Market economies and pro-social behavior: Experimental evidence from Central Asia

The Journal of Socio-Economics 41 (2012) 64–71

Contents lists available at SciVerse ScienceDirect

The Journal of Socio-Economics journal homepage: www.elsevier.com/locate/soceco

Market economies and pro-social behavior: Experimental evidence from Central Asia Stephen M. Rosenbaum a,∗ , Stephan Billinger a , Nils Stieglitz a , Abdumalik Djumanov b , Yerlan Atykhanov c a

Institute for Marketing and Management, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark Faculty of Management, Westminster International University in Tashkent (WIUT), 12 Akhunbabaev St., Tashkent 100000, Uzbekistan c Kazakh-American University, 29 Toraigirov St., 050043 Almaty, Kazakhstan b

a r t i c l e

i n f o

Article history: Received 29 October 2010 Received in revised form 12 September 2011 Accepted 18 October 2011 JEL classification: C72 C91 P30

a b s t r a c t This study seeks to extend the body of knowledge of pro-social behavior in comparative market settings by reporting on a high-stakes ultimatum game and revelation game experiments in two transition economies: Kazakhstan and Uzbekistan. While controlling for cultural differences and framing effects, we find statistically significant differences in fairness and honesty behavior between the two countries. Specifically, subjects in Uzbekistan (in an earlier stage of transition to a market economy) are fairer and more honest than their later-stage Kazakh counterparts. Our experimental findings have implications for the literature on pro-social behavior and market economies, and more generally, on the transmission process between formal and informal institutions. © 2011 Elsevier Inc. All rights reserved.

Keywords: Market economy Pro-social behavior Experimental economics Ultimatum game

1. Introduction In a series of experiments conducted in small-scale societies, Henrich et al. (2001, 2005, 2010) find strong evidence that pro-social norms1 of fairness and reciprocity increase with greater market integration. The intuitive logic is appealing. Greater market integration dilutes the strong community enforcement mechanisms associated with small-scale societies, increasing the likelihood of opportunism. Recognizing this, and motivated to realize the gains from the greater division of labor, exchange parties respond by rewarding attributes such as fairness and reciprocity in dealings with strangers, while punishing incidences of unfair behavior. Accordingly, greater market integration strengthens prosocial norms toward strangers (Henrich et al., 2001, 2005, 2010).

∗ Corresponding author. Tel.: +45 65503262; fax: +45 66155129. E-mail addresses: [email protected] (S.M. Rosenbaum), [email protected] (S. Billinger), [email protected] (N. Stieglitz). 1 Consistent with the emerging literature (cf. Krupka and Weber, 2009), we distinguish between the underlying “pro-social norms” and “pro-social behavior”, where the former may trigger the latter. Accordingly, in the following we reserve the term “behavior” for actions exhibited by the experimental subjects. 1053-5357/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.socec.2011.10.010

However, in their treatment of market integration, Henrich et al. focus exclusively on the role of informal institutions (such as community enforcement mechanisms) in structuring and regulating exchange in rudimentary societies. Accordingly, their notion of market integration fails to account for (a) the vast, impersonal markets and highly specialized division of labor characterizing present day exchange, and (b) that exchange transactions are governed by both formal and informal institutions (North, 1990). The purpose of this paper is to extend this line of research in both dimensions to examine whether pro-social behavior is stronger or weaker in more market-oriented economies. First, we broaden the focus to incorporate vast markets where exchange transactions are more ephemeral and anonymous (Platteau, 2000) to examine predispositions to be fair and honest toward strangers in situations where the gains from trade are enhanced, but where monitoring and community enforcement mechanisms become correspondingly more diffuse (Kandori, 1992). Second, by explicitly incorporating formal institutions which develop concurrently to support exchange in vast markets (North, 1990), we can examine whether pro-social behavior is crowded-in or crowded-out by more developed, market-compatible formal institutions (Bowles and Gintis, 1998; Frey and Oberholzer-Gee, 1997; Reeson and Tisdell, 2008).

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71

We investigate the above by conducting economic experiments with subjects in two transition societies: Kazakhstan and Uzbekistan. Specifically, we report on two high-stake experiments: the ultimatum game and the (novel) revelation game, which respectively tap into norms of fairness/reciprocity and honesty. In the one-shot ultimatum bargaining game (Güth et al., 1982), a proposer offers an allocation of money to a responder. If the latter accepts the proposed split, the payoffs are divided accordingly, whereas if the respondent declines the offer, neither player receives anything. In the revelation game, subjects must individually and privately decide whether to voluntarily reveal a monetary discrepancy in their favor to the experimenter. The canonical model of self-interest predicts that the respective dominant strategies of the above games would be to (a) offer the lowest possible positive offer (which, being preferable to receiving nothing, would be accepted), and (b) pocket the serendipitous money (Rubinstein, 1982). Economies at different stages of transition provide an appropriate setting to tease out the predominance of pro-social vis-a-vis self-interested behavior. While early-stage economies are characterized by relatively small, homogeneous clan and kinship networks, endemic corruption and concern for the collective interest (Duch and Palmer, 2004; Platteau, 2000), more marketoriented, later-stage economies feature more impersonal trade with increasingly heterogeneous counterparts and enhanced pecuniary incentives. In this regard, Kazakhstan and Uzbekistan were purposefully selected on the grounds that, although exhibiting very high degrees of cultural homogeneity, a common historical heritage and similar initial conditions (Alam and Banerji, 2000; Pomfret, 1995), they display wide variations in the degree to which they resemble fully-functioning market economies, as evidenced by their respective European Bank for Reconstruction and Development (EBRD) average transition scores. A number of related studies have conducted experiments in both single- (cf. Bahry and Wilson, 2006; Chen and Tang, 2009; Slonim and Roth, 1998) and comparative transition country settings (cf. Gächter et al., 2004; Herrmann et al., 2008), but without transition (or market integration) being the focal issue. Conversely, the few experimental studies that explicitly focus on transition (cf. Duch and Palmer, 2004; Ockenfels and Weimann, 1999) are conducted in single settings. There is thus a gap in the experimental literature examining the relationship between transition stage and pro-social behavior. In view of the documented role of such behavior in structuring exchange relations (Kahneman et al., 1986), the growing number of exchange transactions conducted with firms from transition countries (Peng, 2003), and the plethora of economies on the transition path (EBRD, 2009), this is clearly a non-trivial issue. The remainder of this paper comprises five sections. Section 2 reviews the prior literature and develops two propositions. Section 3 describes the experimental design and procedures. Section 4 presents the experimental results and discusses their implications, while Section 5 concludes.

