Vojta in the library holding a book

Meet Our Alumni: Experiments are an invaluable tool to understand motivations and constraints in human behavior, says Vojtěch Bartoš

Vojtěch Bartoš, an Assistant Professor at the Department of Economics at the Ludwig-Maximilans-Universität* in Munich, currently visiting the Department of Psychology at the Princeton University, graduated from the PhD in Economics program in 2016. Vojta now shares with us his knowledge and experiences from his field experiments work in several African and Asian countries and explains what led him to work in academia.

Since 2016 you have been an Assistant Professor at the Ludwig-Maximilans-Universität in Munich. How did your CERGE-EI experience help you to get this position?

I owe CERGE-EI a lot. To be honest, when I started my PhD, I was not fully convinced about an academic career. It was one of the options I was considering. Over the course of my studies, I finally started appreciating the beauty of an academic life. I had a great opportunity to learn from the inspiring faculty during lectures, tutorials, and personal interactions.

The coursework during the first two years was intense, but it gave me strong theoretical foundations and a good overview in a range of subfields in economics. Despite being quite long, the two years of coursework gave me plenty of time to read papers from many different fields. I really appreciate this.

I was also learning to enjoy academic exchange in the form of research seminars and conferences, of which CERGE-EI offers plenty at a world-class level. At first, I was a passive consumer. But slowly I started contributing myself – by asking questions in seminars or by meeting the speakers in bi-lateral meetings. Some of them were top researchers in their fields. Interactions with them helped me shape my research ideas and also gave me some confidence that I may also contribute to the broader academic community with my own research. I gained even more confidence when my research ideas attracted funding and I got a chance to do field work. And lastly, I had a chance to visit New York University for a semester as a doctoral fellow. This finally gave me enough courage to start applying for academic positions on the international academic job market. I got very lucky on the market and I finally decided to take the offer from LMU, one of the best universities in Germany.

Why did you choose a career in academia?

I learned to value the freedom academia offers. My job is to seek answers to questions that I am very curious about. I cannot think of many other careers that would give me such freedom. Sometimes it is hard to draw a boundary between my work and personal life. I try to take this as an upside of the job. I am constantly looking for new questions in the world around me. And I keep thinking about ways of answering them. It is great that modern economics is very broad, and thus I do not need to restrict my thinking to fit within some narrowly defined boundaries of the field.

Also, the academic community is full of wonderfully smart and kind people. I enjoy interacting with my colleagues a lot. Since I get to meet researchers from all over the globe, this further broadens my horizons and adds new perspectives. Doing field work brings an extra benefit of interacting with people from outside of academia, across a wide range of socio-economic groups. These interactions are often eye opening and they provide inspiration for further questions that need to be answered.

Your research focuses on behavioral and development economics. You have been involved in several field experiments in Afghanistan, India, Malawi, and Uganda. Can you tell us more about how such field experiments work?

The world is complicated and human behavior is typically driven by a complex set of motivations and constraints. In order to understand their respective role in human behavior, experiments are an invaluable tool. In a typical experiment, we manipulate these motives or constraints one by one and see how this affects behavior. Then we can credibly claim that manipulating domain X affects the outcome we are interested in, in which direction, and by how much, if this is relevant. Such experiments can be done in a laboratory where students sit in front of their computers and we study some general economic theories.

But students are not a very representative sample of the world’s population and we often want to understand the behavior of different groups of people. In field experiments we actually reach out to the population of interest. In the case of some of my experiments, this literally means packing a backpack, heading off to another part of the world, and, for example, spending an extended period of time in remote parts of Afghanistan, living together surrounded only by the local people. And this is the place where we run the study and collect data that I later evaluate back in the comfort of my office.

One example of a field experiment that cannot be done in a laboratory is my paper from Uganda. My colleagues from Prague and London and I  tried to answer the question of why poor people seem to behave more impatiently compared to the rich: they save a smaller share of their budgets, spend larger shares of their budgets on “temptation” goods, or miss investment opportunities that offer potentially very high returns. Maybe the poor just cannot afford to make patient choices? Or are their preferences indeed different due to poverty? It is hard to tell by observing peoples’ behavior in real life. You see people being either rich or poor and comparing their choices misses the fact that there is a plethora of reasons for why they are different. But an experiment can help. We worked with around 300 farmers in rural Uganda, who were all very poor. We randomly manipulated whether we let the farmers think about poverty related concerns or not in an experimental setting. We told people about a problem that may happen to them and let them actively think about how they would solve such a problem. For example, how they would deal with the loss of their entire harvest. We also asked the “control” group, but only about some very mild shocks that would cause no particular problems. Then we measured the farmers’ patience in an experimental task. We let them allocate a set of minutes of entertainment (watching videos on a tablet computer — they enjoyed it very much) and work (a tedious task of sorting beans by colors, which they did not like as much) across two hours, one week apart. To motivate people to take the choice seriously, they had to carry out the task in order to get compensated at the end. We found that those who we made think about poverty related concerns allocated more minutes of entertainment in the first week, at the cost of having to work more two weeks later. This means that mere thinking about poverty makes you impatient. Thus, relieving poverty may not only have the direct material effect of giving someone money, but it may also lead to them engaging in more forward-looking choices.

