Prachi Deuskar: You have been in academia for a long time and now you are on the policy side. What do you think is a good way for the research community and policy to interface?
Viral Acharya: Central banking is common ground for both academics and policy makers to interact in a policy setting. Central banks tend to be research focussed and research intensive. They absorb research, they put out research and they benefit from research. In developed countries, increasingly, central bank governors tend to be people with academic training. The Reserve Bank of India is also moving in this direction at least since the last several appointments.
I believe that banking itself is a great area to pursue research if you want, in the end, to do policy work, even if it is not at a central bank. Even if you have an economic system where markets are developed, banks invariably play an important role as money centres.
Take the global financial crisis, for example. There was a set of academics who believed that banks are not important anymore. It was actually hard to get jobs if you were doing research on banking, and especially if you were doing research on banking regulation. They thought that (banking) research was not that relevant, that it was all about markets. But we saw that when banks are not doing well, they can choke up credit and money flows all across the economy.
I might be biased given my own research area and where I am right now, but I do think that at least one very clear path to having policy impact in a research career, and possibly even taking up a policy role, goes through banking and central banking. Central banking themes apply to monetary policy, financial stability and financial markets development. They include money markets, government securities markets, capital markets and foreign exchange markets and the whole suite of bank supervision, bank regulation and related issues.
When you are doing central banking work, you are of course interacting with banks and taking decisions that affect them. But in the end, you are really thinking about how it is affecting the real sector of the economy. How is it affecting the lay person’s cost of credit? Does the lay person have access to financial avenues to manage his/her savings better over a lifetime? Is it affecting corporations? If yes, then does it translate directly into investment activity, jobs creation and so on? There is something quite rewarding and fulfilling about being able to think about the lay person’s concerns.
When you are doing central banking work, you are of course interacting with banks and taking decisions that affect them. But in the end, you are really thinking about how it is affecting the real sector of the economy.
The other way to have a big impact is, of course, to be an academic who works in markets, with individual financial institutions like banks, insurance companies, hedge funds etc. But your focus there may be too small. While you could be having a huge impact on that one individual entity, does it translate collectively into something bigger? You may not necessarily connect the dots all the way to reaching the lay person or the firms in the economy.
I like the central banking canvas because it is very big. It can be a bit daunting and challenging at times but once you work in that area for a while, you start seeing the bigger picture. That backdrop itself is a inspiring, in my view.
In economics, very often they say that researchers are evaluated for having made contributions in three sub-fields.
Sometimes in the finance profession, however, we tend to be a bit narrow. People tend to be focused on their small sub-field over their entire career. While there are clear gains from specialisation, I think that maybe we could learn a little bit from the slightly broader thinking in economics. And if you are working in banking, and central banking, you automatically end up covering a wide spectrum because banks are at the centre of a lot of these issues. That is another reason why people who are working in the banking and central banking areas may, in the end, actually branch out into different areas. That breadth, along with the depth you gain over a seven to eight-year period within each subfield, can actually lend a rich dimension to academic experience and thinking over time.
If you want to list maybe two or three big questions that are relevant to central banking, and on which research is needed, what would those be?
Just to take a few examples, it would be great to understand how monetary policy gets transmitted into the real sector of the Indian economy. What role is played by the structure of our banking system, the fact that it is a mix of public sector and private sector banks? Over time, as non-banking finance entities have evolved, how has this transmission of monetary policy to the real sector changed? For example, we have a very strong presence of non-banking finance companies in housing finance but less so in areas like large corporate lending.
This is a very big question for monetary policy and central banking in general: When you change interest rates, is what you think going to happen really happening? If not, why is it not happening? If it does not have the desired impact, what may be the reason? This would be a great research area at the intersection of macroeconomics and finance. I think the RBI, ISB and other research organisations can all collaboratively work together to bring new research to bear on these questions.
A second area that has some scope for big data work in economics is thinking about e-commerce penetration. Real time or at least daily data could give us a better signal about how inflation is actually behaving with respect to different commodities. How can these digital transactions be measured more effectively in our GDP estimates? What sort of consumption patterns are shaping up? These issues play a particularly big role when you have structural changes like demonetisation or the rollout of the GST. In an uncertain phase, past episodes do not give you clear signals as to what to expect. Real time data becomes very important. Monthly or quarterly data is okay only when the system is more stationary.
There is a lot of work to do in this space as well. Some of it would have to employ the powerful computing and data techniques out there. About five and ten years from now, it would be nice to start thinking about how we might be generating this data and working with it.
The third question is one that people have been talking a lot about of late. The NITI Aayog has put it upfront and centre as well. The question is understanding the employment situation and measuring employment better. The question is of the first order of importance especially given the kind of demographics we have, the potential young labour force. Research has to address the issue of our large informal economy, which itself might have changed given the big structural changes that have taken place. How do you construct indices to deal with the changing structure and the extent of formalisation in the economy?
These are maybe two or three of the big issues that would be great to work on.
Technology is bringing about a lot of changes in financial services. What are your thoughts on opportunities for central banking through these changes? There will be many, many challenges as well, because regulation typically plays catch-up.
Yes, absolutely. In some sense, digitisation and technology are shaking up the basic assumptions and foundations of a lot of stuff that central banks and other firms in the economy are doing. The big issue to me here seems to be about understanding the implications of technology for productivity, jobs and also for prices of goods and services. How would inflation be impacted given the logistics and distribution capacity that many of these large e-commerce companies have? What are the implications of artificial intelligence being built into various kinds of services that earlier were about repetitive tasks? Now they are saying that even thinking is a repetitive task; being intelligent is a repetitive activity.
Organisms are algorithms.
Exactly. But what I worry about is skilling. There will be jobs with artificial intelligence, but will we have enough training and skilling of large human populations to actually fit into these jobs?
What was beautiful about the jobs involving repetitive tasks was this: even if you managed to succeed in pushing everyone through a skilling system or a good education system, clearly there were differences in skills and quality when people came out of the system. But it seemed that for a while, we had a good job allocation system to deal with this. Society implicitly had an insurance mechanism for those who did not, for whatever reason, come out with the best skills, even when put through a skilling programme.
Now it seems as though that gap between what you can derive from a job when you don’t come out at the top versus when you are at the top is just widening. I think that this phenomenon is also a contributor to income inequality.
Clearly we have to move forward on our education system and skilling to become much more vocational. But, to be able to tailor education constantly for the kind of jobs that exist begs the question of how quickly we need to move the education system. If all the jobs require an extremely high level of competence, are we even currently thinking about the speed at which we have to innovate in education?
There is immense progress being made on online delivery. Children today are at a vastly different level of aptitude, knowledge, awareness and exposure to issues at a much earlier age than we were 20 years back. But as I ponder over these kinds of questions, I don’t have very clear solutions or even clarity.
On the other hand, when I listen to someone like Nandan Nilekani talk about these issues, I always walk out with a very positive feeling about whatever is happening in data and information. We will see how it all plays out. Certainly it seems a big question for academic and educational institutions, researchers as also for central banks.
With demonetisation, there is certainly a big push for digitisation of payments in the economy with Aadhaar and other e-identifiers. There is now tremendous scope for linking things that were not linked before. So in some ways, I think it is an exciting time. Digitisation and technology are now basically starting to touch economics, research, and data management in a very direct way. We have to be ready to adapt to that. As I was saying, thinking about e-commerce data to understand inflation, consumption etc. might actually be a good start. Because maybe in five years’ time, that is how we will be looking at data.