Risk-taking in entrepreneurs, compared to criminals and students: The role of uncertainty and stakes

S

RT

1

2

K1

a

I3

4b5

6

a7

8

A9

R10

R11

A12

13

J14

215

K16

R17

E18

C19

A20

A21

22

t23

d24

n25

c26

a27

b28

n29

30

k31

a Q132

a33

o34

e35

n36

1

RR

EC

TED

PR

OO

F

ARTICLE IN PRESSG ModelOCECO 756 1–11

The Journal of Socio-Economics xxx (2008) xxx–xxx

Contents lists available at ScienceDirect

The Journal of Socio-Economics

journa l homepage: www.e lsev ier .com/ locate /soceco

isk-taking in entrepreneurs, compared to criminals and students:he role of uncertainty, stakes�

lara Faragoa,1, Orhidea Kissa,∗, Janos Borosb

Eotvos Lorand University, Faculty of Education and Psychology, Institute of Psychology, Department of Psychology of Economics and Decision Making,zabella u. 46, 1064 Budapest, HungaryInstitute of Psychology, Peter Pazmany Catholic University, Egyetem u. 1, 2087 Piliscsaba, Hungary

r t i c l e i n f o

rticle history:eceived 17 February 2007eceived in revised form 6 February 2008ccepted 6 April 2008

EL classification:8 002 Z2 Economic psychology

eywords:isk-takingntrepreneursonvictsmount stakedffect heuristic

a b s t r a c t

Entrepreneurs’, college students’ and criminals’ risk-taking were compared in hypotheticalbet situations. The level of uncertainty and the amount staked were varied in gain and losssituations. Potential profit motivates entrepreneurs to choose higher stakes, expected losswill prompt the avoidance of risk. In profit-making situations, college students’ strategiesare different: Students’ decisions are more influenced by probability factors than by theamount staked. Risk-taking tendency of criminals is higher than that of the other two groups,without applying a consistent strategy in taking risk.

© 2008 Published by Elsevier Inc.

Our stereotyped image of various groups is typically linked to their inclination to take risks. We expect increased risk-aking in certain age groups (adolescents), discover it in people’s favorite activities (e.g. hang gliding), in lifestyles (alcohol andrug use), in occupations (entrepreneurs, car racers), and in defiance to social norms (convicts). Risk-taking has positive andegative meaning as well. Entrepreneurs’ behavior when taking risk is considered adaptive, thus has positive meaning, whileommitting criminal act is considered as a maladaptive behavior, thus negative. In the present study we try to distinguishdaptive and non-adaptive risk-taking by concentrating not only on the amount of risk taken by different groups of people,ut on the strategy of the risk-taking behavior. With this aim in mind we focus the attention on the risk-taking of two groups,amely, that of entrepreneurs and convicts, and compare them to a control group, students.

Risk-taking was studied from a normative point of view in the decision-making framework. It is a well-documentednowledge that people are not consistent in their risky choices. The inconsistency toward risk was demonstrated by Tverskynd Kahneman (1981); Tversky et al. (1990). In the same situation formulated in a different way, people preferred different

UN

CO

Please cite this article in press as: Farago, K., et al., Risk-taking in entrepreneurs, compared to criminals and students: The role of uncertainty, stakes,J Socio-Economics (2008), doi:10.1016/j.socec.2008.04.004

lternatives because of different framing, and proved to be risk avoidant in gain, and risk seekers in loss situations. Peopleften choose certain alternatives even if their expected usefulness is smaller than that of the risky alternative. Taking thexample of lottery: It is definitely not worth playing in the lottery, since the probability of the return of the invested money isegligible, but the hope for a big prize and the small amount of the investment clouds rationality. Yet people playing on the

� The research was funded by OTKA, Grant No. TO 37607.∗ Corresponding author. Tel.: +36 1 461 2600; fax: +36 1 461 2697.

E-mail addresses: [email protected] (K. Farago), [email protected] (O. Kiss), [email protected] (J. Boros).1 Institute of Psychology, Peter Pazmany Catholic University, Egyetem u. 1, 2087 Piliscsaba, Hungary.

053-5357/$ – see front matter © 2008 Published by Elsevier Inc.doi:10.1016/j.socec.2008.04.004

dx.doi.org/10.1016/j.socec.2008.04.004

http://www.sciencedirect.com/science/journal/10535357

mailto:[email protected]



dx.doi.org/10.1016/j.socec.2008.04.004

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

NC

OR

RE

CTE

D P

RO

OF

ARTICLE IN PRESSG ModelSOCECO 756 1–11

2 K. Farago et al. / The Journal of Socio-Economics xxx (2008) xxx–xxx

lottery are not risk takers, and many of them strive for high levels of certainty in other areas of life. The example of the lotteryalso proves that decision makers in everyday life do not think along the lines of normatives. Sometimes the expected amountof profit attracts them, disregarding probabilities, while at other times the desire for certainty is their decisive motive, andthey do not plunge into situations of great opportunities.

The decision maker may choose along a criterion, which is in harmony with his/her personal strivings. Game theory hasidentified several such strategies: e.g. persons following the maximax strategy optimistically chooses the alternative in whichthe chances of winning are the highest, regardless of losses and probabilities; maximin is a pessimistic choice, where theperson examines how s/he can get the most out of the bad outcomes; while minimax starts from the regret s/he would feel ifs/he missed a profit by not choosing the alternative. This analysis assumes consistency of choice on the part of the decisionmaker, and the strategies may be translated into the terms of risk-taking (maximax) and risk avoidance (maximin).

