MC exclusive | Inequality fell because incomes declined: Arvind Panagariya

Arvind Panagariya

In his latest paper, Arvind Panagariya, Professor of Economics, Columbia University, and former VC of Niti Aayog, dispels claims of a large rise in poverty in rural and urban India post the Covid-19 pandemic. In the paper, Panagariya states that poverty fell every quarter from July–September 2020. In an exclusive conversation with Moneycontrol’s Shweta Punj, he says that studies claiming a rise in poverty and inequality are flawed. Edited Excerpts.

Startling revelations coming out in your paper, which run completely contrary to a lot of studies and research that has been coming up in the past three years. We’ve had Oxfam, we’ve had Azim Premji University saying that India is more unequal now. Wealth is very unequally distributed. However, you are saying something different. What exactly are your findings?

Okay. So, let me say that this paper about which we are having the current discussion is related to two years before Covid-19 and two years during Covid. So, 2017-18 to 2020-21. Also, I should clarify that this is the survey year. The survey year starts on July 1 and ends on June 30. So, it’s off by a quarter with respect to the fiscal year. What we find on rural poverty is that during that single quarter when the lockdown was very strict, April to June 2020, rural poverty during that quarter did actually rise. But then it resumed declining and it continued to decline in the subsequent four quarters for which we have the data.

So, for the entire 2021, even when you look by quarter, poverty, rural poverty is declining relative to the same quarter the year earlier. Urban poverty, on the other hand, does rise for a little longer. It rises first in that same quarter, April to June 2020. And then it continues to rise in the subsequent three quarters on a year-on-year basis. But then in the last quarter, for which we have the data, which is April to June 2021, even urban poverty declines relative to the same quarter a year earlier. On an annual basis, actually, you only see urban poverty rise in the year 2021. But overall poverty, national level poverty, continues to fall in both 2019-20 and 2020-21. You see inequality declining. What that is representing largely is that incomes fall a lot more at the top end of the distribution. They’re not at the bottom end of the distribution.

You referred to the Periodic Labour Force Survey. Whereas Azim Premji University, which had come out with a very strong report, was widely quoted in the media saying that inequality has gone up. They used high-frequency household survey data. So, I just want to understand these two different conclusions. What is it in the Periodic Labor Force survey data that is missing in the High-Frequency Household survey data?

The Periodic Labor Force Survey also is a continuous survey, only you can analyse it by quarter, not by month. So, it is also in some ways a high-frequency survey. The other one you can do by the month. So, it's a higher frequency for sure. Now, first of all, this is the CMIE survey. This is a private company, which does the survey (the Household survey data that Azim Premji University used). So, PLFS is also a household survey, except that PLFS is primarily devoted to looking at issues of workforce size, labour force participation, the unemployment rate, etc. It tracks work-related characteristics more. It just also happens to collect data on expenditure. And this is the survey done by the National Survey Office, which is part of the Ministry of Statistics and Programme Implementation.

The sample size is roughly the same. They’re both very large surveys. So, there is no issue. We can come to discuss what the differences are and why I put a greater stock on PLFS in a minute. But first, I want to clarify where I very much disagree actually with the way the Premji University report analyses those data. Now, this CMIE survey is an expenditure survey. It (the data) has also been used by another paper, by three co-authors: Gupta, Malani and Woda. So, Gupta, Malani and Woda used exactly the same survey as the Premji University survey. They actually find that inequality fell, and in urban areas, particularly, inequality fell dramatically. Now, the question arises: how come these authors are using the exact same survey as Premji University and Premji University and comes to a very different conclusion?

I mean, so you know this is what we call in our academic language, ‘where is the dead body buried when the murder happens’. So, that is the clue. You have to find the answer to that question. Now, it’s very odd when economists traditionally for a long time, the way we measure inequality is that, look, the household survey gives you what is the income level in the bottom 10 percent, next 10 percent, next 10 percent. So, you got the household data and all you need to do is calculate basically based on those household data and measure inequality, which we all call the Gini coefficient. We don’t need to go into the technicalities of that. But basically, you get the per capita income data and you measure it, which is exactly what the Gupta, Malani and Woda paper does. That’s what they do and they find that inequality actually false. It is very odd that Premji University reaches a very, very different conclusion. In fact, not only did they reach the conclusion that inequality rose, but it rose massively.

Their data says that household incomes went down by 16 percent to 19 percent, and poverty went up exponentially…

Yes. And based on that, the CNN Headline was, ‘India’s billionaires get richer, while Coronavirus pushes millions of vulnerable people into poverty.’ Based on the Premji University study.

Poverty more than doubled during the lockdown. It was 50 percent to 80 percent higher in the post-lockdown period. That’s the highlighted bit from that paper. You’ve said that they devised their own model to come to these conclusions…

Right. It is a long report, so I didn’t scan every word of it. I sort of did not find that particular study persuasive at all.