2. Market economies and pro-social behavior Social norms represent common understandings as to what constitutes appropriate behavior among a significant proportion of a particular social group (Elster, 1989). They induce behavioral conformity, even at personal cost (Azar, 2004), and are enforced by extra-legal sanctions which may be both internal (embarrassment, guilt, and so forth) and external (such as disapproval, ostracism, and such like) (Posner, 2001). Social norms evolve through the mechanism of internalization – the greater number of community members who internalize the norm, the more other members will conform, given a conditional preference for conformity (Bicchieri, 2006). The underlying norms in a given society impact behavior

65

(as evidence, for example, during experiments) either when an individual’s attention is drawn to them or when they observe others behaving consistently in accordance with that particular norm (Krupka and Weber, 2009). Various scholars cite evidence of the role of social norms of fairness, reciprocity, and honesty as countervailing forces to selfinterest motives (Kahneman et al., 1986). For example, Azar (2004) finds evidence of fair behavior as people appear to tip in the same manner for both local restaurants (presumably to encourage good service next time) and when taking taxis (when there is little likelihood of repeat dealings). In experimental settings, offers in the ultimatum game are typically in the region of 40–50% of the stake and responders routinely reject offers lower than 20% (Camerer, 2003), and proposers in dictator game experiments customarily make positive offers even though the receiver has no opportunity to reject the proposed split (Forsythe et al., 1994). Similarly, several scholars attribute evidence of compliance in tax reporting (Wenzel, 2004), handing in full wallets to police stations (West, 2005), revealing invoicing discrepancies in their favor to suppliers (Frank et al., 1993), and returning fully-addressed and pre-stamped “lost letters” containing money to their rightful owners (Zsolnai, 2003), as manifestations of honesty norms. On the basis of the foregoing, the pertinent question becomes: Is pro-social behavior likely to be stronger or weaker in more market-oriented economies? There are intuitive arguments to support each contention. On the one hand, a number of scholars aver that norms of fairness/reciprocation and honesty are likely to be stronger in more market-oriented economies with more developed formal institutions for three main reasons. First, whereas earlierstage transition countries are characterized by exchange typically conducted within internally homogeneous, highly cohesive clan and kinship networks originally designed to circumvent planning restrictions (Platteau, 2000), exchange in more market-oriented later-stage transition countries is more ephemeral, anonymous, and conducted over a larger geographic and temporal space (Kandori, 1992). Accordingly, as market economies dilute the role of community sanctioning enforcement mechanisms, individuals need to establish and maintain a reputation for fair and honest dealings with strangers in order to achieve gains from trade in vast markets (Henrich et al., 2010). Second, by replacing a planned economy characterized by endemic corruption, where advantages were arbitrarily provided to some and not others (Kornai, 1986) with a merit-oriented market economy (where attributes such as performance are, in the main, commensurated accordingly), individuals are likely to be imbued with a greater sense of fairness (Pejovich, 2003). Third, the development of market-supporting formal institutions increases the efficiency of legal ordering mechanisms (enhancing observability and verifiability), reducing the pay-offs for deviant behavior, and thus encouraging fair and honest behavior to strangers (North, 1990). Indeed, in one of the few studies to touch upon pro-social behavior and transition, Ockenfels and Weimann (1999) find that West Germans exhibit significantly more cooperative behavior than their Eastern counterparts. On the basis of the foregoing discussion, we propose that: Proposition 1a. Pro-social behavior is likely to be stronger in more market-oriented economies. On the other hand, several scholars posit that norms of fairness/reciprocity and honesty are likely to be weaker in more market-oriented economies on three accounts. First, noncooperative game theory would suggest that market economy conditions, characterized by greater impersonality, ephemeral contracts and ease of entry and exit (Bowles, 1998), attenuate monitoring problems and mitigate the sanctioning role of clan/kinship networks based on repeated interactions (Axelrod,

66

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71

1984). This renders it increasingly “attractive for some players to deceive and defect and take the relatively low risk of being caught” (Ostrom, 2000, p. 148). Accordingly, it becomes harder to sustain fair and honest practices. This may explain Gurven’s (2004) finding that villagers located in closer proximity to the main market town in the Bolivian Amazon make significantly lower offers in the ultimatum game. He argues that “the increased anonymity of larger, market-based societies could. . . make opportunistic defection a common salient social norm” (p. 19). In a similar vein, Schotter et al. (1996) examine two treatments of the ultimatum game – with and without survival pressure, and conclude that “the meaning of equity or fairness is altered once agents face the types of survival dangers present in market-like institutions” (p. 52). Second, whereas state socialism was characterized by a general lack of private property and absence of pecuniary incentives, fullyfunctioning market economies are highly monetized and oriented more toward extrinsic motivation (Bowles, 1998). Accordingly, the more egalitarian notions and concomitant concerns for the collective in early-stage transition economies are likely to yield to self-interest regards, pecuniary rewards and wealth maximization in later-stage transition societies (Duch and Palmer, 2004; Nee and Matthews, 1996). Third, by enhancing the efficacy of legal ordering mechanisms, the concurrent development of formal institutions reduces the need for individuals to rely on social norms to ensure fair dealings (North, 1990). We therefore propose that: Proposition 1b. Pro-social behavior is likely to be weaker in more market-oriented economies. 3. Experimental design and procedures 3.1. Experimental design The experiments were conducted in 2010 with students at the Kazakh-American University in Almaty, Kazakhstan and the Westminster International University in Tashkent, Uzbekistan. A unique feature of our research design is that we purposively select two settings where differences in cultural background, subject pool composition, and experimental procedures can be constrained as much as practicable (Gächter and Herrmann, 2009; Roth et al., 1991), yet where the two countries exhibit wide variations in the rate of progress to fully-functioning market economies (EBRD, 2009). Each of these methodological considerations will now be discussed in more detail. First, the literature highlights that cultural background may be an important predictor of experimental behavior (Gächter et al., 2010; Henrich et al., 2005, 2010). Thus, it becomes imperative to minimise cross-cultural differences to ensure comparable participant pools.2 In this regard, the neighboring Central Asian republics of Kazakhstan and Uzbekistan were purposefully selected on the basis of the high degree of cultural homogeneity. The two countries share a common history and legacy of colonialism in terms of being undivided prior to being the subject of Tsarist domination for over fifty years before being under Soviet subjugation for the following seventy years. Both countries achieved independence in 1991 and officially embarked on the transition path to a market economy in December 1991. Moreover, both countries share a common predominant religion (Islam) and common ethno-linguistic group (Turkic) (Pomfret, 1995).3 Moreover, the

