The Role of Personal Preferences in Economics

Prof. Dr Thomas Dohmen from the Institute for Applied Microeconomics, University of Bonn, Germany is to deliver an honorary lecture at the upcoming April Academic Conference at HSE University April 9 – 12. HSE News Service spoke with Dr. Dohmen about his work on the Global Preferences Survey (GPS), an international survey of epic proportions that he helped develop and analyze in order to learn whether individuals differ in terms of economic preferences by country, and more.

Global Preferences Survey

People's preferences—particularly regarding risk, the timing of rewards, reciprocity, altruism and trust—are important in economics because preferences like these drive people's decisions, and ultimately, economic outcomes

Many empirical studies provide insights into individual-level preferences within certain populations, but much less is known about how these preferences vary worldwide. This is partly because, until recently, no global data set existed that captured these preferences. Together with my co-authors Armin Falk, David Huffman, Uwe Sunde, Anke Becker and Benjamin Enke, we have developed such a data set, the Global Preferences Survey (GPS).

This data set that we describe in our journal article, ‘Global Evidence on Economic Preferences,’ includes survey measures of time preference, risk preference, unconditional altruism, positive and negative reciprocity and trust for 80,000 adults in 76 countries that represent about 90% of the world's population. The survey data was collected through the framework of the 2012 Gallup World Poll, which ensures a standardized survey method for the 80,000 respondents across countries. The data set is described in detail on the GPS homepage, which also offers an interactive tool to compare preferences across countries, and allows researchers to download the questionnaires and access the data and resources to replicate the results.

In each country, representative population samples of about 1000 respondents – in Russia we surveyed 1498 individuals – answered 12 survey questions that were designed to measure time preference, risk preference, altruism, positive reciprocity, negative reciprocity, and trust. These were selected in an initial survey validation study, in which the GPS preference measures were selected based on their performance to predict choices in incentivized choice experiments for each preference. The survey questions were also pretested in culturally heterogeneous countries.

What the Survey Shows

The survey allowed us to gauge heterogeneity across countries and to answer a set of very important questions, such as: Do countries differ in terms of average preferences?  Are differences in aggregate preference profiles correlated with the cross-country variation in outcomes such as economic development, charitable activities, or violent conflict? Are country-level preference profiles related to differences in geography, culture, language, or religion? How large is cross-country variation in preferences relative to within-country variation? How does individual-level heterogeneity in financial, labor market, or prosocial choices vary with preferences around the world?

Answers to these questions help us to understand why important economic outcomes differ within and across countries. They might also contribute to an understanding of why certain countries are richer than others, why countries have different institutions and forms of social interaction, why different policies work differently in different countries, why mutual agreements can be more easily reached between some countries than others, etc.

How Preferences Differ

People's preferences vary from country to country, but the differences are even larger within countries and these individual characteristics may be even more important

Many of the world's most patient populations—based on their answers to questions about immediate vs. delayed rewards—are located in Europe or the English-speaking world, while risk-taking preferences are particularly prevalent in Africa and the Middle East. Prosocial preferences (altruism, reciprocity and trust) are pronounced throughout Asia and relatively weak in sub-Saharan Africa.

At the individual level, people's preferences differ by gender, age and cognitive ability. For example, women are more impatient, less risk-tolerant and more prosocial than men. Cognitive skills and age are linked to all of the risk, time and prosocial preferences. Yet these relationships, to some degree, are country-specific. While the relationships with risk-aversion and gender are in same direction in most countries, the age profile for patience depends on the country's development level.

At the country level, there are links between preferences and biogeographic and cultural variables, such as absolute latitude, the presence of large, domesticable animals and agricultural suitability. Patience, specifically, is strongly positively related to a set of variables previously associated with comparative development, namely Protestantism and measures of individualism.

 

People's preferences are linked to behaviors and economic outcomes

Within countries, patient individuals are more likely to save and have higher educational attainment, and more risk-tolerant individuals are more likely to be self-employed and smoke. Social preferences are predictive of prosocial behaviors and outcomes, such as donating, volunteering time and providing help to strangers, friends and relatives.

Across countries, preferences vary with outcomes ranging from per capita income to entrepreneurial activities to the frequency of armed conflicts.

Further Exploration

The data from the survey allow us to determine that people's preferences vary by country and within-country around the world, and that these differences are linked to both individual-level and aggregate cultural and biogeographic characteristics. The results also provide evidence that these differences in preferences are relevant to economic outcomes. However, much more can be learned from this global survey. The novel raw correlations we found between country-level profiles and aggregate economic outcomes call for a more detailed analysis of the underlying causes.

Future Research in Behavioral Economics and Collaboration with HSE

I have been successfully collaborating with Hartmut Lehmann since more than a decade. We initially worked on projects in which we used personnel data to better understand how labor market outcomes were shaped during transition. Later, Hartmut and I introduced preference measures in the Ukrainian Longitudinal Monitoring Survey (ULMS) and studied how preferences interact with labor market outcomes. In 2017, we teamed up with Vladimir Gimpelson to study how character skills and preferences in interplay with cognition affect labor market outcomes in Russia. To this end, we also plan to introduce the GPS preference module into the Russian Longitudinal Monitoring Survey (RLMS). HSE is an ideal partner in this endeavor.