Field experiments can also study the effectiveness of interventions aimed at changing people’s lives. One example would be the large-scale field experiments that my colleagues from Munich and Groningen and I have started. We experimentally estimate the effects of an entrepreneurship training for university students at leading Ugandan universities. Most businesses in poorer countries are one-person enterprises and their owners often become entrepreneurs out of necessity, as formal jobs are scarce. Business trainings offered to such entrepreneurs do not deliver encouraging effects. Among the more promising approaches seem to be trainings aimed at teaching a so-called “entrepreneurial mindset” – finding niche business opportunities, actively addressing business challenges, etc. So, we ask if offering such trainings to university students, arguably those with the highest human capital endowments, may help in the creation of companies that grow and that create employment. We use the fact that the program we study is popular and there is more demand for the training than the organization can accommodate. Hence, we give everyone an equal opportunity and choose who gets admitted by chance. This allows us to compare those who got the training to those who did not and attribute any difference between the two groups solely to the training. This project is a large endeavor as it requires surveying more than 2000 students at various points of time. We are still in the process of collecting the data. We hope to find interesting results. And we also hope to see that our results will be able to help to shed better light on problems facing labor markets in a setting such as the one we are studying.

While finishing your PhD studies you published your research paper in the American Economic Review, which is in the top 5 of Economic Journals worldwide. What was the paper about? Has this huge success helped you in your career?

Our paper is precisely about what you asked in your last question. Let me explain. The title of the paper is “attention discrimination”. We developed a theoretical model showing that decision makers, for example, HR managers on labor markets, decide whether to pay further attention to individuals based on the first signal they see about an applicant. They need to be selective because reading further details costs them time. Often, the first signal is a name. But a name can tell the HR manager about your gender or your ethnicity. If an average person from your group is associated with poor performance on the job, the manager may decide to drop your application without reading any further, regardless of how good you are. The chance of discovering a good prospective worker from that group is simply too low. The manager then saves her time but never learns about your true qualities. On the other hand, for other types of markets decision makers are happy to invite almost everyone, without wanting to know too much about them. But here again, receiving a signal about an applicant’s “lower-quality” group affiliation may trigger a warning that requires a deeper check of that applicant’s background. This may sound fine, as only those who are really bad get turned down, but the problem is that this only happens to individuals from the ex-ante “lower-quality” group. Comparably bad individuals from the “higher-quality” group get in. In either of the two types of markets the applicant from a “lower-quality” group is harmed – in the former because of too little attention, in the latter case because of getting too much attention. This exacerbates discrimination.

We then tested the model using a series of field experiments in the Czech Republic and in Germany. We did these experiments online. We created fictitious applications for real job offers and rental housing offers, and we manipulated the names of the applicants in the emails we sent. We then checked whether the decision makers paid attention by opening a website in the application email that we embedded with a hidden tracking code. And we also measured whether the decision makers then followed up with the applicant. The results supported our theory. We also provided some extra support for the findings through surveys with decision makers from both types of markets.

So, to get back to your question, I was obviously trying to send the signal of the article’s success as soon as I could when applying for academic jobs. I am sure that this provided me with quite some positive attention from prospective employers.

You are now visiting Princeton University for 6 months. Is it possible to benefit from the stay, given the situation? What are you currently working on?

I’m visiting the Department of Psychology where I got an invitation from a professor whom I coincidentally met at CERGE-EI during my PhD studies. Well, the timing did not go quite as planned. I had big plans to visit lots of seminars both in economics and psychology, get a chance to meet the faculty, and take classes in psychology. I also had several seminar talks scheduled at nearby universities, and one at Princeton.

All went very well until March 9 when the university president announced the cancellation of all on-campus activities. This came as a shock as it was announced many days before the New Jersey state-wide lockdown on March 21, and one day before the first COVID-19 related death in New Jersey. Seminars were cancelled and classes were shut down. I’m here with my family and the daycare is also closed. Despite all their great efforts to support us remotely, the childcare I need to provide has increased substantially. I still get a chance to interact with a few people who I got to meet virtually and am attending some virtual seminars at US universities that have now popped-up. I also gave a virtual talk. The situation is improving a little, so I hope to get a chance to interact with the faculty some more in June and July.

But the covid crisis also brought an interesting new line of work. I put together a policy brief for the IDEA think tank on how behavioral economics may help in fighting the pandemic. And together with colleagues from Prague and Munich, and in cooperation with PAQ research, a sociological research company, I started a project on tracking the behavior of Czechs during the pandemic. There are several promising research outcomes emerging from this.

What are your next professional plans?

The academic system in Germany is such that my contract at LMU is limited to six years. The deal is that I have a temporary contract and my teaching load is fairly low. This allows me to focus on research. I’m now mid-way through the contract and I will soon start looking for full professor positions. My colleagues in similar positions have been quite successful securing professorships at universities in Germany, Austria, Switzerland, the Netherlands, France, Spain, and Italy. And who knows, maybe there will also be an interesting opening in the Czech Republic, eventually.

*2023 placement update: Senior Assistant Professor, University of Milan, Italy


2 thoughts on “Meet Our Alumni: Experiments are an invaluable tool to understand motivations and constraints in human behavior, says Vojtěch Bartoš”

  1. This year I applied for a Ph.D. in Economics and I have a good research proposal and have strong research skills but my application rejected.
    i am really interested to be a part of CERG for the coming year, can you please give me some information on how to come up with strong Application for 2021?

  2. I am very interested in learning economics at CERGE-EI, but my application was rejected. And, I hope to apply next year. What can I do to improve my application?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.