In this episode of How India’s Economy Works, journalist and author Puja Mehra speaks with economists Arjun Jayadev and Amit Basole, authors of the CSIE working paper India's Labour Productivity Puzzle, about a troubling trend beneath India’s headline growth numbers: a sharp slowdown in labour productivity since 2017.

India remains one of the world’s fastest-growing major economies, employment levels have risen, and female labour force participation has increased. Yet, according to their latest research, workers today are producing far less than they would have if earlier productivity trends had continued. The conversation explores why this matters for wages, living standards, investment, and the broader health of the economy.

They discuss the rise of surplus labour, the difference between employment and productive jobs, and why much of the recent increase in work — especially for women — may reflect economic distress rather than opportunity. The episode also examines weak private investment, manufacturing stagnation, structural transformation, the limits of formalisation, and whether policies like infrastructure spending, digitalisation, and production-linked incentives are truly improving productivity.

The discussion raises a deeper question: can India sustain high growth if output per worker remains stagnant? Tune in for insights on why India’s growth story may be masking a deeper productivity crisis — and what it means for jobs, wages, and the future of the economy

—

TRANSCRIPT

NOTE: This transcript is done by a machine. Human eyes have gone through the script but there might still be errors in some of the text, so please refer to the audio in case you need to clarify any part. If you want to get in touch regarding any feedback, you can drop us a message on feedback@thecore.in.

Puja Mehra: Arjun and Amit, thank you so much for coming to the show again. Both of you have come on in separate episodes, but it's great to have you together in one episode.

Arjun Jayadev: Thanks for inviting us.

Amit Basole: Thanks.

Puja Mehra: Arjun, I want to start by asking you about this new paper that you and Amit have written together about labour productivity in the economy. You've written that since 2018, employment levels have increased in India. Female labour force participation has risen.

India has emerged from the COVID-19 pandemic as the fastest-growing large economy in the world, combining robust output growth with relatively low inflation, which is something that the Ministry of Finance and several other commentators have talked about. And so, in a macroeconomic sense, India's performance appears enviable. However, this sustained output growth has been accompanied by a dramatic slowdown in labour productivity growth, and this, in fact, is a break from trend.

So, what appears to be a happy situation on the macro front seems worrying to you when you look at it closely. Tell us more about what is so worrying about this. What does this slowdown in labour productivity mean?

Arjun Jayadev: All right, yeah. So, thanks very much for having us, Puja. So, yes, this is a paper that Amit and I have written, and it's up on the Centre for the Study of the Indian Economy website.

I'm sure you'll link to it. You know, whenever you think about an economy which is growing and transforming, an economy which is on a healthy path, there are two things which tend to happen for a developing economy. The first is you have output per capita growing, which we're doing, and you have structural transformation happening as well, right, which is you have people moving from surplus sectors into what's sometimes called the modern sector.

Now, what both of these should do over time is to actually increase the output per worker or the growth rate of output per worker. In fact, that is the single most important variable in the long term to understand changes in the output per capita, which is GDP growth per capita. It basically measures how much value each additional worker generates.

Now, here is the interesting thing, right, when there's labour productivity growth, output per worker rises, and when there are booms, it rises really fast. And if India is going through a boom, therefore, you should expect, you know, a real boom, therefore, also of labour productivity. But once we start to look at this particular variable, since around the year 2000, we looked at the growth rate of labour productivity, let's say the trend growth rate from the early 2000s to now, and it's been on a steady upward path, productivity, and the annual growth rate's about 6%.

But about 2017, something really strange happens. It basically flattens and dips sharply relative to the earlier trend. And if you look at a counterfactual, another way of looking at it, it says that if our labour productivity had grown post-2017, as it had in the years prior, workers would be producing 40% more than they do now.

Now, we have to think a little bit about what this also implies. It means that 40% more output per worker, which would go to potentially higher wages, some of that could also be potentially higher profits, it depends on how productivity is split up. But in any case, it at least helps explain some of the kind of things that we're seeing right now, which is much lower wage growth than you might expect from a booming economy and so on.

So this we think is actually a very key variable to understanding the health of the economy.

Puja Mehra: So you're essentially saying that every worker's producing less than what they were earlier?