Risk-taking is always in the fore when characterizing entrepreneurs. This is not surprising, since due to the incalculablebehavior of consumers and competitors, entrepreneurs unavoidably face risks when making decisions (e.g. about entering anew market, or manufacturing a new product). Although different aspects of the behavior of entrepreneurs are emphasizedby various authors (e.g. readiness for innovation by Schumpeter, 1934; recognition of opportunities by Kirzner, 1973; self-confident and independent decision-making in uncertain situations by Knight, 1921), all of the views include the opinionthat entrepreneurs do not shrink from risky solutions if they hope for great profit or success. Stewart et al. (1998) proposethat individual personality features, like risk-taking propensity, innovativeness and achievement motivation are the mostimportant characteristics of entrepreneurship. According to Sitkin and Pablo (1992) risk propensity is the determinant ofrisk behavior in organizations.

Instead of personality characteristics, a number of researches have focused on the process of risky decision-making ofentrepreneurs.

The studies conducted by Palich and Bagby (1995) set out from the observation that entrepreneurs may behave in variousways: Some grab opportunities, others get frightened at risks; some carry flourishing businesses, others become bankrupt.The authors recommend taking a cognitive approach instead of the supposedly futile search for personality characteristics.In their studies, entrepreneurs and non-entrepreneurs had to form opinions about various business scenarios. The authorsfound a difference between the groups in the categorization of the situations: The entrepreneurs found the situations morepositive, noticed the strengths more than the weaknesses, saw more opportunities than threats, and discovered more progressthan failure in these situations. However, no difference between the tendencies of risk-taking was found between the groups.

Forlani and Mullins (2000) complain that despite the importance of risk-taking, there are relatively few studies regardinghow behavior is influenced by the various components of risky behavior, i.e., by risk perception (how dangerous the subjectperceives the situation to be), by expected outcome (extent of profit and loss), by probability (the chances of an event takingplace), and by willingness of taking risks. In their study, entrepreneurs evaluated hypothetical businesses. It was found thatthe respondents were willing to take greater risks if the situation promised great enough profit (they could choose fromfour possible investments: Business risking a smaller or greater amount of money, associated with a smaller or greaterprobability of success). Great risk-taking, however, was present only in cases of large amounts of money. The respondentsavoided highly uncertain situations. It was not only risk itself, but also its different perception and the individuals’ risk-takingtendency also influenced the respondents’ behavior. Risk-taking tendency had no effect on perception, but did on choice. Ina field experiment where the impact of the available resources, the aspiration level, and the risk-taking propensity of theindividual was studied Farago (2005) found that students took much smaller risk in real situations where they had to facereal consequences, compared to a control laboratory group. Risk-taking tendency diminished in the loss, compared to gaincondition.

Shapira (1995) investigated manager’s risk-taking and found that managers focuses their attention on the amount of gainor losses and tend to disregard probability.

Much research and several related theories have identified risk-taking as an important component of criminal behavior.Gottfredson and Hirschi (1990) demonstrated the connection between criminal behavior and poor self-control in their theory,considering risk-taking an important component of self-control. Grasmick et al. (1993) and Romero et al. (2003) consideredself-control as a multidimensional construct whose leading factor was impulsive risk-taking. One of the key momentumsof Farrington’s (1995) theory based on a 24-year longitudinal study is decision-making prior to the criminal act, which isbased on the cost-benefit analysis well known from economics, i.e., if the expected benefit is greater than the cost (beingcaught), the criminal act will be very likely to take place. Naturally, the evaluation of both cost and that of benefit are highlysubjective, they greatly depend on the personality, values, attitude and previous experiences of the person and on the givensituation. Thus, the criminal perpetrator always takes a great risk when making a positive decision.

Violent criminals are doubly risk takers: On the one hand, the criminal act itself involves a risk (counter-attack by theattacked person or attack by the arriving authorities), and on the other hand, the person risks imprisonment if got caught.In his phenomenological approach, Lyng (1993) proposes that there are important instances of similarity between risk-taking and criminal behavior with respect to the experience. In his view, the experience of the extension of control and
U

the bombardment of boundaries add important plus to the actor, thus, risk-taking is a strong reinforcer of self-image. Theexperience and the motivation are similar, but the difference is that well-socialized persons direct their risk-taking behaviorto socially accepted areas, while poorly socialized persons act on their similar tendencies in socially unaccepted fields.

Studying the relationship between personality and criminal behavior, and based on the personality test (MPQ) of 800young people, Krueger et al. (1994) found no difference between criminals and non-criminals along the dimension of risk-

dx.doi.org/10.1016/j.socec.2008.04.004

S

t98

r99

c100

r101

o102

e103

104

h105

“106

r107

i108

w109

1110

H111

i112

H113

H114

H115

H116

g117

H118

f119

d120

2121

2122

Q2123

t124

125

G

A

E

M

R

OR

RE

CTE

D P

RO

OF


K. Farago et al. / The Journal of Socio-Economics xxx (2008) xxx–xxx 3

aking. Lane and Cherek (2000) asked their adolescent subjects to choose between risky and non-risky alternatives in aisk-taking task. The tasks called for big profits with low probability and small losses with high probability. The potentiallyriminal youngsters chose risky solutions more often, gained smaller profits by the end of the experiment, and if taking aisk proved to be a lucky choice, they stuck to it for a prolonged period. Stewart and Hemsley (1984) studied the expectationf winning in criminals, and found that this group expects greater profits in hypothetical criminal situations, because theystimate the level of risk to be lower.