2 The extensive discussion as to constituent elements of national culture is beyond the scope of this paper. Instead, we follow the work of Usunier and Lee (2009) and conceive of national culture as comprising the following elements: nationality/historical background, religion, ethnic group, and language. 3 The country of origin of the subjects may be seen to provide a satisfactory reflection of the overall distribution in their respective countries. In Uzbekistan, 35 of the

two countries exhibited strong similarities in initial conditions. Both countries are landlocked (impeding their integration into the global economy) and both inherited an economic structure which was heavily biased toward agriculture and extractive industries (Zeitler, 2005). Furthermore, the two countries’ social indicators, such as life expectancy at birth and secondary school enrolment at the time of transition, were “almost identical” (Alam and Banerji, 2000, p. 13).4 Second, several steps were taken to constrain variance in subject pool composition. As illustrated in Table 2, participating subjects in both countries were overwhelmingly undergraduate students predominantly studying economics or business, and thereby constitute a relatively homogeneous group in terms of age, education and socio-economic status in their respective societies (Gächter et al., 2010; Güth et al., 1982). Furthermore, the two universities are both located in by far the biggest city in their respective countries, assuaging potential respondent bias that one setting was more Western-leaning or more urban (and by implication, more predisposed to market-type norms) than the other (Bahry and Wilson, 2006). Third, to ensure equivalence of experimental procedures, we followed the standard practices regarding the design of experiments in multiple settings advanced by Roth et al. (1991). To ensure linguistic equivalence (and that the students possessed sufficient English skills to follow the experiments), we selected two institutions which, being affiliated to either a British or American university, use English as the medium of instruction. In addition, purchasing power parity indices were used to ensure that the stake money was equally valued by participants in both settings. Furthermore, to ensure procedural consistency, the first author held pre-experiment meetings to clarify roles, procedures and standard instructions, discuss any potential cultural or linguistic features which may lead to possible misinterpretation by the subjects. Additionally, the first author was present throughout the experiments in both settings.

40 subjects (i.e. 87%) were Uzbek and three (i.e. 7%) were Russian, with the remaining two participants (6%) coming from other ethnic backgrounds. This compares favorably with the overall distribution of Uzbek (80%), Russian (5.5%) and others (14.5%) according to the CIA World Factbook (2010). In Kazakhstan, 31 of the 40 subjects (i.e. 77%) were Kazakh, six (i.e. 15%) were Russian, with the remaining four (i.e. 8%) coming from other ethnic backgrounds. Again, this may be seen to be similar to the overall distribution of Kazakh (53%), Russian (30%) and others constituting 17% as detailed by the CIA World Factbook (2010). Accordingly, the sample in the experiment may be seen to be generally representative of the respective population compositions. 4 Survey data from various datasets as General Social Survey (GSS), World Values Survey (WVS) and Asian Barometer (AB) may be utilized to provide supplementary evidence of cultural similarity. All three above datasets ask the question: “Generally, do you think that people can be trusted or do you think that you can’t be too careful in dealing with people”. The WVS Cultural Map of the World for data collected from 2005 to 2008 suggest that people respond to the above question in a highly culturally homogeneous way. The scores show a high level of intra-cultural homogeneity and inter-group heterogeneity. The scores for “Latin America” include Peru (6.3%), Brazil (9.4%), Chile (12.6%), Mexico (15.6%) and Argentina (17.6%). These contrast with the scores for such “Catholic Europe” countries as France (18.8%), Poland (19.0%) and Spain (20.0%), and the “Orthodox” countries category which includes Bulgaria (22.2%), Russia (26.2%), and Ukraine (27.5%). Similarly, the scores for such “Englishspeaking” countries as the United States, Canada, Australia, and New Zealand all fall within the range of 39.3%–51.2%, whereas the scores for the category of “Protestant Europe” include Finland (58.9%), Sweden (68.0%), and Norway (74.2%). While neither GSS nor WVS datasets contain Kazakhstan and Uzbekistan, AB provides the responses of 800 people in each of 14 Asian nations (including the two focal countries) for 2005. The results show that Kazakhstan exhibited the lowest percentage of respondents who agreed that people can be trusted (16.1%). With the exception of Kyrgyzstan (18.4%), the country with the closest response to the Kazakh participants was Uzbekistan (22.0%). These data provide further support for the present contention that Kazakhstan and Uzbekistan exhibit a relatively strong degree of cultural homogeneity. We would like to thank an anonymous reviewer for bringing this point to our attention.

