Interview , News of 6 March 2017
SOEP People: Five questions to Rainer Winkelmann
Rainer Winkelmann’s research on unemployment and happiness using SOEP data led to his groundbreaking 1998 paper “Why are the unemployed so unhappy?” (written jointly by Liliana Winkelmann), which conclusively demonstrated—for the first time—that unemployment makes people unhappy. It is the most widely cited paper in the history of the SOEP.
Rainer Winkelmann studied economics at the University of Konstanz, Paris IX-Dauphine, and Washington University in St. Louis, and he holds a Ph.D. from the University of Munich (1993). He has taught at Dartmouth College, USA, and the University of Canterbury, New Zealand, and was a visiting professor at Harvard University, Syracuse University, and UCLA. He is a Professor of Economics at the University of Zurich since 2001. His research is in micro-econometrics with applications to social policy issues in the fields of labor, health, and well-being. He is a member of the DIW Berlin Scientific Advisory Board and chaired the SOEP Survey Committee up to the end of 2016.
The video of our interview, “SOEP People: A Conversation with Rainer Winkelmann” can be found in the
1 . Your paper “Why are the unemployed so unhappy?”, published in 1998 in the journal Economica, paved the way for a growing body of research on unemployment and happiness. What did you find out?
First, we found that unemployment matters a lot to individuals’ well-being. For instance, unemployed people on average have a 10 percent lower probability of being happy than employed people. Second, we found out that income is not that important for well-being. This fits with an idea that emerged around the same time that economics is too narrow in focus. It’s not just what makes workers go to work. It’s not just a high per capita GDP that’s needed for a good society. There is much more to it than that—there are other things that people look for and that contribute to their well-being. It’s not just money.
Incidentally, we published another paper three years before the Economica paper in the journal Konjunkturpolitik, where we studied how unemployment affects the household when one partner is unemployed. The SOEP data allowed us to do that because they provide the family context. What we found is not really that surprising: women are very unhappy when their partner is unemployed. This actually means that unemployment is overall even worse than what we described in our Economica paper, because it not only affects the unemployed person but also spills over in the household.
2 . Today, almost 20 years later, do you see policy impacts of that research?
One part of the long-term impact was to put life satisfaction and well-being research on the agenda and say it’s not just money that matters; there should be broader notions. It’s not enough to focus on macroeconomic factors like maximizing GDP; to have high well-being in a country, other things matter as well. That slowly starts to have impact in policy circles. Now the OECD has a “better life index” that takes account of these broader measures of well-being that came from life satisfaction and happiness research. The UN edits a world happiness report. Our research has supported the idea that one should judge progress not just by looking at income changes or GDP growth, but also by looking at other indicators.
3 . Several countries are discussing the introduction of a basic income. What does your research say about how a basic income might affect people’s incentive to work?
For economists, the idea of a basic income clearly has some appeal. As a labor economist, one is very aware of welfare traps: situations where people with low incomes who receive benefits have no monetary incentive at all to start working. Basic income would solve that. I think that our evidence is consistent with the notion that even a basic income would not stop most people from working because actually they like to work; they get social recognition from work. With a basic income, you can also work to supplement your income and have a higher income as a consequence. In this sense, there is some link between our unemployment research and the discussion on basic income.
4 . Your research has high policy relevance, but you’re also known in the SOEP community as an innovator and expert in micro-econometric methodologies.
Most of my research is really one step before policy-oriented research. I develop microeconomic methods and am happy if people use them in applied research that goes into policy reports, but I don’t have to be the person that actually does that. I find it more interesting to be guided by my curiosity than by current policy issues, so I think a bit more long-term about what to work on and what fascinates me. I find research fascinating because you can make discoveries—you think about questions that no one has addressed before. Whether it gets published in the end is almost secondary. We’ve written papers that were never published but I still thought it was a good experience and a good idea to do that research.
5. As a data user for over 30 years, you’ve seen numerous changes and innovations in the SOEP study…
The most important thing for my purposes was that from the start in 1984, the SOEP included a life satisfaction question, which no other survey had at the time and which was quite visionary. I think that has paid off nicely for the SOEP and for many researchers. We now have 32 years of data this year, so there are also tremendous opportunities for future research to look at long time series of consistent measurements in life satisfaction.
Another point that I think is important about the SOEP is that success breeds success. Once the SOEP was there—it was early and was doing good things—others picked up on it. A research community developed around the SOEP. That also makes the SOEP more attractive to you as a young researcher because you benefit from the experience, from the acknowledgment that this is a good dataset, and it becomes easier for you to publish. There are also the SOEP user conferences. All these aspects are important points when deciding what data to use.