So, this is where Dr. Panagariya, again, India has been on a different track than the rest of the world, because inequality did go up in other parts of the world, but according to your paper in India, inequality actually declined.

So, frankly, I do not know what exactly happened to inequality in the other parts of the world. I need to check that. In a way, my instinct would be that actually as far as inequality is concerned, very likely it should have fallen in other countries also, particularly in a country like the United States, where also massive transfers of income got made which were not to the top 5 percent at all. And there is one simple argument you see, that when incomes rise rapidly, this is during growth, right? Intuitively, you would expect actually that it generates billionaires because somebody has to create the wealth. And let’s say even if they keep 5 percent of their wealth for themselves and 95 percent goes to others, that will still create a billionaire. It’ll create a billionaire, because you have many billion dollars’ worth of wealth being created. So, in fact, when incomes are rising, you would expect inequality to actually go up. It doesn’t always happen. But it depends on what measure of inequality you’re looking at. But intuitively, it would seem in an economy that’s rising at 8 percent, 9 percent, 10 percent — you will see that inequality might be rising. You are reversing that process. The whole GDP during the year 2021 fell by 6.6 percent in India.

Yeah. So that was the year of negative growth (contraction) also.

Yeah. So, if the incomes are declining that sharply, where are they going to decline more? They will decline more at the top end.

Now, when I go back to that summer, that April-June 2020 summer, we know there was a mass migrant exodus that happened when people went back to their villages. And according to your research, you’re saying that rural poverty saw a modest rise. How’s that possible? So, these people probably lost their jobs around that time when they went back? Are you saying that the cash transfers that were taking place, the food grain schemes, that’s what provided the cushion?

Okay. So there are multiple factors at work here for the rural poor. Now, remember that the general intuition had been that somehow when incomes fall, it must be the poor who get hit hardest. But in this case, my intuition was different for the reason that the rural economy had continued to do quite well. So, if you check the agricultural growth, in the 2019-20 fiscal year, only the last 10 days of that year are part of COVID. But 2019-20 incomes, of course, continued to have an impact on what people were able to consume in the following year. I mean, it does help to have it. That growth in agriculture was 5.5 percent. Now, remember our trend, if you take the last 30 years’ period, the trend of agricultural growth is not more than 3 percent.

At best, it’s 3 percent. Now, 5.5 percent for agriculture is exceptional. Then, the following year 2020-21, again the fiscal year, also 3 percent, which is certainly better than the trend growth of about 3 percent or less. So, agriculture actually did extremely well. We all remember the reports coming out. They were tracking how sowing activity is progressing. And I remember that soon after Covid, reports were that sowing activity was actually excellent and sowing was about 5 percentage points more than it had been in the previous year. I remember all this. So, agriculture did well. So that’s clearly one factor. But the social programmes of the government also helped. So, for example NREGA, I think they reacted very quickly. It is a demand-driven programme. And so, even though it was only the last 10 days of 2019-20, which was Covid, the expenditure on NREGA did rise by about Rs 100 billion. So, it was Rs 618 billion in fiscal year 2018-19, and it rose to Rs 717 billion in 2019-20. So, there’s already Rs 100 billion expansion in the NREGA that year.

The following year, 2020-21, which is the year we find almost no reduction in poverty, at least in the survey year, 1.1 trillion rupees, that was a big jump, about Rs 400 billion jump in NREGA expenditure. So, that was also obviously rural. That’s a rural scheme. The government also very quickly moved to accelerate cash transfers, you know that Rs 6,000 are transferred every year under PM-Kisan. So, the government had begun to advance those payments sooner around the lockdown time. So, at least liquidity-wise, that was helpful, obviously, for some of the families. But much more important, I would say, is the Prime Minister’s Garib Kalyan Ann Yojana (PMGKAY), which gave another 5 kg per person per month free foodgrain to 75 percent of households in rural areas and 50 percent of households in urban areas.

Now this survey, luckily, the way it asks the question, when asking what was your expenditure, the households are also asked to include in it imputed value of anything that did not go through a market transaction. So, it includes the expenditure on any food grains that they may have produced themselves on their own farm, imputed value of what they received from the government. So, this 5 kg also gets included. So, it gets included in the expenditure and so the expenditure data, therefore, captures the effect of that programme. So, these are the factors I think that kept rural poverty going down after that strict lockdown.

Right. So, according to your paper, rural India has fared better than urban India. What were the challenges that urban India faced? And has urban India recovered from them?

Okay. So now urban India was a slightly different story because all the contact intensive activities, where obviously COVID had a much greater impact, were sort of primarily urban activities. Look at, for example, transportation, travel, hotels, that particular sector, that’s urban, that’s contract intensive. Construction. Again, not exclusively urban, but predominantly urban construction. And even a lot of manufacturing, particularly formal sector manufacturing, which obviously was more impacted. If it’s rural areas, you’re running your small little shop, you can still continue to, whatever it is, if it’s a carpenter or if it is ironsmith, you know, you can carry on that activity, because the enterprise is just your family in many cases.

