the ability levels of the entrepreneurs. They do however find a new factor, time lag, that moderates dispositional optimism. This still needs to be measured with over optimism and overconfidence.

(21)

The above reviewed literature should provide the different views and findings of the relevant literature that discusses ‘’What does overconfidence and overoptimism mean for

entrepreneurs, their businesses, and their markets?’’. The clear overview of this literature will be presented after the next paragraph, first there is one last issue to be discussed.

4.4. What is the social benefit of overconfidence and overoptimism among entrepreneurs, and how can overconfidence and overoptimism be solved.

This is the final issue that this research identified within the literature about overconfident and overoptimistic entrepreneurs. There is some literature concerning ‘’What is the social benefit of overconfidence and overoptimism among entrepreneurs, and how overconfidence and

overoptimism can be solved’’. First of all, de Meza and Southey (1996) proposed a model made from a number of observations. They studied the implications of the overoptimistic

entrepreneurs. They propose that the banking sector will have a dampening effect on

entrepreneurs excessively entering markets. Banks should stem the rush of overly-optimistic entrepreneurs entering markets by capitalizing the profits of overoptimistic decisions through high interest rates. If the bank decides to not capitalize these earnings, entrepreneurs are drawn to excessive bank loans much as lemmings are drawn to the sea.

The idea that he proposes is interesting because there has been some critique over banks not lending enough (de Meza and Webb, 1987). De Meza and Houghton (1996) give a new insight into this thought by stating that if banks lend more, entrepreneurs will enter markets even more, causing more excess entry and more failures, which in turn could jeopardize the banking industry. This study also complies with Hayward, et al. (2006), which states that overoptimistic entrepreneurs perform unreasonable financial behavior.

Bernardo and Welch (1997) studied the role of overconfident entrepreneurs within groups. Entrepreneurs are defined as independent spirits, innovators, leaders, or change agents. Overconfidence is defined as the overestimation of the quality of their own information. Bernardo and Welch (1997) argue that overconfidence leaves entrepreneurs less likely to imitate their peers. Therefore entrepreneurial activity can provide valuable information in their social group. They further argue that overconfidence poses small costs on entrepreneurs, but that they provide large benefits in revealing their private information to their groups.

This article states that it is beneficial for entrepreneurs in a group to be overconfident, however there can also be too many entrepreneurs. Too many entrepreneurs will make too many mistakes eventually causing costs to the social group instead of benefits. The optimal

(22)

amount of entrepreneurs in a group is dependent on a number of factors, for example group size, information precision, and degree of overconfidence. One could argue that this is also the case in the real world. Entrepreneurs dare to take on risky ventures and will get us further than we are today, for example by putting a lot of money to invent smartphone. However according to the theory of Bernardo and Welch (1997) when there are too many entrepreneurs there will be taken huge risks, which eventually will turn out to be not so beneficial, caused for example by excess entry. These examples are however only proposition made by this research, not containing any empirical evidence.

Townsend, Busenitz and Arthurs (2010) propose a solution to reduce the tendency for entrepreneurs to start-up a company. They studied that time-lags in the decision to start a new venture significantly reduces the likability that this decision will eventually be made. The time lag was measured in months, thus by letting entrepreneurs wait at least a month to decide whether they should start a firm, will reduce this tendency.

This does however raise a problem. First of all this article does not state anything about the time lags affecting overconfidence of entrepreneurs. They tested this in a hypothesis, but did not find any relation. Secondly this article does not state which entrepreneurs will be less

inclined to start a business by these time lags. Weinberg (2009)states that entrepreneurs differ in their amount of overconfidence. Because this article did not find any relation between time lags and overconfidence of entrepreneurs, this could mean that by introducing time lags, the less overconfident will be demotivated to start a venture, but the entrepreneurs reporting the highest overconfidence would still be inclined to start a business. This article proves that time lags can decrease over entry of entrepreneurs, but cannot state if this is beneficial for the markets (Townsend, et al., 2010)

