DIRECT EVIDENCE ON RISK ATTITUDES AND MIGRATION David A. Jaeger, Thomas Dohmen, Armin Falk, David Huffman, Uwe Sunde, and Holger Bonin* Abstract���It has long been hypothesized that individuals��� migration pro- pensities depend on their risk attitudes, but the empirical evidence has been limited and indirect. We use newly available data from the German Socio-Economic Panel to measure directly the relationship between mi- gration and risk attitudes. We find that individuals who are more willing to take risks are more likely to migrate. Our estimates are substantial compared to unconditional migration probabilities, as well the effects of conventional determinants of migration, and are robust to controlling for a variety of demographic characteristics. We find no evidence that our results are the result of reverse causality. I. Introduction Gtioning eographic mobility plays an important role in the efficient func- of markets. Moving people and capital to where they can be most productively utilized is essential to any working economy. Because migrants may grease the wheels of the labor market (Borjas, 2001), it is important to understand the determinants of geographic mobility. While it has long been hypothesized that individuals��� will- ingness to take risks may play an important role in migration, and therefore in the efficient functioning of labor markets, there is no direct evidence on whether risk attitudes in fact influence individuals��� migration decisions. This paper seeks to fill this gap in the literature. To motivate why attitudes toward risk taking might be important for the migration decision, consider that individuals derive utility from consumption and leisure. It is quite reasonable to assume that indi- viduals have more information about income, consumption, and leisure opportunities in their current location compared to other potential locations. This relatively greater uncertainty over some or all of the arguments of the utility function is one sense in which migration is a fundamentally risky activity, leading to a tendency for individuals who are more willing to take risks to have a higher probability of migrating. The direction of the relationship between risk attitudes and migra- tion is potentially ambiguous, however. Risk attitudes could also potentially affect migration decisions if risk-averse individuals desire to live in regions with (known) lower variances of the income distribution. More risk-averse individuals might favor locations with lower variances to avoid uncertainty in income, while lower-risk- averse individuals may use migration as a means of improving their chances of receiving a higher-than-average wage.1 In the German context that we examine, we consider the former argument regarding imperfect information to be more applicable. To be sure, there are some differences in income variances across regions in Germany, but they are small compared to those that exist in developing countries or between developing and developed coun- tries.2 We therefore hypothesize that general uncertainty about other locations is the more important channel through which risk attitudes determine intra-German geographic mobility and that the average mover is relatively more willing to take risks.3 To our knowledge, no previous empirical study has examined the relationship between migration and risk attitudes, directly measured.4 We use newly available data from the German Socio-Economic Panel (SOEP) that includes direct measures of attitudes toward risk and find that being more willing to take risks is a positive, statistically significant, and quantitatively important determinant of migration. We find that being relatively willing to take risks is associated with an increase of at least 1.5 percentage points in the probability of ever migrating between 2000 and 2006, even after conditioning on indi- vidual characteristics. This effect is substantial relative to the uncon- ditional migration propensity of 5.8 percent. We also present results indicating that our findings are highly unlikely to be the result of reverse causality���that migration causes changes in risk attitudes. II. Data The SOEP is a representative panel survey of the resident adult population of Germany that began in western Germany in 1984 and was expanded to include eastern Germany in 1990.5 The SOEP surveys the head of each household in the sample, as well as all other household members over the age of seventeen on a wide variety of economic, political, and attitudinal issues. Of most interest to us are the 2004 and 2006 waves of the SOEP, which contain a novel set of questions about individuals��� risk attitudes. Our primary focus is on the question that asked individuals for their attitude toward risk in general, allowing respondents to indicate their willingness to take risks on an eleven-point scale, with 0 indicating complete unwilling- ness to take risks and 10 indicating