But the large part of the urban economic activity is again manufacturing activity and you remember how Rajiv Bajaj, the two-wheelers that he produces, he was actually… I remember that Rajiv Bajaj was very upset about this, that the government had imposed a strict lockdown. He was very opposed to it, which was quite remarkable because he was the sole one actually I remember who very clearly came out. The point is that manufacturing also was asymmetrically impacted in the urban areas. So, these were challenges. But now, we’re coming out of lockdown. So, urban activities certainly are coming back and we will see. Very soon, I think the data will become available and we will try to see what happens in 2021-22. Those data we have not analysed yet.

You mentioned that Bihar, surprisingly, did not see more people going into poverty, whereas states like Rajasthan and Madhya Pradesh saw poverty going up and inequality increasing. Was it because of the state schemes that probably were not targeted very well? What were the reasons there?

Yeah. So, once you get to the state, one needs to know a lot more information than I have. Some states are very clear, right? I mean, take Maharashtra, for example. Maharashtra clearly the COVID impact was much more severe and very much concentrated in Mumbai and Pune, and it was prolonged. I remember this. And that’s where you see the urban inequality or urban poverty actually rise most dramatically. There, it goes up in my data, in my estimates from 7.6 percent in 2019-22 to 29.3 percent, so it’s more than a 10-percentage-point increase in urban poverty in Maharashtra. Karnataka was another place, you again see a fairly large increase in poverty there. Now, as you get to the state level, you do get some anomalous results and it’s hard for me to speculate why. I mean, one of them was West Bengal where in my data, I see no increase actually. Poverty is declining, both rural and urban, and overall, even in 2020-21.

Which is fascinating. So is it because these states fought COVID better? Or is it because they spent a lot more?

Yeah. I mean, I can’t say. Like, for example, in West Bengal’s case, I mean, ultimately, rural, everywhere generally I think it’s a rare case where rural poverty rises. But why did urban poverty not rise in West Bengal? The only thing I can speculate is that perhaps urban economic activity, which is context intensive, is not that concentrated. I mean, that’s speculation. I do not know. That’s pure speculation.

India wears the infamous crown of being one of the most unequal countries. It’s something that’s constantly thrown at us. Are we slightly more equal now? Have we improved on that?

I don’t think we should accept that. I don’t accept that we are one of the most unequal. We are not.

The Oxfams of the world love to comment on us.

Yeah. Because the Oxfams of the world dislike billionaires. So, if we are looking at that kind of measure of inequality. See, one of the problems with the inequality issue, in my mind, always is that there are so many different measures. Hence, there is no question. So, measures of inequality would inevitably rise if you’re growing fast. I mean you give me a single country, which has grown 6 percent or 7 percent a year on a sustained basis, and does not experience increased urban-rural inequality. It has got to happen, because wealth gets created in urban centres. Even if it gets created in rural centres, by the time enough wealth is accumulated, they become urban. I mean, look at Shenzhen. It was a bunch of fishing villages, 300,000 population. Today, it is the most urban spot on the face of the Earth. And it’s about something like 13 million population and per capita income of about $25,000 to $30,000 per person. So, the point is that some form of inequality will always rise if you’re growing rapidly.

The common measure we use is Gini coefficient. And if we are looking at that, then we are not certainly equal. …you have mentioned Oxfam. I didn’t comment around that time, because basically my mind ignores whatever Oxfam says because it is so driven by its own preconceived notions. Now, how in the world are they saying that more billionaires got created during COVID? I mean, just think about it — when economic activity is declining, in the 2020-21 fiscal year, GDP fell by 6.6 percent. When such a large decline in GDP happens, where is it going to happen? It is going to happen at the top end.

There’s also a report by the State Bank of India, which said that inequality has declined. And there was also a report by the National Bureau of Economic Research, which has…

That’s the one. That’s the Gupta, Malani and Woda report. And they are using exactly the same data set that Premji University does. So, now like, for example, it’s very intuitive that during COVID, we all stopped going to restaurants, meaning people who had higher incomes, they stopped going to restaurants, they stopped taking vacations. So, the expenditure of those who were at the top end of the income distribution fell a lot more than the expenditure of those at the lower end of the income distribution. They’re basically having survival consumption: food, shelter, clothing. I mean that is what they are consuming, and that expenditure fell a lot less. So inequality, particularly because we are measuring inequality by expenditure here, I mean expenditure-wise, inevitably inequality intuitively ought to go down when there is such a sharp decline, and especially when you’re getting hit on contact-intensive activities. If incomes are rising and inequality falls, that’s a great thing. But in this case, for 2019-20…

Incomes are coming down, which is reducing inequality.

Yes, right. So that’s not a good way to achieve declining inequality.