Interview with Dr. Abhinash Borah

By Evita Rodrigues (ER) & Gulhaider Zaidi (GZ)

It brings us immense amount of elation to share with you, our interview with Dr. Abhinash Borah (AB). His insights into the realm of Behavioural Economics are a must-read. So, read on!

GZ: The economics courses taught in colleges usually depict people as cold-blooded optimisers and rational. We do have courses in Behavioural Economics as well but these courses are in their initial stages of development and do not constitute a major portion of the syllabi. Such a bland education fails to give economics students hands-on-experience and practical knowledge. Do you think, it is time we reviewed college syllabi? How do we go about breaking the norms of rationality in this venture?

AB: Yes, personally speaking, it isn’t a statement about universities but if you give me complete autonomy about deciding the economics syllabi, I would do things a lot differently than they are being done. Methodologically speaking, there are two ideas that act as a glue and tie various fields of economics together. One is the idea about understanding behaviour- everything that we do in this discipline starts with the idea of understanding individual behaviour, in that we share the goal with a lot of social sciences as well. The second idea is a bit unique to economics. It is that we are extremely interested in the aggregation question- what are the aggregate consequences of behaviour at the individual level in terms of aggregate economic outcomes.

It is important to give students a sense of what we do in economics and why it is important.The first course in economics for undergraduates should not be about tools but appreciating the relationship between individual behaviour and aggregation so that they can think about a wide range of problems and this should be an entry-level course. Of course, the technical tools do have to come in later, but we must teach students why these tools are interesting for the kind of questions we are dealing with.

Usually, what happens is that some students get disinterested at the undergraduate level, precisely because they come in with some picture and we throw technical things at them which are dry and they end up saying, ‘Why are we even doing this?’.

Also, in terms of the curriculum, when we teach intermediate microeconomics, in addition to teaching Samuelson and how we think about neoclassical economics, we also need to teach them about criticism and the two must thence go hand in hand. What also happens is that the applied world and the theory world are kept separate, which is where modern curricula have to change. Something, like the idea of the experiment such as the field of ‘lab experiment’ is an important one in economics these days. In an intermediate microeconomics course, what is taught to students ends up becoming tenets of behaviour, which should not be the case. It must be constantly stressed that theory is empirically motivated and thus the idea of testing and experimentation must be introduced to students much earlier. Part of why economics becomes dry for undergraduates is because of a disconnect between experimentation and course content, hence, also becoming too abstract.

I would introduce something like Game Theory much earlier into the syllabus which would bring in more aspects of fun.Sometimes, our coursework lacks a degree of motivation for students for which we could load it with a lot more fun things which would make the understanding of strategic behaviour more engaging. After this, if the students find this interesting, then they can put in the hardwork of studying these fields. A lot of exciting content tends to come late on in an undergraduate course, so this is something we can meaningfully reshape.

Source: Sim Economics Society

ER: Could you tell us about your research program in decision-making under uncertainty? What subjects have you engaged with and what have been some of your findings?

AB: A lot of my research is about decision-making under uncertainty. I like to work in this area called ambiguity. When we teach you anything about uncertainty in the undergraduate or Masters level, we say that whatever uncertainty there is, it can always be captured by some probabilities over uncertain events. There’s a distinction in Economics that goes back to Night, which Keynes has also talked about, that there can be certain kinds of events where it is impossible to assign any probabilistic assignments. In light of this, ambiguity precisely refers to those situations of uncertainty, where you might not be able to capture the uncertainty itself and this is what some of my work is on.I also like to think, not just from the decision-theoretic perspective but also about some of these models and what are their implications on larger questions. For instance, I’m very much interested in voting decisions and social influence on decision-making.

Source: Decision Innovation

GZ: A very popular notion in Behavioural Economics is that of ‘default options’. If we talk about ‘Organ Donation Volunteer Rates’, we see that some European Countries like Austria and Belgium follow a policy, whereby the citizens who do not intend to donate their organs have to opt-out. Only around 10% of the people actually opt-out. Do you think, making citizens organ donors by default is a moral move, given that we all have rights over our body and most of the citizens are not able to opt-out of the programme because they feel that they are breaking a supposed ‘norm’? Is nudging them in this way justified?

AB: The default option seems to have very non-trivial consequences and sure there’s a place for it but I feel what we need to be careful about is that we should never pursue policies at a level which can be even remotely interpreted as a manipulation. I think, because we have these behavioural biases, when we look at our consumption, we see that we are systematically manipulated. The benchmark for policymaking needs to be so much higher, our hands need to be so much cleaner. We need to make sure that when we engage in policymaking, we aren’t systematically or implicitly manipulating people into doing things, that shouldn’t have a place in policymaking because our ethical standards need to be higher. Some people say ‘Look, Apple has your information, so what if the Government does too , except my standards for the Government are a lot higher than my standards for Apple!’

ER: What have been some of the significant examples of Behavioural Public Policy under the present government in India? How do you see our voting choices and preferences being an influence in the run-up to the 2019 elections?

AB: The one event that stands out is of course demonetization. It’s difficult to pinpoint the rationale because the goal post keeps shifting, but some of the arguments that came out like demonetization and shifting to an electronic form of payment were behavioural.