Lovallo and Kahneman (2003) also offer a solution to reduce the overconfidence and over optimism bias. Small evidence has been provided by Cassar (2010) that entrepreneurs are biased by the inside view. Lovallo and Kahneman (2003) go further and state that competitor neglect and exaggerating our abilities and control are affected by this inside view. They also propose a solution to reduce these biases. They state that this could be done through forcing the entrepreneurs and managers to apply an outside view. They offer 5 steps in realizing this. First a set of past projects need to be selected to reference to. Secondly they need to assess distributions of possible outcomes. Thirdly they need to predict the position of the project within this distribution. They then need to assess their prediction reliability and finally correct their estimates. This research provides a good, step wise approach of reducing the inside view bias.

(23)

However, there still needs to be empirical evidence proving that the inside view, does in fact affect overconfidence or overoptimism.

This could be a possible solution for the inside view, but it is not yet clear whether entrepreneurs use this inside view. Therefore this solution is not applicable to the

overconfidence and overoptimism literature until empirical evidence exists of overconfidence or overoptimism being influenced by the inside view. Through this literature study the three key issues were discussed and conclusions will be drawn in the following part of this study.

5. Conclusion

Having discussed the literature, this study will try to order the articles and make conclusions to answer the issues about overconfident and overoptimistic entrepreneurs. First the general overview will be presented. This includes the different results linked to the different constructs of overconfidence and overoptimism, and the results that needs further researching. Secondly the issue surrounding the definition of overconfidence and overoptimism will be discussed. Finally future research will be proposed within the discussion section of this study. This includes the found results that need additional research as well as new interesting literature fields.

5.1. The Overview

Table 1: The Main Overview of the literature

(24)

Overestimation: Overprecision: Overplacement: Findings: -New entrepreneurs are overly optimistic. (Cooper, et al. 1988).

-Nascent entrepreneurs are overly optimistic (Koelinger, et al., 2007; Cassar, 2010). -Excess entry can be explained by overoptimistic nascent entrepreneurs (Camerer and Lovallo, 1999; Koellinger, et al., 2007; Townsend, et al., 2010).

-New and established entrepreneurs are more overly-optimistic than non-entrepreneurs (Puri and Robinson, 2006)

-Overestimation of one’s ability is related to the illusion of control, and because of this affect risk propensity and entry decisions (Simon, et al., 1999; Keh, et al., 2002). -Overly optimistic people make less reasonable financial choices than reasonable optimistic people.

-The level of optimism is negatively related to performance, and moderated by experience and environmental dynamism (Hmieleski and Baron, 2006)

-New entrepreneurs are more

overconfident than managers (Busenitz and Barney, 1997).

-Established entrepreneurs are more inclined to introduce pioneering products (Simon and Houghton, 2003)

-New entrepreneurs over place themselves (Cooper, et al. 1988). -Excess entry can be explained by entrepreneurs overplacing themselves (Camerer and Lovallo, 1999; Koellinger, et al., 2007) -Experience of failure dampens the effect of overplacement on low committed, new and established entrepreneurs (Ucbasaran, et al., 2010).

Findings that need

further research:

-People who have a higher self-efficacy will be more inclined to be an entrepreneur, which is moderated by education, experience, and risk propensity (Zhao, et al., 2005). This research can be useful when there is empirical evidence for a direct link between ones self-efficacy and overestimation of one’s ability.

-Environmental complexity, and environmental dynamism affects ones overestimation of one’s success. Also more overconfident founders start their ventures with smaller resource endowments, commit greater resources to focal opportunities, and reduce liquidity of their venture, increasing the chance of failure. (Hayward, et al., 2006). This research can be useful when founded by empirical evidence.

-Overoptimism about performance is caused by the inside view (Cassar, 2010; Lovallo and Kahneman, 2003). This statement needs more empirical evidence. -Time lags affect dispositional optimism in a way that reduces market entry (Townsend, et al., 2010). This research needs evidence from a field study, and the measure of dispositional optimism needs to be adjusted to be applicable to overoptimsim.