If we investigate risk-taking of entrepreneurs and convicts, it can be assumed that the level of risk-taking tendency isigher in both entrepreneurs and criminals, than in other subjects. We can assume that entrepreneurs are experienced andprofessional” risk takers, since risk-taking plays such an important role in their life. It can be expected, that they have a moreational or at least more consistent strategy of risk-taking than convicts, the latter having a disregard for conditions. Sincet is vital for entrepreneurs to successfully avoid big losses, and convicts tend to underevaluate risk and act in an impulsiveay we expect criminals to be even more risk seekers than others.

. Hypotheses

1. As entrepreneurship include risk-taking, we expect the level of risk-taking to be higher in the entrepreneur group thann the control group (college students).

2. Risk-taking of convicts will be higher than in the other two groups.

3. We expect people to take bigger risk in the loss situation than in the gain situation.

4. Convicts do not strive for security; therefore, they will take big risk in both situations.

5. The size of stakes will influence risk-taking more than the differences of probabilities in all groups, since people ineneral tend to disregard probabilities.

6. It will be possible to identify a consistent strategy in the case of entrepreneurs and students – we assume, maximinor the former, minimax for the latter group. Convicts will not have a consistent strategy, they will not differentiate betweenifferent situations.

. Method

.1. Samples

95 entrepreneurs, 86 college students and 49 convicts filled out the questionnaires in 2002 and 2003. The distribution ofhe samples was as follows:

Entrepreneurs

enderMales 68Females 27

ge25–35 2936–45 2946–55 3055+ 6

ducationSecondary 42

UN

C


Higher 48No data 5

onthly net incomeUnder 70,000 Forints 17Between 71,000–250,000 Forints 46251,000+ Forints 20No data 12

esidenceBudapest 36Town with more than 50,000 inhabitants 20Town with less than 50,000 inhabitants 24

dx.doi.org/10.1016/j.socec.2008.04.004

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

CO

RR

EC

TED

PR

OO

F



No data 15

Number of employees0 201 172–3 194–10 2111–50 950–100 4

As can be seen the respondents were small business entrepreneurs. This distribution does not reflect the intention of theinvestigators, only that these were the persons who were available for the study. The majority of the questionnaires weredistributed by university students by the so called snowball technique. Due to the difficulties of availability, the criterion forselection was having risked their own property in their business.

2.2. College students

The questionnaires were filled out by 1st and 2nd year students of the Faculty of Foreign Trade of the Budapest BusinessSchool.Gender

Males 15females 71

Age18–19 520–25 81

Convicts in prison

GenderAll adult males

Type of crime committedBurglary 24 (3 of them with a history of burglary)Armed robbery 25 (21 of them with a history of robbery)

Age21–25 925–35 2936–45 1046–55 1

EducationPrimary (8 grades) 41Secondary 8

The questionnaires were distributed by two psychologists in the prison.

2.3. Materials

In the present study, we used the bidding paradigm. We offered gain and loss situations to the subjects. The two alterna-tives had equal expected value. In such case, no normative prescriptions are violated either by always choosing the certainalternatives (these persons are called risk-avoiders) or by always choosing uncertain alternatives that promise greater profits(risk takers). The size of the bid, and the degree of uncertainty (we offered different probability situations) was manipulated.The strategies of entrepreneurs were compared with those of convicts, and the responses of college students were used ascontrols to be compared with them.

The respondents had to reply 10 questions in the questionnaire, 5 of them related to winning situations, 5 related tolosing situations2. In each question a certain and a risky alternative was offered. The value of the certain alternative (extentof certain profit or loss), and the magnitude of stake and the probability of occurrence in the uncertain alternative werevaried from question to question.

Let us see one example of the questions

UN


(a) You certainly get 300,000− Forints.(b) There is a 20% chance that you get 1,500,000− Forints and a 80% chance that you get nothing.

The characterization of the situations can be seen in Table 1.

2 See questionnaires in Appendix A.

dx.doi.org/10.1016/j.socec.2008.04.004



Table 1Characterization of the situations

Situation no. Value of certain profit (win) Size of stakes Probability of improving the starting position

1 Great Small Great2 Great Great Small3 Small Small Great4 Small Great Small5 Moderate Moderate Moderate

Value of certain loss Probability of worsening of the starting position

6 Great Small Great

3155

156

c157

(158

t159

d160

161

t162

t163

r164

165

a166

t167

168

g169

t170

171

a172

173

b174

a175

176

(177

178

p179

s180

TM

WWWWW

T

LLLLL

TT

CTE

D P

RO

OF7 Great Great Small

8 Small Small Great9 Small Great Small

10 Moderate Moderate Moderate

. Results

First, we compared the risk-taking of the male and female participants. This is important because the group of convictsonsisted of males only, the students group consisted mainly of women (71 vs. 15) and the entrepreneurs were mainly males68 vs. 27) The difference between the genders were significant (males 4.78; females 3.88; p = 0.001). For winning situations,he significant difference disappear, and the remaining difference for the loosing block is explained by the only significantifference of question 10.