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71

Despite the above commonalities, one marked difference between the two countries is their respective success in progressing to the status of fully-functioning market economies, as typically assessed in the transition literature by reference to the EBRD average transition scores (Zeitler, 2005).5 The EBRD average transition scores measure legislative and regulatory reforms that foster institutional development. In 2009, the average transition scores for Kazakhstan and Uzbekistan were 2.96 and 2.15, respectively (EBRD, 2009). The widely divergent transition paths have been extensively documented in the literature. While Kazakhstan opted for big bang reforms in accordance with the Washington Consensus strategies of stabilisation, liberalization, and privatization, Uzbekistan chose a more gradual path with slow sequencing of reforms (Alam and Banerji, 2000; Pomfret, 1995; Zeitler, 2005).6 Accordingly, we propose that observed behavioral variation may be ascribed to differences in the degree of market orientation, rather than to differences in cultural background, initial conditions, subject pool composition, or experimental procedures. 3.2. Experimental procedures In terms of recruitment, students were asked to participate in an “economics experiment” by recruiters in both undergraduate classes and on posters, and a show-up sum equivalent to USD 5 was promised to each subject who signed up, arrived on time, and completed the experiment. Monetary rewards were emphasized such that students were also advised that they may earn in excess of the show-up fee at the end of the experiment depending on their decisions. Consistent with the research objective of investigating normative behavior toward strangers, we employed a double-blind procedure (Hoffman et al., 1994) in which subjects were guaranteed anonymity with respect to both fellow students and experimenters. Recruiters were informed not to disclose more information to the subjects prior to the experiment. No subject had ever participated in an experiment before, and only one subject in each country had some prior knowledge of game theory. The age range of the participants in Kazakhstan was from 16 to 22 years of age, and between 18 and 24 in Uzbekistan. Both mean ages and standard deviations of both subject populations were similar. Likewise, the proportion of male to female subjects is consistent with comparable studies (cf. Gil-White, 2004). The demographic characteristics of the participants are illustrated in Table 1. The experiment consisted of three separate phases. In the first, pre-game phase, the students were randomly assigned an ID code by a foreign researcher. Participants were told that these ID codes would be used for matching their decisions with the decisions of others, while preserving complete anonymity. Subjects then completed a 1-page pre-game hard copy questionnaire on their demographic characteristics. The subjects were subsequently randomly assigned7 to two separate rooms connected by a hallway,

5 This is an unweighted average of eight reform indices, measuring the extent of the implementation of liberalization and privatization policies. The average scores range from 1 to 4+, where 1 represents no change from a rigid centrally planned economy, 2 indicates some progress, 3 indicates substantial, comprehensive progress, and 4+ represents the standards of a fully-functioning market economy. 6 Further evidence of significant variation in formal institutional frameworks is provided by the Index of Economic Freedom published by the Heritage Foundation (2010) which measures formal institutional development by examining liberty and economic freedoms in such spheres as business, trade, investment and property rights. Here, Uzbekistan scores 47.5 and is classified as “repressed”, whereas Kazakhstan with 61.0 points is categorized as “moderately free”. 7 This random entitlement (where subjects drew chips out of a hat) can be contrasted with contest treatments, where the best scoring subjects in general knowledge tests are assigned the role of proposers in the ultimatum game. This approach was rejected on the grounds that the latter approach can produced biased

67

where they were seated separately and prevented from communicating. A native experimenter in each room then read from a standard script in English, and demonstrated the tasks for the ultimatum game with the relevant decision cards, vectors, and envelopes. To ensure complete understanding of the rules and procedures, each student received written instructions and had the opportunity to field procedural questions in their own language (Roth et al., 1991).8 In each country, the money was physically placed on a table before the subjects to reinforce to pecuniary nature of the experiment. The second phase consisted of the ultimatum game experiment. Consistent with a number of leading studies in the ultimatum game (cf. Cameron, 1999; Chen and Tang, 2009; Gil-White, 2004), the sample size was 20 bargaining pairs at each university. The experiment was run by hand. No examples of allocation decisions were provided at any time during the instructions so as not to bias the subjects in any way. To ensure that the stake represented a relatively large opportunity costs for other-regarding behavior, each bargaining pair played for a sum of USD 60 in Kazakhstan and USD 16 in Uzbekistan. This stake size, constituting two days’ average income in each of the two countries (Asian Development Bank, 2009), represents double the standard amount in ultimatum bargaining games conducted worldwide (Camerer, 2003; Gowdy et al., 2003). Both proposers and respondents were told that they would only play this game once to preclude reward or punishment considerations (Gächter and Herrmann, 2009). After being allowed three minutes to consider their decision, the proposers wrote their offers underneath their ID codes and placed their decision card in an unmarked envelope, which was then sealed, collected by an experimenter and randomly distributed to the responders in the adjacent room. The responders were then given an equal length of time to decide upon their course of action – crossing either the “accept” or “reject” boxes, before placing their response in the envelope, sealing it, and returning it to the experimenter. After anonymously completing a hard-copy, post-experiment survey, subjects were individually escorted to an adjacent room occupied by the cashier (the foreign researcher). The final stage comprised the revelation game, a novel feature of this research design. In the confines of this private room, the foreign researcher provided a sealed envelope with corresponding ID number to each subject. Only one subject was present in the payment room at any time. The rooms were specifically chosen to ensure consistency in both countries such that the distance between the two individuals was equivalent in both settings. The remaining subjects were seated in a holding room under the supervision of a member of the experimental team and instructed to refrain from communicating with others. In the private room, the subject was correctly informed about the payoff from the first experiment. However, the actual monetary amount contained in each envelope exceeded the stated payoff on the enclosed pay-out form by the equivalent of USD 5. The subjects were then given two instructions. First, they were requested to take the envelope to an adjacent desk where the stool was directed away from the foreign researcher and count the money in private. This allowed the latter to unobtrusively observe the counting process, without the former suspecting that he/she was being monitored. Second, if they were satisfied that the cash amount in the envelope corresponded with the stated winnings on the accompanying pay-out form, they were

results (for example, responders have been known to accept stingy offers because they perceive the proposers to be more worthy – Hoffman et al., 1994). 8 Pilot studies were conducted with undergraduate students at the University of Southern Denmark in autumn 2009, facilitating the clarification of instructional ambiguities and rectification of procedural problems prior to adoption in the main study. All experimental materials are available upon request.

