.png)
Hamlet may appear indecisive and inconsistent, but he was in fact far more rational than many confident politicians. Indecisiveness can even help us make rational choices, says Ozgur Evren, a professor at the New Economic School. Yet circumstances themselves often force us to act more quickly. In an interview with GURU, Ozgur discusses the decision theory that he studies.
Ekaterina Kravchenko
– What is the main difference between decision theory and behavioural economics? Are they two perspectives on the same subject?
– Both decision theory and behavioural economics study how people make decisions and both rely on experimental and empirical evidence. However, their areas of focus differ. Behavioural economics builds models in which behavioural patterns systematically deviate from the rational benchmark. These models examine human behaviour within a given institutional framework, with the main focus being on the consequences of these deviations – their impact on markets, prices, savings and other economic variables.
By contrast, decision theory focuses on the foundational building blocks of models – behavioural patterns known as axioms. It formulates requirements for preferences and constructs mathematical models of choice behaviour. As theorists, we propose different classes of models. Two approaches can be distinguished here. The normative approach focuses on prescriptive choice and preference axioms. The best-known examples are the rational choice model based on the completeness and transitivity axioms, and the expected utility model for decision making in conditions of uncertainty. The other approach is descriptive: these models permit deviations from standard assumptions and describe how choice is actually made; they also incorporate preference incompleteness, context dependence, self-control problems, and other systematic deviations from the normative standard.
The first key axiom of the standard model of rational behaviour is transitivity, which is a consistency condition. Suppose that, when choosing between X and Y, you choose X, and then, when choosing between Y and Z, you choose Y. What would you choose when comparing X and Z? If you choose X, your preferences are transitive. Another key axiom is completeness, which assumes that a person can compare any two available alternatives and either prefer one to the other or regard them as equally valuable.
These axioms may seem too abstract and far removed from everyday life. After all, is this really how we shop in a supermarket? In real life, the person in our example might suddenly choose Z instead, thereby violating the transitivity axiom. Decision theorists therefore propose classes of models that can be calibrated depending on the particular task. As with any economic model, the ultimate goal is not merely to describe observed behaviour but also to make predictions – for example, to anticipate a person’s choice in new circumstances or assess how their choice would change in response to a particular economic policy measure. This is precisely why models and their assumptions must be clearly formulated.
– In your view, where does theory connect best with practice?
– Research on decision-making is relevant to both consumer markets and marketing. One popular topic is behavioural finance, which examines investors’ decisions, their attitudes towards risk, and systematic biases in information processing.
There is significantly less research on behavioural anomalies in politics and voter behaviour, at least in relation to axiomatic decision theory. Voting in elections is perhaps one of the most complex phenomena to analyse. Over time, behavioural political economy may develop further, and politicians may take into account these studies when making decisions.
– What has been the most surprising finding in your research?
– Of course, my best research paper is yet to be written. I am currently working on an exciting project with one of my PhD students, Kirill Savin. We have found that by relaxing the assumption that a person can always compare any two alternatives, we can improve the accuracy with which we predict their choices. In other words, indecisiveness can make a person’s behaviour more «predictable» because it becomes easier to rank or compare some options without violating consistency conditions elsewhere.
– So, can being indecisive ever be useful?
– Many researchers in my field have always regarded indecisiveness as a rational mode of behaviour. There is nothing wrong with admitting that you do not know what to choose. Is it wise to admit that you do not know what to choose? I think so.
– Hamlet is a classic example of a character who hesitates for a long time before finally taking action. However, in his paper «Hamlet and Rational Choice», Jim Leitzel of the University of Chicago argues that the protagonist’s delay is consistent with economic rationality.
– As far as I can tell, Hamlet acted wisely by trying to find the truth and gather information. However, he waited too long, missed some opportunities, and ultimately brought about a tragic ending. It is easy to judge with the benefit of hindsight. We should also take into account that a person’s psychological state can change. While revenge may provide immediate satisfaction, it routinely yields to subsequent moral guilt. Determining a severe penalty for a crime is a profound decision that alters the lives of everyone involved, demanding deep reflection. It is precisely for this reason courts hear cases carefully before reaching a verdict.
– Psychological states are not an economic issue.
– Multi-self models in economics focus on uncertainties regarding the decision maker's own moral judgments, preferences, etc. A decision maker faces such uncertainty if he or she does not really know what he or she wants, and this is precisely what we mean by indecisiveness. In such a situation, we may end up regretting any decision we take if it proves wrong when we finally learn what we should have done, and that moment of truth will always come. There is a well-developed theory of regret concerned with such uncertainties. In principle, a sophisticated person may also approach this as an optimal stopping problem. An action should be taken when the marginal benefit of obtaining new information (based on introspection or other possible channels) falls below the marginal cost of waiting.
– Could not making a decision be considered the best possible decision?
– It’s an interesting question. Experience suggests that hasty, potentially regrettable decisions are among the worst outcomes. However, it is much harder to think things through, stay calm, and find the right solution than it is to make a quick decision. It is hard to control one’s impulses. Moreover, the brain needs a great deal of energy to function, so deliberation is a costly process. This is why people may rationally ignore certain information.