-Banks should stem the rush of overly-optimistic entrepreneurs entering markets, else entrepreneurs will be drawn to founding a business, like lemmings are drawn to the sea (De Meza and Southey, 1996)

-Forcing an outside view on managers and entrepreneurs, their overoptimism can be significantly reduced (Lovallo and Kahneman, 2003). This needs empirical support.

-Entrepreneurs are overconfident because they are affected by individual age, firm comprehensiveness, and external equity funding (Forbes, 2005). This research needs to be redone with a sample containing a bigger scope.

-Overconfidence leaves entrepreneurs less likely to imitate their peers. Hereby providing valuable information in their social group, of which the entire group can benefit (Bernardo and Welch, 1997). This study needs empirical support.

(25)

There are some interesting findings within the overview that are going to be discussed here. First of all, it is clear that most of the studies used overestimation as construct for either overoptimism or overconfidence. This means that by using the overprecision and overplacement constructs, there is a research field open for discovery. It is for example not yet clear whether overprecision affects performance, if environmental factors affect overprecision. An interesting finding is that overprecision did not explain either risk propensity or the decision to start a firm (Simon, et al, 1999; Keh, et al., 2002), while overestimation did provide evidence for this (Simon, et al, 1999; Keh, et al., 2002; Camerer and Lovallo, 1999; Koellinger, et al., 2007; Townsend, et al., 2010). Also experience only reduces the overplacement construct in certain cases (Ucbasaran, et al., 2010), where with overestimation it looks as if any business or entrepreneurial experience lowers the overestimation construct.

5.2. The issue with overconfidence

First of all, the construct overprecision is going to be discussed. Overprecision is defined as an excessive certainty regarding the accuracy of one’s beliefs (Moore and Healy, 2008). Parker (2009) argues that overconfidence reflects the underestimated variance of the perceived outcomes by people. With this he implies that overconfidence would have only one of the constructs used by past research, named overprecision.

Within this literature study 24 % of the articles researched around the overprecision construct. The articles came to an agreement about the definition of overconfidence. As stated in the general literature about overconfidence, they agree with their conclusion of overprecision being the construct of overconfidence. Therefore overprecision will be defined, as stated by Simon and Houghton (2003), as the real overconfidence

Secondly, overestimation is going to be discussed. Overestimation was defined by Moore and Healy (2008) as a construct of overconfidence. This construct is defined as an overestimation of one's own ability, performance, level of control, or chance of success. Parker (2009) argued that an over-optimistic person overestimates the probability of high-x outcomes and underestimates the probability of low-x outcomes, which also ends up at the overestimation construct (Parker 2009; Van den Steen, 2004)

In this literature study, 60 % of the articles used this construct. Of this 60 %, two third used the overestimation to refer to overoptimism, delusional optimism, or unrealistic optimism. Of the one third using overestimation to refer to overconfidence, over half used this construct to argue further. They argued that overestimation of one’s ability ,either implicitly or explicitly, state

(26)

that the participants also place their abilities above others, causing overplacement. This could mean that the ones using overestimation to refer to overconfidence see this as a transitional phase towards overplacement. This article argues that the overestimation of one's chance of success and performance are constructs of overoptimism. There is however room for discussion about the overestimation of one's ability. This part of the overestimation construct is mostly argued as overconfidence, therefore the relation to the overestimation of one’s ability, and the overestimation of one’s performance, and chance of success should be measured to determine whether they belong to the same construct. For the time being however, we argue that the definition of Parker (2009) also implies the overestimation of one’s ability. Therefore also overestimation of ability is seen as overoptimism.

Finally, overplacement is going to be discussed. Moore and Healy (2008) argue that overconfidence occurs when people believe themselves to be better than others, such as when a majority of people rate themselves better than the median. In contrast Ucbasaran, et al. (2010) state that comparative optimism is defined as the tendency of people to report that they are less likely than others to experience negative events, and more likely than others to

experience positive events.

