Ekaterina Zhuravskaya, professor at the Paris School of Economics and winner of the George Gamow Prize RASA-USA 2024, told T-invariant about how modern economic science works, how to evaluate the results of sanctions, and what research into Goebbels’ propaganda can tell us about the current military mobilization of Russian society.
T-invariant: Economic science is still not always perceived by hard sciences as a proper science, and is seen as a set of opinions or models. How could we argue with such an opinion?
Ekaterina Zhuravskaya: Over the last 25 years, which is roughly equal to my life as a researcher so far, economic science has revolutionized itself. The field largely moved away from pure theoretical models, opinions, and descriptive studies, as you put it in your question, toward models that have predictive power and testable implications necessarily coupled with rigorous testing of those predictions. Advances in data availability (including individual-level data from administrative sources, surveys, and digital sources), enhanced computing power, and the development of econometric analytical techniques led to what is now usually referred to as the “credibility revolution” in economics. The field now prioritizes rigorous empirical methods to establish causal relationships rather than relying solely on theoretical models or correlational evidence. Causal identification, using approaches like randomized controlled trials, natural experiments, and instrumental variables, is routinely used to isolate the effect of one variable on another. Obviously, my research also uses state-of-the-art methods.
That said, it is important to recognize that social science—and economics, in particular—is not an exact science. It is a science because we make important out-of-sample predictions about human behavior, test them, and find significant predictive power in our theories. Yet, it is often harder than in other disciplines. In chemistry, the quality of testing theoretical predictions depends on how clean the dishes are. In my field, political economics, keeping the dishes clean—namely, eliminating the effects of other factors—is often challenging, even in randomized controlled trials for the following reasons. First, political attitudes and culture change, meaning context matters. Second, humans learn from research results and may adjust their behavior. Third, the very nature of randomized controlled trials on humans is challenging because individuals may change their behavior differently across treatments, not due to the treatments themselves but because they infer something about what the experimenters aim to test. All these factors can potentially contaminate political economics research results. Thus, one needs to take special care to find proper placebo treatments and “clean the dishes better.”
Speech by Ekaterina Zhuravskaya at ISET – International School of Economics. Photo: Facebook
T-i: What kind of tasks do you address using these methods?
EZ: I am an empirical political economist, specializing in media and institutions. This means that I study how media affects people’s attitudes and behavior and how institutions affect economic and political outcomes. These are questions set in a global context and not related to a particular geographic area. To study them, one needs at a minimum at least two things: data on important outcomes and treatment, and sources of exogenous variation in this treatment. I have worked on the US, France, Poland, the Middle East, China, Central Asia, and Argentina, as well as global data covering over 100 countries. Naturally, some of my work is on Russian data and transition countries because I know the context well, and this helps in finding both the data and the sources of exogenous variation.
T-i: Your Gamow award is given for your contribution to the understanding of institutions, but if you yourself would have to pick one of your most important results, what would it be?
EZ: This is a tough one. There are many results which I would consider interesting and important (and which are rather well cited). But, if necessary, in retrospect, one could single out my work on the political persuasion effects of traditional media in immature democracies and autocracies. When I started working in this area, we only had vague hypotheses that propaganda should work. Otherwise, why would every autocrat engage in it? Yet, it was a priori not obvious, because one could expect people to recognize the bias and counteract it. Our work (and I use “we” here not as a royal “we,” but to acknowledge my wonderful coauthors) has rigorously shown across different contexts that:
-
Mass media propaganda is effective on average, and people do not fully unravel the bias.
-
It is often less effective than milder slanted news.
-
There are important heterogeneities in the effects of propaganda: it is most effective on positively predisposed and may backfire on those who are (a) negatively predisposed and (b) knowledgeable about the subject of the news.
Now, everyone in the profession considers these results well-established facts. But this is largely because of our findings. Importantly, these results have been replicated in many different contexts, first by us and later by other researchers who followed.