– Cognitive costs cannot always be measured or quantified.
– I agree, quantitative measurement is difficult, but these issues are certainly important. For example, when you are hungry or upset, you might make a poor decision. Consider investing, for example. You see warnings everywhere: «Think carefully. Do not make hasty decisions. Do not click immediately.» Yet people still rush and sometimes lose a great deal of money.
If you work outside the financial sector, you may not have enough time to analyse investments properly. If this is the case, you should ask yourself whether you are exposing yourself to too much risk by managing your investments yourself. It may be better to entrust your money to a professional manager.
– Has your specialisation in decision theory influenced the way you make decisions in life?
– I tend to think carefully about important decisions. This way of thinking is sometimes appreciated and sometimes not. For example, it can cause problems in a relationship if a partner does not want to discuss things in detail. Overall, though, I cannot complain. I find it beneficial to think things through.
– In many countries, over-indebtedness may result from a preference for current consumption. Why do people tend to prioritise immediate gratification over long-term financial stability?
– People have different circumstances, and there is nothing wrong with borrowing as long as they understand that they will have to reduce their spending in the future in order to repay the loan. If this is a deliberate strategy, it is a rational choice.
Nevertheless, people tend to think too little about the future. Our perception of time is often distorted, with present gains seeming larger and future costs being underestimated. Procrastination is a classic example of such behaviour: a person decides to do something tomorrow, but then postpones it again when tomorrow arrives. This kind of time-inconsistent behaviour departs from the standard model of dynamic choice behaviour.
– For students who plan to pursue a PhD in economics, could you please tell us how promising your field is in terms of its research potential?
– Decision theory is a smaller field than behavioural and experimental economics, and it is considerably more mathematically demanding. Our work involves functional models and is, in many ways, a mathematical exercise, yet a substantial part of it is also dedicated to analysing experimental data.
Researchers in decision theory have long been concerned with testing axioms. However, a hypothesis may be testable in theory without being easy to test using real-world data. In the future, we may place greater emphasis on empirical data. The expansion of large-scale datasets and the emergence of new data sources will shape the research questions we pursue. We may then study machines in much the same way as we currently study human beings (such research is already under way, editor’s note).
– In the past, research could take several years to complete, and by the time it was published, it was often no longer relevant. How has this process changed?
– The publication cycle has become shorter. Nowadays, editors can reject papers within just a few days without sending them out for peer review. As a referee, I am often given deadlines of one to two months. If the process takes longer than that, it suggests that someone is not doing their job properly. The number of academic journals has grown. In the past, the leading journals for microeconomic theory were primarily the Journal of Economic Theory and Games and Economic Behavior. Now, there are four or five leading journals in this field, and they aim to make decisions on submissions within two to three months.
– Economic modelling is often criticised for being overly mathematical. What is your view?
– It can be difficult for a non-specialist – or even for a specialist from another field – to understand an article published in a reputable economics journal. This is not necessarily a bad thing. If you read an article in a civil engineering journal, you may understand less than you would understand from an article on economics. As technical researchers, we communicate with one another through models.
However, despite their technical sophistication, these models often have serious weaknesses. For example, unlike engineering structures, economic models cannot be expected to produce consistently reliable results. They may fail to provide sufficient predictive or explanatory power. Economists should be aware of these weaknesses and caveats and should not be overconfident in their recommendations. There are no perfect solutions in economics. We simplify models in order to make predictions; without such simplification, it would be impossible to make sense of the incredibly complex real world. Sometimes, we fail to see beyond the model. There is also a risk of pursuing technical sophistication for its own sake. Such a tendency is a warning sign. Solving a difficult mathematical problem and presenting it as an economic contribution will get you nowhere, except perhaps in a mathematics journal. Conversely, if you have excellent intuition but dislike mathematics, economics may not be the field for you either. Economics today is not what it was a century or two ago. It has become more mathematically sophisticated, and that is a good thing.
– Nobel laureate Paul Samuelson joked that economists work for the applause of their peers. One of the main criteria used to evaluate a scholar is the number of publications they produce. But how difficult is it to keep up with the required publication pace?
– The pressure to publish research papers is enormous. Leading universities require academics to publish a certain number of papers within a given timeframe. This «publish or perish» principle can create distortions, for example, it may lead to an overemphasis on paper quantity or researchers may spend excessive time pursuing top-tier journals when alternative venues might offer a more suitable fit for their specific work.
But at the end, quantity and quality of publications are natural criteria for evaluating the career of a researcher. To succeed, researchers must seek a balance between quantity and quality. This requires one to anticipate potential challenges, and avoid spending too much time on a project that has little chance of success.
– If you work in academia, you must be prepared to receive harsh criticism.