68

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71

Table 1 Demographic characteristics of subjects by national group and experimental role. Kazakhstan

Uzbekistan

Proposers

Respondents

Proposers

Respondents

Country of origin (number of each)

Kazakh (15) Russian (3) Other (2)

Kazakh (15) Russian (3) Other (2)

Uzbek (18) Russian (2)

Uzbek (17) Russian (1) Other (2)

Mean age in years (sd)

18.60 (1.46)

19.10 (1.21)

19.45 (1.54)

20.50 (1.10)

Number of men/women

11/9

9/11

15/5

14/6

Number of undergraduates/postgraduates

20/0

16/4

20/0

18/2

Major

Economics (9) International relations (8) Others (3)

Economics (6) International relations (6) Others (8)

Business administration (13) Economics (4) Others (3)

Business administration (9) Economics (8) Others (3)

Observations: ultimatum game

20

20

20

20

Observations: revelation game

40

40

Table 2 Descriptive statistics – ultimatum game.

Mean offer (%) Standard deviation Mode (%) Minimum offer (%) Maximum offer (%) N (proposers/responders) Rejection rate (%)

Kazakhstan

Uzbekistan

46.25 9.81 0.50 (32.5%) 12.50 62.50 20/20 0.00

55.00 11.20 0.50 (15.0%) 41.67 95.83 20/20 10.00

requested to sign a declaration to that effect.9 Upon doing so, the subject was released from the payment room and escorted individually out of the building by an experimenter, before the next subject was admitted from the holding room. 4. Results and discussion 4.1. Ultimatum game Table 2 summarizes proposer and responder behavior for both countries, while Fig. 1 reports offer distributions in each country. Compared to ultimatum game experiments run elsewhere, there are several unique features in our data. In terms of proposer behavior, the range of offers in Kazakhstan was between 12.5% and 62.5% with a standard deviation of a little under 10, whereas the offer range in Uzbekistan was between 41.67% and 95.83% with a standard deviation of about 11. In Kazakhstan, the modal offer (50%) was offered 32.5% of the time, whereas Uzbek proposers offered the equal split in 15% of cases. In Kazakhstan, proposers offered, on average, 46.25% of the stake. This figure is consistent with offer sizes exceeding 40% reported in other transition countries like Romania (Munier and Zaharia, 2003), Russia (Bahry and Wilson, 2006), and Slovakia (Slonim and Roth, 1998), and in the upper bracket of offer sizes in the Oosterbeek et al.’s (2004) metaanalysis of 37 ultimatum games in 27 countries. In contrast, the mean offer size in Uzbekistan was 55%. This figure exceeds the

9 At this juncture, it is important to raise the issue of deception, which is conventionally proscribed by experimental economists concerned about contaminating future subject pools (Hertwig and Ortmann, 2001). Consistent with the distinctions in the experimental literature (cf. Hertwig and Ortmann, 2001; Hey, 1998; McDaniel and Starmer, 1998), conveying false information to experimental subjects is deceptive, whereas withholding certain information from subjects is not. The revelation game is clearly an example of the latter. Note also that the subject was not deceived since the information provided by the experimenter was factually correct. We would like to thank an anonymous reviewer for bringing this important point to our attention.

proportional offer levels of all studies reported in the meta-study by Oosterbeek et al. (2004) (including the hitherto highest offer of 51% for Paraguay reported in Henrich et al. (2001)) and is higher than the average offer of 48.49% for Malaysian subjects reported in Chuah et al. (2007). By comparison, the modal offer (50%) in fullyfunctioning market economies is 40–50% and the average offer lies between 30 and 40% of the total (Camerer, 2003). We follow prior ultimatum game experiments (cf. Cameron, 1999; Roth et al., 1991) by adopting a non-parametric approach to examine differences in average offer levels using a Mann–Whitney U-test. The test statistic attests that the differences in the mean proportional offers between the two subject pools are statistically significant (p = 0.032). An examination of the ultimatum game offers revealed one outlier; one person offered more than 95% in Uzbekistan. Although this represents an extreme outlier, prior studies report similar behavior with examples of offers exceeding 80% in Indonesia (Cameron, 1999) and 90% in Nigeria (Gowdy et al., 2003). Furthermore, to check whether our results are driven by the extreme outlier, we subsequently examined the results without that particularly high offer, and found that the results remain stable. The mean offer in Uzbekistan drops to 53%, still well-beyond the offer level in Kazakhstan, while the differences in mean proportional offers between the two subject pools remain significant at the 5% level.10 Turning to responder behavior, the number of rejections is relatively low. In Kazakhstan, none of the offers were rejected by the respondents. While not entirely common in ultimatum game experiments, zero rejection rates have been evidenced in a number of studies in non-western societies (Gowdy et al., 2003; Henrich et al., 2001). In comparison, two offers in Uzbekistan (of 55% and 58%) were rejected. This constitutes a rejection rate of 10%, which is well within the range established by prior studies utilizing

10 Our ultimatum game utilizes the sequential “game” method as described in the foregoing. One shortcoming of this method concerns the difficulty in disentangling a proposer’s preference for fairness and his/her concern that the offer may not be sufficiently high to be accepted. To address this issue, various scholars (cf. Henrich et al., 2006) have proposed the “strategy” method as an alternative, whereby respondents state contingent choices (accept, reject) for different offer levels, which are then binding. However, while this approach may elucidate more nuanced respondent behavior, it is not without methodological problems. For example, Oxoby and McLeish (2004) note that respondents are required to reflect and consider responses to a range of different offers in a “cold” environment, such that their responses are less likely to be influenced by positive and negative emotions, which are important aspects in normative behavior. Furthermore, previous studies report no significant difference in behavior between the two methods in trust games (Solnick, 2007), prisoners’ dilemma games (Brandts and Charness, 2000), or indeed, ultimatum games (Oxoby and McLeish, 2004). On the basis of the above, the game method was invoked in the present experiment.