We did not find useful to compare risk-taking for age. Our student group were clearly the group of young people comparedo the others. According to the results of almost all studies, young people are bolder risk takers than older people. Accordingo our hypothesis, we expect the other two groups to take bigger risk, not the students. If we will not find the expectedesults, than we can attribute the negative results to the effect of age and blame the unmatched sample for the lack of results.

The responses to the questions were coded dichotomically: Risky choices scored 1 certain choices scored 0. First theverage scores of the responses are shown in the three groups for all questions, for the average total win, for the averageotal lose, and for the combined average total risk-taking.

As it can be seen in Table 2, the highest values were given by the convicts not only for the wins and losses, but for thelobal risk-taking indicator as well, while the entrepreneurs scored higher than college students, i.e., they are more riskakers than the students.

The differences among the three groups were statistically significant in the global indicator (ANOVA: F = 7.706, p < 0.001)nd in the total loss indicator (F = 3.135, p < 0.05).

In winning situations risk-taking did not differ significantly among the three groups, there was only a tendency of differenceetween the college students and the entrepreneurs (t = −1.806, p = .0.072). In losing situations, however, college studentsnd convicts differed from each other significantly (t = −2.38, p < 0.01).

UN

CO

RR

E


It was analyzed whether risk-taking of winning and losing situations differed from each other, as prospect theory suggestsFig. 1).

The differences in all cases are statistically significant (total sample: t = −9.698, p < 0.001; entrepreneurs t = −6.745,< 0.001, college students t = −5.151, p < 0.001, convicts t = −4.803, p < 0.001), i.e., everybody took much greater risks in losing

ituations.

able 2ean values of risk-taking

Entrepreneurs College students Convicts

Average S.D. Average S.D. Average S.D.

in 1 0.21 0.41 0.18 0.39 0.31 0.46in 2 0.43 0.50 0.21 0.41 0.31 0.46in 3 0.34 0.48 0.50 0.50 0.41 0.50in 4 0.44 0.50 0.21 0.41 0.40 0.50in 5 0.42 0.50 0.40 0.49 0.47 0.50

otal Win 1.83 1.44 1.49 1.15 1.88 1.7

ose 6 0.82 0.39 0.72 0.45 0.73 0.45ose 7 0.39 0.49 0.42 0.50 0.54 0.50ose 8 0.71 0.45 0.54 0.50 0.65 0.48ose 9 0.41 0.50 0.48 0.50 0.63 0.49ose 10 0.60 0.49 0.45 0.50 0.65 0.48

otal Lose 2.49 1.29 2.61 1.38 3.18 1.25otal risk-taking 4.69 2.09 4.20 1.77 5.06 2.16

dx.doi.org/10.1016/j.socec.2008.04.004

D P

RO

OF



Fig. 1. Risk-taking in winning and losing situations.

181

182

183

184

TEFig. 2. Risk-taking of entrepreneurs in winning and losing situations.

UN

CO

RR

EC


Let us examine the questions in which the respondents were the most risk takers. Remember, score 1 was given for risk-taking, and zero score was given for choosing the certain alternative. For any question, group average over 0.5 means morerisky choices than risk avoidance in the group.

The level of risk-taking in the three groups is demonstrated in Figs. 2–4.

Fig. 3. Risk-taking of college students in winning and losing situations.

dx.doi.org/10.1016/j.socec.2008.04.004

F



185

a186

c187

T188

189

s190

191

192

o193

c194

Q3195

s196

t197

w198

199

3200

201

(202

203

s204

205

206

207

208

209

w210

211

212

213

p214

3215

216

217

218

219

CO

RR

EC

TED

PR

OO

Fig. 4. Risk-taking of convicts in winning and losing situations.

Every group proved to be risk takers in situations 6 and 8. People want to take risk in losing situations, where the stakesre not high. The value of the certain alternative is great in one case and low in the other. Thus, this variable has no effect onhoice. In both situations, the chances of loosing some more is high, indicating that probability does not deter from risk-taking.hus, if the stakes are not high, people are ready to take big risks in order to avoid losses.

Group average was 0.5 or higher in Situations 6, 8, and 10 in the entrepreneur group, Situations 3, 6, and 8 in the collegetudent group, and Situations 6, 7, 8, 9, and 10 in the convict group.

Convicts are risk takers in all losing situations.There is only one case in the gain situation which involved more risk-taking than risk avoidance: College students (as

pposed to the other two groups) were risking a profit in situation 3 where the initial value is small and a little more moneyould be won with great probability (i.e. if stakes are low).

Now let us examine the situations with great uncertainty, 50%, and moderate outcome (5. for gain and 10 for loss). Collegetudents took no risks in situation 10, entrepreneurs and convicts, however, took risks in order to avoid loss, even whenhere was great uncertainty. Convicts, as opposed to the other two groups, are more likely to accept great uncertainty even ininning situations (situation 5). The differences in response tendencies were further studied by comparing the three groups.

These general differences were analyzed in a pair wise comparison by t-test.

.1. Comparison of entrepreneurs and college students

Only tendencies of differences were found in the total risk indicator (t = −1.688, p = 0.093) and in the total win situationst = −1.806, p = 0.072) in favor of the entrepreneurs.