complete willingness to take risks.6 Our analysis uses responses on the scale as an index of Received for publication February 28, 2007. Revision accepted for publication December 8, 2008. * Jaeger: City University of New York Graduate Center, University of Cologne, and IZA Dohmen: ROA, Maastricht University, IZA, and DIW Falk: University of Bonn, IZA, CEPR, and DIW Huffman: Swarthmore College and IZA Sunde: University of St. Gallen, IZA, CEPR, and DIW Bonin: ZEW and IZA. We thank Daron Acemoglu, Deborah Cobb-Clark, and two anonymous referees, as well as participants of seminars and conferences at the Australian National University, the Berlin Labor Network, the European Association of Labor Economics meetings (Prague), Econometric Society European Meetings (Budapest), ESRI (Dublin), Hebrew University, IZA, Rutgers University, RWI-Essen, Verein fu ��r Socialpolitik, and Virginia Commonwealth University for comments. David Jaeger thanks IZA for support while he worked on this paper. 1 Most of the previous literature on the role of risk attitudes for migration has focused on this sorting argument and on migration as a means of diversification of family income, largely in developing countries. See, for example, Smith (1979), Levhari and Stark (1982), Katz and Stark (1986), Xu (1992), and Daveri and Faini (1999), among others. Heitmueller (2005) posited a model in which risk-averse individuals are less likely to migrate and calibrated the model using actual data, but did not estimate how risk aversion determines migration propensities. 2 Moreover, Gottschalk and Smeeding (1997) showed that Germany has lower income inequality than many other developed countries. 3 We present evidence below that Germans perceive migration as a risky endeavor. 4 In work that postdates ours, Conroy (2007) has presented work that uses a direct measure of risk attitudes in Mexico. Using the Mexican Family Life Survey, he found a positive correlation between risk aversion and migration for young Mexicans. 5 For a detailed description of the SOEP, see Wagner, Burkhauser, and Behringer (1993), and Schupp and Wagner (2002). Additional details can be found at http://www.diw.de/english/sop/ (retrieved February 14, 2007). 6 The exact wording of the question (translated from German) is: ���How do you see yourself? Are you generally a person who is fully prepared to THE REVIEW OF ECONOMICS AND STATISTICS 684 The Review of Economics and Statistics, August 2010, 92(3): 684���689 �� 2010 The President and Fellows of Harvard College and the Massachusetts Institute of Technology

willingness to take risks (which we refer to as the risk index), as well as a binary indicator for whether someone chose a value of 6 or higher on the scale (which we refer to as the risk indicator).7 The latter measure minimizes potential problems from different use of scales by the survey respondents. We view the preferences represented by the responses to these questions as fixed over the period of time in our sample, but also investigate the possibility of reverse causality, such that migration affects risk attitudes, by using the 2006 wave of the SOEP that includes the risk question for a second time. The risk question we use here has been experimentally validated and shown to be a reliable measure of an individual���s actual propen- sity to take risks. Dohmen et al. (2005) used a pool of 450 subjects with characteristics comparable to the respondents of the SOEP who answered the same general risk question from the SOEP questionnaire that is used in this paper. After completing the survey questionnaire, these subjects participated in real-stakes lottery experiments. The responses to the general risk question turned out to be good predictors of actual risk-taking behavior in the paid experiment. Dohmen et al. (2005) also showed that responses to this risk question predict other behaviors involving risk, such as holding stocks, being self-employed, or smoking. Thus, we are confident that the general risk question is a behaviorally valid measure of an individual���s underlying attitude toward risk.8 The smallest geographic unit defined in the publicly available version of the SOEP is the Raumordnungsregion (literally, ���spatial district,��� although we refer to them as regions in the rest of the paper). Germany is divided geographically into 97 such regions, which are defined by the Bundesamt fu ��r Bauwesen und Raumordnung (Federal Office for Building and Regional Planning) and reflect an aggregation of Landkreise and kreisfreie Sta ��dte (administrative districts, some- thing akin to counties in the United States), taking into account economic agglomeration and commuting flows. Each region captures a center of economic activity and its surrounding area and corresponds to a labor market. We define a migration as a move from one region to another.9 For our analysis, we restrict the sample to the 2000 through 2006 waves of the SOEP, including