Here, it is worth remembering a joke, which I find funny mostly because there is some truth to it. In the eyes of other social scientists, economists are those who spend a lot of time and resources collecting data to show what other social scientists (who do not use the rigorous methods I mentioned) already know very well. Some of the results I just described fall into this category. Yet, before our work on radio persuasion in Germany during the Weimar Republic and right after its fall, leading up to the March 1933 elections, many claimed that the effects of Hitler’s speeches on the radio in January-March of 1933 was more persuasive than the Weimar anti-Nazi messages during the year before Hitler’s appointment as a chancellor. We show that this claim is wrong and now we understand why. Overall, if we do not look at the data, hypotheses remain only hypotheses.
T-i: To what extent is your research driven by the political agenda?
EZ: As a scholar of propaganda and, more generally, autocratic regimes, I am very interested in what is going on in Russia now. Needless to say, it is very important to fully understand how autocrats mobilize the population to fight an aggressive, criminal, unjust war. Whether I will be able to translate this interest into research output remains to be seen. The closest work to my current interest in Russia is my previous research on the effects of Goebbels’s radio propaganda in Nazi Germany, in which, with my wonderful coauthors, we show that, despite being quite effective on average, Nazi propaganda backfired (i.e., produced the opposite effect) on certain groups of the population.
T-i: Do you feel like you are a part of the Russian scientific diaspora? If so, what does this identity mean to you?
EZ: We speak Russian at home, and I work a lot with Russian-speaking academics based outside of Russia. I would say that this naturally makes me part of the Russian-speaking diaspora. Whether this also makes me part of the Russian diaspora is a difficult question, especially given the tragic events unfolding due to Russia’s horrific invasion of Ukraine. I try to view this identity as honoring a shared intellectual heritage and fostering constructive dialogue. As a social scientist, I feel a responsibility to uphold universal values like truth, critical thinking, and collaboration across borders, while condemning aggression, propaganda, and hatred—elements that are integral to Putin’s regime. There are strong role models in history, and I hope the Russian scientific diaspora can live up to them. One could think of parallels with WWII. Erich Maria Remarque and Hannah Arendt were vocal representatives of the German diaspora.
T-i: What role, in your opinion, should Russian intellectual and scientific emigration play in the conditions of “eternal Putin”?
EZ: It needs to support Ukraine and all anti-war efforts. Academics – and especially social scientists – cannot claim being uninformed or unaware and should take a clear position in this conflict.
T-i: Sanctions against Russia have not brought the promised effect. Why and how could they have acted differently?
EZ: The important question here is what the right counterfactual is. Expectations for sanctions were clearly too high. As academics, we need to compare the impact of sanctions not to a wishful-thinking benchmark but to a realistic counterfactual (what would have happened without the sanctions). By this standard, sanctions have been successful in undermining Putin’s capacity to win the war in Ukraine.
Without sanctions, Russia would have still had access to hundreds of billions of dollars of sovereign reserves. It would have received many more petrodollars. This would have allowed Putin a lot of things he does not have. Putin would have been able to buy modern Western military and dual-purpose equipment and would have been able to use
these resources to recruit many more soldiers. Overall, we should recognize these successes of sanctions and learn from them. Sanctions should continue and become stricter, especially, secondary sanctions, which target firms and banks in the third countries, who continue doing business with Russia.
T-i: In different countries, the DEI ideology in the academic system looks different and has different effects. In the US, there are heated discussions about this. In Germany, a significant part of the standard DEI is not applicable due to historical reasons. And how is this felt in France?
EZ: In France, there is no tangible self-censorship or other limitations of academic freedoms imposed by political correctness or DEI. While there is growing awareness of gender and socioeconomic disparities, in society and in academia, discussions around race, ethnicity, and other identity-based issues in universities are still largely open.
About gender issues in academia: There are certainly important systemic biases that lead to the underrepresentation of minorities and women in academia. These biases are tangible and can and should be corrected through a policy response. But this should and can be done through changing culture during the impressionable years, much earlier in life than at the age at which people enter universities.
Text: Alexandra Borisova
Alexandra Borisova 14.11.2024