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71

69

Fig. 1. Ultimatum game behavior in Kazakhstan and Uzbekistan.

similar experimental designs (cf. Güth et al., 1982; Oosterbeek et al., 2004; Shupp et al., 2006) and also resembles findings in other transition countries such as a mean rejection rate of approximately 12% in Slovakia (Slonim and Roth, 1998) and of 10% in Mongolia (Gil-White, 2004). Interestingly, both of the two offers rejected exceeded 50%. Although this behavior may, prima facie, appear counterintuitive, other studies conducted in transition countries show similar results. Bahry and Wilson (2006) for instance found that Russian responders rejected “hyper-fair offers” (more than half the stake) 56% of the time, suggesting that the legacy of Soviet egalitarianism may invoke a “very distinct concept of fairness in which hyper-fair offers are unacceptable” (Bahry and Wilson, 2006, p. 46). Subsequent regression analyses with selected demographic variables show that offer levels in the ultimatum game cannot be explained by the subjects’ gender (p = 0.150), their level of education, i.e. whether they were undergraduates or postgraduates (p = 0.482), or their choice of major (0.156). 4.2. Revelation game One of the distinguishing features of our research design is the introduction of the revelation game as a novel experiment for studying the truthful disclosure of information toward strangers. As reported in Fig. 2, only 7.5% (N = 3) of all Kazakh subjects revealed the windfall, while fully 45% (N = 18) of Uzbek subjects revealed the excessive payout and duly returned the money to the foreign researcher. The figure for Uzbekistan is quite remarkable given the excess amount constituted over half a day’s national average wage, thus representing a relatively high opportunity cost for behavior deviating from the Nash equilibrium predictions. The finding that over one quarter of all 80 subjects chose to disclose the serendipitous amount to the foreign researcher is slightly lower than the 33% of subjects revealing invoicing discrepancies in the Frank et al. (1993) study, or the 44% of subjects returning lost letters as reported by Zsolnai (2003). However, these differences are eminently plausible in the light of the fact that the former study used hypothetical, rather than real, money, and that the

Fig. 2. Revelation game behavior in Uzbekistan and Kazakhstan (N = 40 in each country).

70

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71

latter experiment involved overt, as opposed to covert, monitoring opportunities. Nonparametric Mann–Whitney U-tests show that the difference in revelation behavior between the two groups is highly significant at the 1% level (p = 0.014). Subsequent Probit regressions controlling for possible demographic effects revealed that differences in gender (p = 0.190), education level (p = 0.176) or choice of major (p = 0.621) were found to be insignificant in explaining variations in revelation behavior between the two subject pools. Regression analysis further showed that behavior in one game is not explained by behavior in the other. As a result, we conclude that subjects exhibit differences in revelation behavior not attributable to experimental conditions or demographic characteristics other than country. In sum, we observed evidence of significantly fairer, more honest behavior in dealings with strangers in early-stage Uzbekistan than in later-stage Kazakhstan. Together, these findings lend support to proposition 1b, that pro-social behavior is likely to be weaker in more market-oriented economies. 4.3. Discussion The above findings appear to contradict those of the few tangential studies examining social behavior and transition to market economies. For example, when playing the public good and solidarity games a number of years after reunification, Ockenfels and Weimann (1999) found that individuals from Eastern Germany exhibited markedly less pro-social behavior than their Western counterparts. Furthermore, the above two scholars and their colleagues replicated the solidarity experiment 20 years after reunification and found that systematic differences persisted in solidarity behavior (Brosig-Koch et al., 2011). However, plausible explanations for these differences exist. In the German studies, stake sizes were significantly lower than in the current study, reducing the monetary incentive to act selfishly, particularly if the experimenters were not able to fully account for the real income differences between the two parts. Additionally, their findings may have been confounded by the fact that the experiments were conducted a number of years after the fall of the Berlin Wall, whereas subjects in the present study were all currently in the midst of the transition process. Moreover, they examined different types of norms from the present study, and prior studies have shown that certain types of normative behavior may be more malleable to the transition process than others, despite a common economic, political and ideological legacy (Ardichvili and Gasparishvili, 2003). Additionally, our findings seem to be at odds with various ethnographic studies that find pro-social norms co-vary positively with market integration (cf. Henrich et al., 2005, 2010). However, these contradictory findings may be accounted for by the fact that the current study broadens their concept of market integration to incorporate both vast market economies and formal institutional development. By extending our analysis from rudimentary, small-scale societies to highly monetized market economies, we suggest that the enhanced opportunities for pecuniary gain induce a shift to more selfish behavior (Bowles, 1998; Schotter et al., 1996). Furthermore, we propose that the presence of stronger formal institutions exerts a crowding-out effect on their informal counterparts (Gneezy and Rustichini, 2000; Reeson and Tisdell, 2008). Specifically, by facilitating more efficacious enforcement of better-defined property rights, stronger formal institutions reduce the relative costs of relying on legal ordering mechanisms to regulate behavior (Frey, 1998; North, 1990), thus crowding out the role played by social norms in regulating exchange interactions with strangers (Frey and Oberholzer-Gee, 1997).

5. Conclusion While the role of pro-social behavior in structuring and regulating exchange relations is well-established, the degree to which such behavior prevails in different formal institutional settings is less known. This paper reports on two laboratory experiments examining evidence of pro-social behavior in early-stage Uzbekistan and later-stage Kazakhstan. While carefully controlling for differences in cultural background, participant pool composition, and experimental procedures, we found statistically significant differences between subject pools in both ultimatum and revelation games, suggesting that Uzbeks were fairer and more honest than their Kazakh counterparts. It would thus appear that pro-social behavior becomes weaker in more market-oriented economies. Our findings also have normative implications. While firms may have previously been deterred from entering collaborative arrangements with counterparts domiciled in relatively weak formal institutional settings, the observed prevalence of other-regarding preferences in Uzbekistan may provide encouragement that contractual obligations may, indeed, be upheld. Conversely, in view of the incomplete nature of contracts (Frey, 1998), firms should exercise caution in entering agreements with firms from later-stage transition economies, where reliance on formal institutions may not provide sufficient redress against heightened self-interested motives. Our study elucidated two elements of normative behavior, namely fairness/reciprocity and honesty. In view of the multidimensional nature of pro-social norms, future studies which attempt to elicit other forms of normative behavior (such as cooperation or altruism) in similar settings would further our knowledge of this complex area. Relatedly, in light of the differences in findings between our study into fairness and honesty and those of Ockenfels and Weimann (1999) into cooperation, more research specifically aimed at investigating the different ways in which transition may affect different types of behavior or norms is required. Additionally, in view of the cross-sectional nature of the current study, future studies conducted in other transition countries may enhance the external validity of the findings in this paper. Similarly, and related to the discussion point in the previous section, future research providing longitudinal evidence would similarly enrich our understanding the relationship between formal and informal institutional development. Finally, we wish to emphasize the exploratory nature of the current study which means that our conclusions remain tentative. In view of the attested importance of normative behavior in exchange relations (Kahneman et al., 1986) and the burgeoning number of exchange interactions with firms from transition economies (Peng, 2003), future contributions to this subject area will be both timely and pertinent.