I find that slightly problematic because these forms of behaviour that you choose are very endogenous decisions and are functions of constraints you face. Sometimes, what is pushed as behavioural decisions has no basis really, because some of the constraints that people work with are very different. One of which I find very problematic with shifting from cash to cashless is that the modes of payment have ramifications for self-control and other things like that and so we need to be slightly circumspect when we have these kinds of grand visions of work.

Also, with respect to something like Swachh Bharat and sanitation, the desire there again is to change behaviour. While it looks to stop open defecation, we shouldn’t be naïve about it because there are always second and third order effects and we need to understand how these play out. And so, whenever we talk about behavioural change, some of it of course, is desirable. While proposing policies, we also need to see their consequences and how they play out.

To come back to the election question, yes, I’ve become extremely frustrated with the process. Elections or any social decision-making process, as problematic as it might be, is some way of aggregating individual preferences, and I just don’t think that aggregation happens anymore. We are systematically manipulated through electoral systems.

Let’s take an experiment. Let’s think about a policy under two different electoral regimes. We can have two policies, one where we focus on education, health and things that materially change people’s lives. Secondly, another policy where we play to people’s biases – more religious tension, caste bias, more divisions in society. These are 2 policies – A and B. Let’s presume people don’t have large biases, but very small biases – slightly ethnocentric for whatever reason. Now, let’s think about 2 different electoral systems. In the first system, all of our names are put in a lot and one person’s name is picked out of that lot. Then, she/he is told to choose between A and B. I can bet my life on this, if you make the differences between the two policies clear enough, and there is experimental evidence to back me up on this, that the person is going to choose A over B if the material benefits are good enough.

Let me change the institution slightly and let everyone put her/his votes for A and B into a ballot. Then, we will randomly pull out one chit and whatever is written on it, whether A or B, becomes the policy. In the second setting, I can bet that the policy chosen is going to be B. What is happening here is that now I know that my chances of influencing the outcome are so much lower, and so we see a lot of biases.

The second system is a lot closer to what we’re seeing in the real world where elections have become this massive space in which basically what is getting aggregated are our biases. The aggregation of material things that we would like to see- better schools, better hospitals or better roads – are not getting aggregated. Politicians and political parties understand that, and so election campaigns operate accordingly – when do you hear about better education and healthcare in election campaigns? It’s all about playing into biases. And this process has become a lot easier now, because of technology and social media. I mean, Trump just needed a Twitter account to win an election! I’m personally very disillusioned by this process because what we’re seeing is a fundamental aggregation of biases. I don’t know how we can change that yet, but it’s just not working.

ER: When it comes to notions such as biases or something that you have inherited, can you create a mathematical model that incorporates and satisfactorily explains the phenomenon?

AB: Well, it’s hard… What is the defining characteristic of a mathematical model? Its consistency. We can agree or disagree with the model, but consistency makes it easier to model. Things which are volatile and show no consistency are hard to model. So, what we say is these behavioural tendencies are deviations from the perfectly consistent model, but they still retain some consistency and that’s exactly what we use. For instance, take the theory of loss aversion – the most important idea is the reference point. Loss aversion is essentially an idea of reference dependency. But prospect theory itself has nothing to say about what the reference point is and in economics we have no nice theories about reference dependency because the reference point is such a volatile thing. So generally anything that lacks consistency is hard to model.

GZ: How do you think the theory of loss aversion can be applied to stock markets where people have to take that dicey step towards investing?

AB: One of the things that is a great puzzle in finance is stock market participation – why do so few people participate in stock markets? This is a relevant question especially if you want to think about reasons for inequality. Picketty’s great observation about why inequality is rising was precise because rates of growth of capital income are very much higher for some people than for others. If you look at people who don’t even have a mutual fund or don’t take part in the stock market, you don’t have any form of capital income – not only is your base lower, but the rate of growth is lower because capital is growing faster than anything else. This is a puzzle because the standard model clearly says that you should always hold a position on assets. One thing that we can model using loss aversion theory is the idea of why people might not hold a position where the standard model might not give you an answer. More generally, loss aversion is a big idea in terms of how people hold onto certain stocks precisely because they don’t want to realize the loss even when the stocks are tanking, which is a behavioural tendency that influences positions.

ER: Could you recommend some books for students getting started with Behavioural Economics?

AB: ‘Thinking Fast and Slow’ by Daniel Kahneman is a very nice read. ‘The Friendship that Changed the World’ is not a behavioural economics book but talks about the friendship between Amos Tversky and Daniel Kahneman and how that whole project evolved – those are good things to get started with. If you want more than chatty introductions and want to get into behavioural economics, the ‘Round-table Series in Economics’ has a good set of survey articles which give a broad picture of what the research findings are. They cover all the big areas in Behavioural Economics, for example, time preferences. ‘Misbehaving : The Making of Behavioural Economics’ by Thaler can get students excited in the area, and then, of course, you can follow it up with something like the ‘Round-table Series’, there is one in Behavioural Economics (edited by Camerer) as well as Behavioural finance (edited by Thaler) which can give you a good flavour of a research paper.

Read the article on the lecture here.