data since the most recent survey refreshment in 2000, in order to have a large balanced panel and also provide a sufficient number of observed migrations. We concentrate on prime-age individuals who were between 18 and 65 years of age during the entire survey period, leaving us with a sample of 10,115 individuals with six years of migration data.10 III. Risk Attitudes of Migrants and Nonmigrants Figure 1 shows the distribution of responses to the general risk question for movers (individuals who changed region at least once between 2000 and 2006) and stayers (individuals who did not change region in that period). While both distributions have a modal value of 5 on the eleven-point scale, the distribution for movers has less weight on the left-hand tail and more weight on the right-hand tail. A greater proportion of movers than stayers clearly responds that they are relatively more amenable to taking risks. In table 1 we present for movers and stayers the average of the risk index, as well as the share of the sample for which the risk indicator is equal to 1, stratified by a variety of demographic characteristics. As reflected in figure 1, the averages of the risk index and risk indicator are substantially larger for the 5.8% of the sample who moved than for those who never moved within the sample period. Moreover, those who moved more than once are more risk friendly than those who moved only once. These results are a first strong indication (albeit not conditional on any individual characteristics) in favor of the hypoth- esis that migrants are less risk averse than nonmigrants. Across nearly all of the demographic categories (sex, age, educa- tion, marital status, and place of origin), we find strikingly consistent take risks or do you try to avoid taking risks? Please tick a box on the scale, where the value 0 means ���not at all willing to take risks��� and the value 10 means ���very willing to take risks.��� ��� German versions of all risk questions are available online at http://www.diw.de/deutsch/sop/service/ fragen/fr2004/personen_2004.pdf (retrieved February 14, 2007). 7 Robustness checks conducted by Dohmen et al. (2005) suggest that choosing a threshold of 6 and above on the eleven-point scale does not affect the behavioral validity of the responses. 8 The SOEP also asked about risk attitudes in other domains such as career and finance. We explore their relationship with migration in Jaeger et al. (2007). 9 We have also estimated the models in the paper using definitions of moves based on distance as well as between German federal states (Bundesla ��nder). The results are all qualitatively similar. When looking only at intraregional moves���those in which individuals changed dwelling but not Raumordnungsregion���we find that risk attitudes are substantially less important determinants. This supports our argument that what is primarily driving the results is uncertainty about other regions. See Jaeger et al. (2007) for these results. 10 This implies that only individuals born between 1940 and 1983 are in our sample. We eliminate from the sample individuals who had missing information on any of the variables used in the analysis, as well as twenty individuals whose information on moving dwelling and moving region was inconsistent. Note that because an individual had to be in our sample in 2004 in order to have answered the risk question, any attrition in our sample that we can relate to risk attitudes occurs only over two years. Nevertheless, of the 106 individuals whom we observed in 2004 but left the sample in 2005 or 2006, the mean of the risk index is 4.76, versus 4.52 for individuals who remained in the sample for the entire period (this difference is not statistically significantly different from 0). The averages of the risk indicator for the samples of attriters and nonattriters are 0.443 and 0.324, respectively, which are statistically significantly different from one another. One should keep in mind, however, that most sample attrition occurs because individuals have moved and cannot be found by the SOEP interviewers. Thus, the attrition bias in our results would tend to be negative (the true relationship between risk attitudes and migration would be more strongly positive than our results indicate), and therefore our estimated effects of risk attitudes on the propensity of migrating should be taken as lower bounds of the true effects. FIGURE 1.���GENERAL RISK ATTITUDES FOR STAYERS AND MOVERS, 2000���2006 Source: Authors��� tabulations from the 2000���2006 waves of the SOEP. The index is an individual���s response to a question in the 2004 waves of the SOEP asking about ���willingness to take risks, in general��� on a scale from 0 to 10, where 0 indicates ���unwilling to take risks��� and 10 indicates ���very willing to take risks.��� Movers are individuals who changed region at least once between 2000 and 2006. Sample size is 10,115. NOTES 685