References Alam, A., Banerji, A., 2000. Uzbekistan and Kazakhstan: a tale of two transition paths. World Bank Policy Research Working Paper 2472, November 2000. Ardichvili, A., Gasparishvili, A., 2003. Russian and Georgian entrepreneurs and nonentrepreneurs: a study of value differences. Organization Studies 24, 29–46. Asian Barometer, 2011. (accessed 04.08.11). Asian Development Bank, 2009. Key Indicators for Asia and the Pacific: Highlights. Asian Development Bank, Manila, Philippines. Axelrod, R., 1984. The Evolution of Cooperation. Basic Books, New York, NY. Azar, O.H., 2004. What sustains social norms and how they evolve? The case of tipping. Journal of Economic Behavior and Organization 54, 49–64. Bahry, D.L., Wilson, R.K., 2006. Confusion or fairness in the field? Rejections in the ultimatum game under the strategy method. Journal of Economic Behavior and Organization 60, 37–54. Bicchieri, C., 2006. The Grammar of Society: The Nature and Dynamics of Social Norms. Cambridge University Press, New York, NY. Bowles, S., 1998. Endogenous preferences: the cultural consequences of markets and other economic institutions. Journal of Economic Literature 36, 75–111.

S.M. Rosenbaum et al. / The Journal of Socio-Economics 41 (2012) 64–71 Bowles, S., Gintis, H., 1998. The moral economy of communities: structured populations and the evolution of pro-social norms. Evolution and Human Behavior 19, 3–25. Brandts, J., Charness, G., 2000. Hot vs. cold: sequential responses and preference stability in experimental games. Experimental Economics 2, 227–238. Brosig-Koch, J., Helbach, C., Ockenfels, A., Weimann, J. Still different after all these years: solidarity behavior in East and West Germany. Journal of Public Economics, in press. Camerer, C.F., 2003. Behavioral Game Theory: Experiments in Strategic Interaction. Princeton University Press, Princeton, NJ. Cameron, L.A., 1999. Raising the stakes in the ultimatum game: experimental evidence from Indonesia. Economic Inquiry 37, 47–59. Chen, K., Tang, F.-F., 2009. Cultural differences between Tibetans and ethnic Han Chinese in ultimatum bargaining experiments. European Journal of Political Economy 25, 78–84. Chuah, S.-H., Hoffmann, R., Jones, M., Williams, G., 2007. Do cultures clash? Evidence from cross-national ultimatum game experiments. Journal of Economic Behavior and Organization 64, 35–48. CIA World Factbook, 2010. (accessed 14.09.10). Duch, R.M., Palmer, H.D., 2004. It’s not whether you win or lose, but how you play the game: self-interest, social justice, and mass attitudes toward market transition. American Political Science Review 98, 437–452. EBRD, 2009. Transition Report 2009. European Bank for Reconstruction and Development, London, UK. Elster, J., 1989. Social norms and economic theory. Journal of Economic Perspectives 3, 99–117. Forsythe, R., Horowitz, J.L., Savin, N.E., Sefton, M., 1994. Fairness in simple bargaining experiments. Games and Economic Behavior 6, 347–369. Frank, R.H, Gilovich, T., Regan, D.T., 1993. Does studying economics inhibit cooperation? Journal of Economic Perspectives 7, 159–171. Frey, B.S., 1998. Institutions and morale: the crowding-out effect. In: Ben-Ner, A, Putterman, L.G. (Eds.), Economics, Values and Organization. Cambridge University Press, Cambridge. Frey, B.S., Oberholzer-Gee, F., 1997. The cost of price incentives: an empirical analysis of motivation crowding-out. American Economic Review 87, 746–755. Gächter, S., Herrmann, B., Thöni, C., 2004. Trust, voluntary cooperation, and socioeconomic background: survey and experimental evidence. Journal of Economic Behavior and Organization 55, 505–531. Gächter, S., Herrmann, B., 2009. Reciprocity, culture and human cooperation: previous insights and a new cross-cultural experiment. Philosophical Transactions of the Royal Society 364, 791–806. Gächter, S., Herrmann, B., Thöni, C., 2010. Culture and cooperation. Philosophical Transactions of the Royal Society 365, 2651–2661. Gil-White, F.J., 2004. Ultimatum game with an ethnicity manipulation: results from Khovdiin Bulgan Sum Mongolia. In: Henrich, J.P., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H. (Eds.), Foundations of Human Sociality: Economic Experiments and Ethnographic Evidence from Fifteen Small-Scale Societies. Oxford University Press, Oxford. Gneezy, U., Rustichini, A., 2000. A fine is a price. Journal of Legal Studies 29, 1–17. Gowdy, J., Iorgulescu, R., Onyeiwu, S., 2003. Fairness and retaliation in a rural Nigerian village. Journal of Economic Behavior and Organization 52, 469–479. Gurven, M., 2004. Economic games among the Amazonian Tsimane: exploring the roles of market access, costs of giving, and cooperation on pro-social game behavior. Experimental Economics 7, 5–24. Güth, W., Schmittberger, R., Schwarze, B., 1982. An experimental analysis of ultimatum bargaining. Journal of Economic Behavior and Organization 3, 367–388. Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H., McElreath, R., 2001. In search of Homo Economicus: behavioral experiments in 15 small-scale societies. American Economic Review 91, 73–78. Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H., McElreath, R., Alvard, M., Barr, A., Ensminger, J., Smith Henrich, N., Hill, K., Gil-White, F., Gurven, M., Marlowe, F.W., Patton, J.Q., Tracer, D., 2005. Economic man in cross-cultural perspective: behavioral experiments in 15 small-scale societies. Behavioral and Brain Sciences 28, 795–855. Henrich, J., McElreath, R., Barr, A., Ensminger, J., Barrett, C., Bolyanatz, A., Cardenas, J.C., Gurven, M., Gwako, E., Lesorogol, C., Marlowe, F., Tracer, D., Ziker, J., 2006. Costly punishment across human societies. Science 312, 1767–1770. Henrich, J., Ensminger, J., McElreath, Barr, A., Barrett, C., Bolyanatz, A., Cardenas, J.C., Gurven, M., Gwako, E., Henrich, N., Lesorogol, C., Marlowe, F., Tracer, D., Ziker, J., 2010. Markets, religion community size, and the evolution of fairness and punishment. Science 327, 1480–1484. Heritage Foundation, 2010. (accessed 07.10.2010).