On the basis of the responses to the individual questions, it can be stated that entrepreneurs make greater risks than collegetudents in:

Situation 2 (small certain profit, high stakes, great risk3; t = –3.283, p < 0.001).Situation 4 (small certain profit, high stakes, great risk, t = −3.434, p < 0.001).Situation 8 (small certain loss, low stakes, great risk t = −2.401, p < 0.01).Situation 10 (moderate certain loss, moderate stakes, moderate chance––great uncertainty, t = 2.000, p < 0.05).The significance of stakes in entrepreneurs is striking. In cases of winning they make greater risks if the stakes are high,

hile in cases of losing they make greater risks if the stakes are low. Probability does not seem to influence their choice.College students make greater risks than entrepreneurs in:Situation 3 (small certain profit, low stakes, small risk, t = 2.337, p < 0.05).Thus, college students take greater risks than entrepreneurs in situations where the starting value, the stakes, and

robability of the worst outcome are all low, thus, they are careful.

.2. Comparison of convicts and college students

UN


Convicts make greater risks than college students in all questions. The difference is significant in:Total risk-taking (t = −3.560, p < 0.001).Total loss indicator (t = −2.387, p < 0.01).Situation 4 (small certain profit, high stakes, great risk, t = −2.338, p < 0.05).

3 Here great risk means: high probability of the worst alternative.

dx.doi.org/10.1016/j.socec.2008.04.004



Table 3Effects of starting position, stakes and uncertainty in the different groupsQ5

Entrepreneurs College students Convicts

1, 2 vs. 3, 4 (great vs. small starting position) <** <** <*1, 3 vs. 2, 4 (small vs. great stakes) <** >** >1, 2, 3, 4 vs. 5 (small vs. great uncertainty) <* <* <*6, 7 vs. 8, 9 (great vs. small starting position) > > <6, 8 vs. 7, 9 (small vs. great stakes) >** >** >6, 7, 8, 9 vs. 10 (small vs. great uncertainty) < > <

Table 4Variability of the responses (S.D.) in the win and lose situations of the 3 groups

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

Q4245

246

247

248

249

250

251

252

253

254

255

NC

OR

RE

CTE

D P

RO

OF

Entrepreneurs Win Entrepreneurs lose College students win College students lose Convicts win Convicts lose

0.095 0.186 0.141 0.119 0.069 0.065

Situation 10 (moderate certain loss, moderate stakes, great uncertainty, t = −2.328, p < 0.05).There is a great difference between convicts and college students in risk-taking, and this is true primarily in situations of

loss and in those of great uncertainty.

3.3. Comparison of entrepreneurs and convicts

There is no difference between the two groups in the total indicators.Convicts take greater risks in:Situation 9 (small certain loss, high stakes, low probability, t = −2.509, p < 0.01).This is a situation of loss where the stakes are high and the probability of a good outcome is low.In the next analysis, we attempted to separate the effects of the starting position, i.e., those of the initial values, stakes

and probabilities.

3.4. Effects of starting position, stakes and uncertainty in the different groups

The effects of starting positions in the different groups can be seen in Table 3.All three groups avoid risks in winning positions if the value of the certain alternative is high as opposed to low. Thus,

they do not give up a high certain profit for an uncertain gain. On the other hand, they take risks in cases of small certain profits.In cases of losses all three groups make greater risks if the stakes are low. Except for the convicts, who do not differentiate

high or low certain losses, the two other groups take risks mainly in high certain losses. Stakes show consistent reversal ofpreference only in the entrepreneur group: They are more willing to take risks in cases of high stakes in winning situations andin cases of low stakes in losing situations. Situations 7 and 9 (great loss with low probability) are so repulsive for entrepreneursthat they not only give up their strategy of risk-taking used in winning situations, but they take outright smaller risks inthese losing situations than in similar winning situations (averages scores: Situation 2 = 0.43, Situation 7 = 0.39, Situation4 = 0.44, Situation 9 = 0.41).

If we want to highlight the differential features of the different groups, we can say that

- Entrepreneurs are motivated to take risk by the possibility of large gain, while students prefer high probability.- Convicts are less afraid of big losses when taking risk than the other two groups.

To demonstrate the variability of responses to the different risk-taking situations, we compared the standard variationsof the means separately to win and lose situations for each groups Table 4.

The means differ significantly both within the gain and within the loss situations in the entrepreneurs’ (p < 0.001) andstudents’ group (p < 0.001), but not in the convicts’ group (for gain: p = 0.201) (for loss: p = 0.512). We can conclude that whilethe former two groups differentiate between situations having different characteristics when taking risk, convicts use similarstrategies regardless of the situation.

4. Discussion

U


The results analyzed by several methods point in the same direction. On the one hand, some definite general tendenciesappear that show the typical reactions manifested in experimentally studied bet situations, independently of the group-membership of the respondents (entrepreneurs, convicts or college students).