These definitions do not overlap entirely. The definition by Moore and Healy talks more about being better, Ucbasaran, et al. (2010) argues more about the likelihood of experiencing events. However, one could argue that these two are related. Hence If someone overplaces their ability, they could also overplace their chance of success (Cooper, et al., 1988). Besides this these two definitions did not have any contrasting literature, which is why this article is still going to argue these two definition are part of one construct. Because Moore and Healy (2008) do however identify a difference between overplacement and the other two constructs,

overplacement should be seen apart from overoptimism and overconfidence. Overplacement however needs a definition to avoid confusion. This research proposes that overplacement should be defined as the definition provided by Moore, and Healy (2008).

These results suggest that different constructs need to be seen apart from each other, as they provide different results, as for example overprecision and overestimation do. Therefore the three constructs identified by Moore and Healy (2008) should be identified differently to avoid confusion. This study identifies that overprecision can be called overconfidence, and overestimation can be called overoptimism. There is however room for future research about the overplacement construct. Within table 1, no contrasting findings are presented between overestimation and overplacement, which could imply them being related. Moore and Healy (2008) state that they are negatively related through task difficulty. Further research is

(27)

necessary to either present a good definition of this construct, or provide evidence that they these two constructs are related, if the task difficulty is ruled out.

6. Discussion

This article proposes an overview of the literature concerning overoptimism, overconfidence, and entrepreneurship. After selecting the relevant literature, the selected articles were reviewed. This has been done to create a clear overview about the literary differences, and the issue about overconfidence. Table 2 provides a summary of the critical review of all the relevant literature. This summarized table can be used to quickly read all the literature about

overconfident and overoptimistic entrepreneurs. Besides this, table one provides an overview of the findings that are robust, and the research that needs to be redone, structured per construct of overconfidence and overoptimsim. This table clearly shows studies that are done within each construct, providing an overview of possible research gaps and new future research

possibilities

To get to the relevant literature this study provided clear searching criteria. Besides this, critically reviewing of the articles has been done structurally and consistently. One could argue that the critical reviews are biased through the author. Since only one person critically reviewed the articles, the results are biased by the perceptions and views of this person. This research therefore proposes that this study should be redone, to test the robustness of the findings. Besides this, the overplacement construct has remained its name and therefore not provided with an appropriate definition. Therefore this research proposes to further research this construct and provide a definition.

There are also some future research propositions that came out of the literature review. First of all, Forbes (2005) states that an understanding of how quickly entrepreneurs

susceptibility to overconfidence changes over time is necessary. One could argue that Koellinger et al., (2007) provide evidence for this, by stating that overconfidence changes as one moves from being a nascent entrepreneur to being an established entrepreneur. However this evidence does not yet exists for the overprecision construct, or as we now call, the real overconfidence.

Secondly, relation between self-efficacy and the constructs overestimation and overplacement should be further researched. It could be that higher self-efficacy creates an overestimation of one’s success. Since self-efficacy can be partially explained by learning in entrepreneurship courses, entrepreneurial experience, and risk propensity, it could also

Referenties

GERELATEERDE DOCUMENTEN

Het plangebied, met een oppervlakte van 7.380m², bevindt zich aan de zuidwestelijke zijde van het  centrum  van  Gent,  gelegen  binnen  het  bouwblok  dat 

Tja, soms is de oplossing direct

Thus, the present study adopts a qualitative approach and explores psychology, science and engineering stu- dents’ conceptualizations of mental health through semi-

heterostructures grown on Si(001), employing a high temperature stable, sacrificial oxide template mask to obtain freestanding cantilever MEMS devices after substrate etching..

Two flexure hinge types are optimized for high support stiffness and high first unwanted eigenfrequency for two different working ranges, ±5.7° and ±20°.. We show how multiple

The presented approach for a target oriented integration of Industrie 4.0 in lean production systems integrates design thinking elements into the value stream mapping

The measured 21st harmonic yield for the cluster jet (black circles), calculated 21st harmonic yield for pure monomers (blue line ) and the liquid mass fraction, g, (red circles)

Reference points A, B, C, and D on well identified corners are selected and measured on the panoramic image to compute its pose by the developed method of oblique