71

Herrmann, B., Thöni, C., Gächter, S., 2008. Antisocial punishment across societies. Science 319, 1362–1367. Hertwig, R., Ortmann, A., 2001. Experimental practices in economics: a methodological challenge for psychologists. Behavioral and Brain Sciences 24, 383–451. Hey, J.D, 1998. Experimental economics and deception: a comment. Journal of Economic Psychology 19, 397–401. Hoffman, E., McCabe, K., Shachat, K., Smith, V.L., 1994. Preferences, property rights, and anonymity in bargaining games. Games and Economic Behavior 7, 346–380. Kahneman, D., Knetsch, J.L., Thaler, R., 1986. Fairness as a constraint on profit seeking: entitlements in the market. American Economic Review 76, 728–741. Kandori, M., 1992. Social norms and community enforcement. Review of Economic Studies 59, 63–80. Kornai, J., 1986. The Hungarian reform process: visions, hopes, and reality. Journal of Economic Literature 24, 1687–1737. Krupka, E.L, Weber, R.A., 2009. The focusing and informational effects of norms on pro-social behavior. Journal of Economic Psychology 30, 307–320. McDaniel, T., Starmer, C., 1998. Experimental economics and deception: a comment. Journal of Economic Psychology 19, 403–409. Munier, B., Zaharia, C., 2003. High stakes and acceptance behavior in ultimatum bargaining. Theory and Decision 53, 187–207. Nee, V., Matthews, R., 1996. Market transition and societal transformation in reforming state socialism. Annual Review of Sociology 22, 401–435. North, D.C., 1990. Institutions, Institutional Change and Economic Performance. Cambridge University Press, Cambridge. Ockenfels, A., Weimann, J., 1999. Types and patterns: an experimental East-WestGerman comparison of cooperation and solidarity. Journal of Public Economics 71, 275–287. Oosterbeek, H., Sloof, R., Van de Kuilen, G., 2004. Cultural differences in ultimatum game experiments: evidence from a meta-analysis. Experimental Economics 7, 171–188. Ostrom, E., 2000. Collective action and the evolution of social norms. Journal of Economic Perspectives 14, 137–158. Oxoby, R.J., McLeish, K.N., 2004. Sequential decision and strategy vector methods in ultimatum bargaining: evidence on the strength of other-regarding behavior. Economics Letters 84, 399–405. Pejovich, S., 2003. Understanding the transaction costs of transition: it’s the culture, stupid. Review of Austrian Economics 16, 347–361. Peng, M.W., 2003. Institutional transitions and strategic choices. Academy of Management Review 28, 275–296. Platteau, J.-P., 2000. Institutions, Social Norms and Economic Development. Harwood Academic Publishers, Amsterdam, NL. Pomfret, R., 1995. The Economies of Central Asia. Princeton University Press, Princeton. Posner, R.A., 2001. Social norms and the law: an economic approach. American Economic Review 87, 365–369. Reeson, A.F., Tisdell, J.G., 2008. Institutions, motivations and public goods: an experimental test of motivational crowding. Journal of Economic Behavior and Organization 68, 273–281. Roth, A.E, Prasnikar, V., Okuno-Fujiwara, M., Zamir, S., 1991. Bargaining and market behavior in Jerusalem, Ljubljana, Pittsburg, and Tokyo: an experimental study. American Economic Review 81, 1068–1095. Rubinstein, A., 1982. Perfect equilibrium in a bargaining model. Econometrica 50, 97–109. Schotter, A., Weiss, A., Zapater, I., 1996. Fairness and survival in ultimatum and dictatorship games. Journal of Economic Behavior and Organization 31, 37–56. Shupp, R., Schmitt, P., Swope, K., 2006. On the role of the hostage in ultimatum bargaining games. Journal of Socio-Economics 35, 399–411. Slonim, R., Roth, A.E., 1998. Learning in high stakes ultimatum games: an experiment in the Slovak Republic. Econometrica 66, 569–596. Solnick, S.J., 2007. Cash and alternative methods of accounting in an experimental game. Journal of Economic Behavior and Organization 62, 316–321. Usunier, J.-C., Lee, J.A., 2009. Marketing Across Cultures. Pearson Education, Harlow, UK. Wenzel, M., 2004. The social side of sanctions: personal and social norms as moderators of deterrence. Law and Human Behavior 28, 547–567. West, M.D, 2005. Law in Everyday Japan: Sex, Sumo, Suicide and Statutes. University of Chicago Press, Chicago, IL. Zeitler, A., 2005. Between Russia and China—Central Asia’s transition experience. Osteuopa-Institut München Working Papers No. 258, December 2005. Zsolnai, L., 2003. Honesty versus cooperation: a reinterpretation of the moral behavior of economics students. American Journal of Economics and Sociology 62, 707–712.