Economic Systems And Risk Preferences: Evidence From East And West GermanyVW Staff
Economic Systems And Risk Preferences: Evidence From East And West Germany
Technische Universität Darmstadt
January 14, 2016
For standard economic models it is typically assumed that preferences are given and stable. But do economic systems shape individuals' risk preferences? Using the reunification of East and West Germany as a natural experiment I evaluate differences in financial risk taking comparing Eastern and Western German households for almost two decades after the fall of the Berlin Wall. Controlling for a large set of socio-economic variables East Germans having been “treated” by a command economy were more prone to taking financial risk than West German citizens. The differences were quantitatively relevant after the fall of the Iron Curtain and almost vanished by 2008.
Economic Systems And Risk Preferences: Evidence From East And West Germany – Introduction
Standard economic models assume that preferences are given and stable. Exchange of endowments through markets and a price mechanisms does not alter the valuation of goods. However, it is increasingly questioned that preferences actually are independent from social and societal influences (Fehr and Hoff, 2011). In particular, may it be the case that people have different risk preferences depending on whether they interact in a market or in a command economy?
Peoples' attitudes towards risky behavior is important. It substantially determines economic decision making in many domains. Typically, savings decisions have to be made under uncertainty so that risk attitudes may not only affect individuals' wealth accumulation. On a larger scale, more risky behavior by a populace may also be constitutive for the stability of financial markets. Besides savings decisions there are numerous other domains where uncertainty is a crucial feature: educational investments, occupational choices, or real estate purchases. Moreover, the topic bears considerable policy relevance. Without taking into account preference endogeneity policy evaluations will be biased. If preferences are endogenous it becomes important to distinguish between the changes that policies have on the choice set and the behavioral changes that come about by altered preferences.
A recent literature suggests that culture and the political environment affect people's preferences for redistribution (Corneo, 2004; Alesina and Fuchs- Schündeln, 2007), attitudes toward democracy (Fuchs-Schundeln and Schundeln, 2015), solidarity (Ockenfels and Weimann, 1999), moral standards (Falk and Szech, 2013), conspicuous consumption (Friehe and Mechtel, 2014), or financial risk taking (Osili and Paulson, 2008). It is, however, inherently difficult to meaningfully measure a causal effect of markets on people's preferences because preferences and institutions very likely co-evolve. The after World War II split of Germany constitutes an exceptional possibility to analyze the effect of markets versus a command economy on people's risk preferences. From 1949 to 1990 Germany was divided into a market oriented Federal Republic of Germany (FRG) and a communist German Democratic Republic (GDR) which installed a command economy. Before World War II Germany was a unified and rather homogeneous country. Comparing the risk preferences for East and West German citizens after the former having lived for up to four decades in a command economy might be insightful.
I use the post cold war four waves (1993, 1998, 2003, 2008) of the German income and consumption survey (“Einkommens- und Verbraucherstichprobe”, EVS) in order to investigate whether East German citizens differ from West German Citizens with respect to the riskiness of their wealth positions and savings decisions, controlling for a large set of socio-economic variables including households' incomes and wealth. I conjecture that having been socialized in a market economy as opposed to a command economy might be relevant for shaping peoples' risk preferences.
What determines risk attitudes is not so well understood and astonishingly results vary from study to study quite substantially. Drawing on survey data Donkers et al. (2001), Dohmen et al. (2011), or Dohmen et al. (2012), for example, show that socio-economic variables such as gender, age, height, or parental background have an effect on the willingness to take risks whereas Guiso and Paiella (2008) find that households' attributes are of little help in predicting risk aversion. Black et al. (2015b) analyze the stock market participation of Swedish adoptees and relate it to the investment behavior of their biological and adoptive parents finding that a substantial proportion of risk-attitudes is environmentally determined. Their result, however, goes against the findings in Barnea et al. (2010) and Cesarini et al. (2010) who compare the investment behavior of identical and fraternal twins. Their findings suggest that parental influences play little role beyond the genetic influence. Using a compulsory school reform in Sweden as an exogenous variation in education, Black et al. (2015a) derive evidence for an effect of education on risky behavior which they measure in terms of stock market participation and risky asset holdings. Such an effect seems, however, to only hold for men and not for women.
See full PDF below.