– Of course! Criticism helps reveal the weaknesses of the research. Researchers look for flaws in one another’s work, and this is how the research output ultimately advances. Feedback can often feel harsh, so an academic career requires a thick skin. You must always be ready to answer the questions, and try to learn from them, even when you feel that the question was not so great. After all, smart people tend to be confident, and that may occasionally lead to strange or arrogant questions and comments. I have always found giving presentations to be the most challenging part of the job.
– Is competition more intense in other fields of economics than in decision theory?
– Decision theory is not a very popular field. It requires a unique set of technical skills. But in the job market, universities often prioritise hiring experts in applied fields, such as finance and macroeconomics. This is a general issue for microeconomic theorists.
– Just as motorised transport changed not only the speed of travel but also the entire logistics system, AI has the potential to transform economic relations. However, forecasts about the impact of AI on the labour market and the economy are based on the current economic structure. But can we really make predictions about a system that we are unable to imagine?
– It is a difficult task, I agree. One Nobel laureate (Paul Krugman – editor’s note) once predicted that the internet would have no greater impact than the fax machine, but he turned out to be wrong. Mistakes happen.
Economists hold a wide range of views about the risks and benefits of AI. Market participants appear enthusiastic, but there is already talk of an «AI bubble». The leaders of the industry started talking about the need to limit the pace of its advance, and setting up rules. By now, it is clear that advanced models can solve deep mathematical problems. Overall, it is a jungle out there, but certain patterns appear: AI will influence many professions, including academia. It certainly has the potential to increase productivity. However, there are also significant risks associated with it, namely security risks and growing income inequality.
– Students make extensive use of AI when writing their term papers and theses. Academic circles are debating how to assess work produced with the assistance of AI.
– Education signals graduates’ abilities to the labour market. AI has made this assessment more difficult, but well-designed assignments should enable us to distinguish between students’ abilities and AI-generated performance.
As a teacher and researcher, I am interested in assessing students’ intellectual potential, as well as their ability to tackle problems and think independently. If a student outsources some tasks to AI, I am no longer assessing what that student is capable of; instead, I am assessing their ability to work with algorithms – and anyone can use Google. In exams, I assess students’ problem-solving abilities, as these provide an indication of their capacity for sustained effort and the depth of their thinking.
Employers are satisfied with NES graduates precisely because we are properly measuring students' intellectual potential, aside from the excellent training we provide.
– You were born in Turkey, studied in the United States, and now work in Russia. Economists often draw parallels between Russia and Turkey, as both are emerging markets facing high inflation and exchange-rate volatility, significant state involvement in strategic sectors, and considerable geopolitical risks. How would you characterise the Turkish economic model, and in what ways is it similar to the Russian model?
– You correctly identified the similarities, noting that both countries are emerging markets grappling with inflation and currency volatility. In Turkey, for example, inflation has been a chronic problem since the 1970s, at times reaching annual rates of 100% in the 1990s (inflationary spikes have recurred – for example, the peak of the latest inflationary cycle was 85.5% in October 2022 – editor’s note). Normally, authorities try to curb inflation by raising interest rates, but this is a painful measure that suppresses demand. In Turkey, the lira's depreciation also causes inflation, and it is perceived by the public as an indication of a bad economic situation, which hurts the government's image. In 2019, the government began intervening heavily in the central bank’s operations, particularly through unconventional and non-transparent interventions in the currency market. Vast amounts of foreign-exchange reserves were sold in an attempt to halt the lira’s decline. Meanwhile, they also adopted a low interest policy. Quite predictably, inflation went out of control, even in dollar terms. Turkey is now considered an increasingly expensive place to live.
It is also important to note that Turkey's economic model is based on the export of goods. They are finding it increasingly difficult to compete with countries such as Vietnam, India, and Egypt in terms of labour costs. Overall, government interference in monetary policy has severely damaged the country’s economic model.
A fundamental difference between Russia and Turkey lies in the independence of their respective central banks. Also, Russia's natural resources provide a steady source of foreign currency, while the Turkish economy is much more dependent on capital inflows and export goods with low profit margins.
– From a research perspective, which aspects of your home country interest you most?
– As a decision theorist, I wonder why people supported some policies over the past 20 years, especially the constitutional changes that shattered the foundations of Turkish democracy. Perhaps they did not anticipate the economic consequences, or perhaps they were guided by emotions.
The interactions between financial flows and political outcomes are also an interesting issue. The initial economic success of the Erdogan period owed much to his good relations with the European Union and the United States, as well as an influx of foreign capital. After the coup attempt of 2016, however, investment flows reversed, and they remained negative until Erdogan’s victory in the 2023 presidential election. After that election, large investment banks started expressing strong interest in Turkish financial instruments. To some extent, this is related to Mehmet Simsek, who rejoined the economic team as Minister of Treasury and Finance. Indeed, the central bank began raising interest rates. But inflation remained higher, which makes it difficult to explain the sudden influx of capital. More importantly, the risks that threatened the foundations of Turkish democracy only worsened, thanks in part to the financial stability provided by the investment flows. Overall, it appears that the decisions of large financial institutions carry profound political implications, which do not always align with the long-term public interest.