One of the robust relationships was the willingness to take risk in loss and avoid risk in the win situation. In situationswhere the respondents had to choose between keeping a certain profit and taking the chance of losing it in hope of an even

dx.doi.org/10.1016/j.socec.2008.04.004

S

g256

t257

n258

o259

t260

261

w262

263

T264

t265

m266

267

c268

r269

o270

b271

p272

n273

t274

s275

l276

277

a278

279

g280

t281

s282

o283

e284

285

r286

c287

288

r289

u290

t291

s292

293

t294

l295

o296

t297

n298

i299

a300

i301

c302

g303

T304

l305

t306

i307

d308

a309

t310

w311

I312

h313

314

a315

h316

NC

OR

RE

CTE

D P

RO

OF



reater profit: People generally choose certain profit (“A bird in the hand is worth two in the bush”), i.e., they do not like toake risks. This is all the more so when the certain profit is great enough and the further gain achievable with further risk isot substantial. If the frame is loss, where the choice is between suffering certain loss and taking the risky alternative wherene could get rid of the loss, but if unlucky, could meet an even greater loss, people take significantly greater risks than inhe case of profit-frame.

This general tendency proves that keeping the profit is an important motive in situations that are considered as winning,hile in losing situations, people are more willing to take the chance of an even greater loss for the sake of avoiding the loss.

As we have pointed it out, the expected value of risky and certain alternatives were the same in the same situation.heoretically it could be expected that even if people do differentiate between positively and negatively introduced situations,hey would consistently choose between the strategies of striving for safety and taking risks. If this is not so, it means that the

agnitude of stakes, the probability, or the initial value plays a greater role than the others when taking a risk is considered.No group risked certain big profit (winning situation) in the cases of low stakes. Risk-taking proved to be the most

onspicuous in losing situations where the probability of further loss was rather high, but its level was low (low stakes). Inisky situations, however, where the level of further loss was threateningly high, people were less willing to take risks, in spitef the high probability of getting rid of the loss completely. Thus, it can be concluded that risk-taking in loss is influencedy the level of stakes: if the stakes are high, we tend to take less risks than if the stakes are low. Less attention is devoted torobability (the chances for the occurrence of a good or a bad event) that is also important in the evaluation of risks; it doesot influence taking risks. In losing situations convicts differ from the other two groups: They are less diverted from risk-aking by the possible loss inherent in risk-taking than the others, and they seem to weigh risky situations less––we foundmaller differences in their reactions to the different situations of risk-taking than in the other groups. They differentiateosing situations even less so either by the level of stakes or by the extent of certain alternative.

In addition to the general tendencies, let us turn our attention to the differences among the groups in their risk-taking,nd find the characteristics of the entrepreneurs and the convicts.

As we have mentioned, nobody risks certain great profit in case of low stakes, but entrepreneurs are willing to risk certainreat profits in case of high stakes (despite the low probability of a favorable outcome) in winning situations. As opposedo this, college students choose the uncertain, but greater profit-promising alternative if the value of the certain profit ismall, if the stakes are low and the probability is high. Thus, in winning situations entrepreneurs are motivated by grabbingpportunities even if it is unsafe. Low stakes, however, motivated the risk-taking behavior of only college students, notntrepreneurs. In the preferences of convicts, it was uncertain situations (50% of probability) that ranked first.

In losing situations entrepreneurs definitely wanted to avoid great loss (even if its probability was low): These situationsanked among the more cautious solutions for them. College students and convicts were relatively more daring here (asompared to the other situations).

It can be concluded that the level of stakes bears greater significance for the entrepreneurs in both situations, and theole of probability definitely falls back. It is another important difference among the groups, that it is convicts who acceptncertainty the most, situations with 50% probability rank higher in the rank order of risks for them than in the otherwo groups, and entrepreneurs tolerate uncertainty more than college students do, the difference between them reachingtatistical significance in the case of loss.

What conclusions can be drawn from all this with respect to the risk-taking of entrepreneurs? The outstanding role ofhe level of stakes and the definite loss of importance of probabilities produces a somewhat paradoxical situation. If weook at it from the aspect of how much the respondents strived at improving their situation, i.e., instead of concentratingn the dimension of profit–loss we focus on the dimension of improving or worsening the starting position, we will findhat in positively framed situations they strongly strive at improving the situation even if the probabilities are low, while inegatively framed situations they are reluctant to take great risks (perceived as great probably because of the stakes) even

f the probability of improvement is high. If the stakes are low (low perceived risk), however, the direction is reversed: Inwinning frame they are not motivated for improvement, while in a losing situation they are. Our results concerning the

mportance of stakes for entrepreneurs were confirmed by a recently published article (Kamalanabhan et al., 2006) theyollected data on two measures of risk-taking from entrepreneurs, prospective entrepreneurs and non-entrepreneurs. Theroups differed significantly only on the Magnitude of Loss Questionnaire, but not on the Choice Dilemma Questionnaire.heir results also highlight the significance of loss. The authors concluded that the riskiness in business ventures is not theow probability of success but the high stakes involved in entrepreneurship. Translating this into business life we will seehat a profit-making entrepreneur is more willing to take risks, while a losing colleague will be more cautious. It is alsonteresting how our results modulate the process of decision made known as endowment effect by Thaler (1980). Thus, theesire to justify former decisions makes entrepreneurs not to accept the loss of bad investments. Therefore, they invest morend more money into the business; try to save what cannot really be saved. This effect could be seen in our study only ifhe prospect of further loss was not perceived as threatening by the respondents. If the stakes were threateningly high, theyould rather accept the loss. If they calculated with probabilities as well, their reactions would, perhaps be just the opposite:
U

nstead of enduring further small losses repeatedly, they could make the most of a good opportunity. It works against them,owever, that in case of failure they could not justify their decisions that brought big losses.

Our results also support those of Forlani and Mullins (2000), according to which members of the entrepreneur groupre willing to take greater risks if the situation promises a big enough profit. As opposed to the Forlani and Mullins study,owever, in highly uncertain situations our respondents did not avoid risk-taking; in fact, in highly uncertain loss situations

dx.doi.org/10.1016/j.socec.2008.04.004

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

NC

OR

RE

CTE

D P

RO

OF



– not studied by Forlani and Mullins – they proved to be definitely risk takers, rather than risk-avoiders. Thus, entrepreneurscould well tolerate a high degree of uncertainty in our bet situation.

As to the characterization of the risk-taking of convicts, it can be highlighted in the results that convicts’ general level ofrisk-taking is higher than that of the other groups, and that they take greater risks in uncertain situations than the other twogroups. They took greater risks in all of the losing situations than in the winning situations. It can be concluded from this thatthe differentiation of the situations is very rough. Contrary to the other two groups, convicts are not threatened by enduringthe (small) possibility of very big losses. Thus, our research has also confirmed the higher risk-taking tendency of criminals.This seems to be a part of their life strategy, which is not surprising, as they spend a great part of their lives in dangerousplaces like the outskirts of cities, residential areas and pleasure grounds of slums, prisons: the places of crime. It is impossibleto live in these territories “non-dangerously”; they must take risks 24 h a day in order to survive. Socialization seems to bedifferent in this milieu that forms risk-taking strategy. They are motivated by the smaller, but more certain profit; they dowait for “striking oil” at any price. Yet at the same time, the possibility of failure is inherent this lifestyle, often becoming areality indeed, like being beaten in a fight, or being caught in a burglary. They do not want to avoid this loss; they accept ifthey lose, a prison-sentence does not mean a tragic situation for them; it belongs to their everyday life, accepted by boththeir environment and their relatives.

If we compare our results with those of Slovic et al. (2002), we had to count on the outcome that – since we offered onlya hypothetical situation, and, thus we actually asked for preference judgments and the respondents did not have to makereal bets in real money – in winning situations the respondents would choose the alternative in which they could win asmall amount of money with great probability. The reason is, that affects heuristics, as we have pointed out, accentuatesprobabilities more in these situations when making judgments. This was really true for college students and convicts, butnot for entrepreneurs. If we assume after Lopez (1993) that entrepreneurs are motivated by the perception of opportunitiesand the motive for success, then – as a result of the operation of affect heuristics – the emotional value of the stakes will be sohigh that their attention is turned toward that, and its appeal will not be diminished by the relatively low level of probability.The significance of the stakes can be observed in losing situations as well: Failure has a negative emotional value, and thepossibility of a great loss pushes the role of probabilities into the background again. The effect of individual differences inaffective reactions on risk-taking was demonstrated by Peters and Slovic (2001) in an experiment. They compared personswho described themselves as reacting more sensitively to either negative or positive events. Persons belonging to the firstgroup did not chose from the deck of cards that included the possibility of losing a lot, while the other group chose the deckthat promised the great hit. Thus, the individual differences in affective reactions have an effect on the learned patternsof risk-taking and its behavioral manifestation. Similar differences can be assumed between groups of similar people. Thispossible explanation supports the emotion-based interpretation of the differences found among the different groups.

Our investigation may justly be criticized on the ground that comparisons were not made on matched samples; unfor-tunately, however, we could not create the conditions for this. This is why it came at the just right moment that despite theyounger sample, level of risk-taking was biased toward the entrepreneur group; since the age differences would predict anopposite relationship; this circumstance only reinforces our results. A shift in gender in favor of women, however, slightlydecreases the power of the differences found, especially for the difference between the students and the other two groupsmainly for situation 10.

Finally, it must be mentioned that the conclusions are based on the investigation of artificial, hypothetical, and laboratorydecisions and bet situations. Asking about different types of risks, and conducting the study in real situations would probablyproduce different reactions and a number of other influences that were not studied here.

Appendix A

I am inviting you to an imaginary winning game. Please imagine yourself to be in each of the following decision-situation,as if you really received or had to pay the amounts of money indicated below. You can always choose from two possibilities.Please circle your choice.

1. (a.) There is an 80% chance that you get 400,000− Forints and a 20% chance that you get nothing, or (b.) you certainly get320,000− Forints.

2. (a.) You certainly get 300,000− Forints, or (b.) there is a 20% chance that you get 1,500,000− Forints and an 80% chancethat you get nothing.

3. (a.) There is a 90% chance that you get 200,000− Forints and a 10% chance that you get nothing, or (b.) you certainly get180,000− Forints.

4. (a.) You certainly get 160,000− Forints, or (b.) there is a 10% chance that you get 1,600,000− Forints and a 90% chance
U

that you get nothing.5. (a.) There is a 50% chance that you get 500,000− Forints and a 50% chance that you get nothing, or (b.) you certainly get

250,000− Forints.6. (a.) There is an 80% chance that you must pay 400,000− Forints and a 20% chance that you have to pay nothing, or (b.)

you certainly must pay 320,000− Forints.

dx.doi.org/10.1016/j.socec.2008.04.004

S

373

374

375

376

377

378

1379

380

R381

F382

F383

384

F385

G386

G387

388

K389

390

K391

K392

K393

394

L395

L396

L397

P398

399

P400

401

R402

403

S404

S405

S406

S407

408

S409

S410

411

T412

T413

T414

ED

PR

OO

F



7. (a.) You certainly must pay 300,000− Forints, or (b.) there is a 20% chance that you must pay 1,500,000− Forints and an80% chance that you have to pay nothing.

8. (a.) There is a 90% chance that you must pay 200,000− Forints and a 10% chance that you have to pay nothing, or (b.) youcertainly must pay 180,000− Forints.

9. (a.) You certainly must pay 160,000− Forints, or (b.) there is a 10% chance that you must pay 1,600,000− Forints and a90% chance that you have to pay nothing.

0. (a.) There is a 50% chance that you must pay 500,000− Forints and a 50% chance that you have to pay nothing, or (b.) youcertainly must pay 250,000− Forints.

eferences

arago, K., 2005. Kockazatvallalas vizsgalata terepkıserletben. Magyar Pszichologiai Szemle 3, 343–371.arrington, D.F., 1995. The psychology of crime: influences and constraints on offending. In: Bull, R., Carson, D. (Eds.), Handbook of Psychology in Legal

Context. John Wiley & Sons, pp. 291–315.orlani, D., Mullins, J.W., 2000. Perceived risks and choices in entrepreneur’s new venture decisions. Journal of Business Venturing 15, 305–322.ottfredson, M.R., Hirschi, T., 1990. A General Theory of Crime. Stanford University Press, Stanford, CA.rasmick, H.G., Tittle, C.R., Bursik, R.J., Arneklev, B.J., 1993. Testing the core empirical implications of Gottfredson and Hirschi’s general theory of crime.

Journal of Research in Crime and Delinquency 30, 5–29.amalanabhan, T.J., Sunder, D.L., Manshor, A.T., 2006. Magnitude of loss scaleevaluation of entrepreneurial risk-taking using magnitude of loss scale. Journal

of Entrepreneurship 15, 37–46.irzner, I.M., 1973. Competition and Entrepreneurship. University of Chicago Press, Chicago.night, F.H., 1921. Risk, Uncertainty and Profit. Kelly and Millman, Inc., New York.rueger, R.F., Schmutte, P.S., Caspi, A., Moffitt, T.E., Campbell, K., Silva, P.A., 1994. Personality traits are linked to crime among men and women: evidence

from a birth cohort. Journal of Abnormal Psychology 103 (2), 328–338.ane, S.D., Cherek, D.R., 2000. Analysis of risk taking in adults with a history os high risk behavior. Drug and Alcohol Dependence 60, 179–187.opez, L.L., 1993. Reasons and resources: the human side of risk taking. In: Bell, N.J., Bell, R. (Eds.), Adolescent Risk Taking. Sage, pp. 29–57.yng, S., 1993. Disfunctional risk taking: criminal behavior as edgework. In: Bell, N.J., Bell, R.W. (Eds.), Adolescent Risk Taking. Sage, pp. 107–131.alich, L., Bagby, D.R., 1995. Using cognitive theory to explain entrepreneurial risk taking: challenging conventional wisdom. Journal of Business Venturing

10, 421–438.eters, E., Slovic, P., 2001. The springs of action: affective and analytical information processing in choice. Personality and Social Psychology Bulletin 26,

1456–1475.omero, E., Gomez-Fraguela, J.A., Luengo, M.A., Sobral, J., 2003. The self-control theory construct in the General Theory of Crime: an investigation in terms

of personality psychology. Psychology, Crime, and Law 9, 61–86.chumpeter, J., 1934. The Theory of Economic Development. Harvard University Press, Boston.hapira, Z., 1995. Risk Taking: A Managerial Perspective. Russel Sage Foundation, New York.itkin, S.B., Pablo, A.L., 1992. Reconceptualizing the determinants of risk behavior. Academy of Management Review 1, 9–38.lovic, P., Finucane, M., Peters, E., MacGregor, D., 2002. The affect heuristic. In: Gilovich, T., Griffin, D., Kahneman, D. (Eds.), Heuristics and Biases. The

Psychology of Intuitive Judgment. Cambridge University Press, pp. 397–421.

UN

CO

RR

EC

T


tewart, C.H.M., Hemsley, D.R., 1984. Personality factors in the taking of criminal risks. Personality and Individual Differences 5 (1), 119–122.tewart, W.H., Watson, W.E., Carland, J.C., Carland, W.J., 1998. A proclivity for entrepreneurship: a comparison of entrepreneurs, small business owners, and

corporate managers. Journal of Business Venturing 14, 189–214.haler, R., 1980. Toward a positive theory of consumer choice. Journal of Economic Behavior and Organization 1 (1), 39–60.versky, A., Kahneman, D., 1981. The framing of decisions and the psychology of choice. Science 211, 453–458.versky, A., Slovic, P., Kahneman, D., 1990. The causes of preference reversal. American Economic Review 80, 204–217.

dx.doi.org/10.1016/j.socec.2008.04.004

Risk-taking in entrepreneurs, compared to criminals and students: The role of uncertainty and stakes

Documents

Transcript of Risk-taking in entrepreneurs, compared to criminals and students: The role of uncertainty and stakes