Arjun Jayadev: Compared to the 2017 rate, it's stagnated. So it's periods when it's dipped, it's about the same as around 2017. It used to grow at about 6%.

It's been stagnant since about 2017, which is why if you compound how much it would have been had output per worker grown post-2017 like it did before, we're about 40% shortfall.

Puja Mehra: So GDP would have been 40% more?

Arjun Jayadev: Output per worker would have been 40% more, output per capita, because you're actually the entire workforce is supposed to buy. But certainly output per capita would also have risen.

Puja Mehra: Amit, let me come to you. The productivity slowdown post-2017 seems non-trivial and pervasive across sectors. Do we have any idea of what has caused this?

What is the reason this has happened? It does come a year after demonetisation, the year in which we know investments had collapsed. So is there some link with that?

Because although there was later on COVID, but COVID happened in all countries, all countries were affected. So what is the reason for what we are seeing?

Amit Basole: Right. So just to start with that international bit, one of the things we do try to do in the paper is understand India's specificity with respect to this phenomenon by putting it in the context of other developing countries. So we use the World Bank data on value added per worker and try to see among the large developing economies, is India really standing out when it comes to this kind of productivity stagnation that we were talking about?

And that does seem to be the case. And this is important to establish the fact that we are somehow different than the average developing country, because as you said, COVID was a very large international shock. And a lot of other things that have happened in the economy in the last few years are of that kind, which have affected every country, whether that is deglobalization and the uncertainty around tariffs, the wars that are going on, the pandemic, of course, all of these are global events likely to affect every country.

And we want to see if there's anything over and above this that also India seems to be having. And it does seem to be the case that there's a lot of data in the paper to show that India's productivity stagnation is more severe than what is to be expected on average. Now, why might that be the case?

Certainly, factors internal to India then come to the fore, or maybe an interaction between Indian factors and the global factors. You may remember, and your readers may also recall, that before the pandemic, there was quite a bit of discussion around a slowdown in the Indian economy, which at that time was a very severe slowdown. I remember something like five or seven consecutive quarters in which growth had been slowing down in the years 18-19.

And several explanations were offered, including the twin balance sheet problem, the accumulated NPAs, and then what had been done to take care of that problem, etc. And you yourself have written quite a bit about all of this as well. So there were certain, I think, factors there, which had already started to go wrong, whether that you trace it back to the boom years of the first decade, and the accumulated bad debt, and that kind of story could be told, and that feeds back into weak investment.

There's, of course, the huge international element of China, and then the vacating of China, and which countries came up in its place, and why India was not able to capitalise on that. Maybe we can come back to that in the later part of the programme, which is the big question, which is, why haven't we been competitive with respect to labour-intensive industrialisation and so forth, where a lot of good jobs should come from? So there's legacy problems, infrastructure competitiveness issues, there's more medium-term issues like the debt overhang, or other things like that, demonetisation, GST, any number of those kinds of things.

A summary of it seems to be that once the COVID bounce-back was over, you know, we had a purely base effect bounce-back in terms of productivity and so on, there were still unresolved problems in the Indian economy that had not gone away, and that then sort of come back to get you back into a kind of largely stagnant productivity space, and it probably comes back in a big way to weak private investment, which you had also alluded to.

So maybe we can go back to that question of what we know about behaviour of private investment, why it might be weak, and so on. One last thing I'll say, just what Arjun had said in the beginning about structural transformation, this is very important for readers or listeners to appreciate that when we talk about robust employment generation in this period, the bulk of it has been in the surplus labour sector. So India being a dual economy, it's very crucial to distinguish are people working in self-employment, in agriculture, in other words, in employment that is created by themselves.

Nobody needs to employ them, or is it a result of labour demand, which means somebody is investing money and hiring workers. That's a very different situation. So we use the word jobs to describe that, and any employment can be created which need not be jobs.

So a lot of recent employment, particularly for women, has been of the kind where women have gone in to earn some livelihood, but that is not the result of creation of labour demand necessarily, which means in turn it is not because of investment. So investment picking up is what leads to jobs, and we need to understand whether or why that has not happened and so forth.

Arjun Jayadev: Maybe you can just jump in and say one thing, Puja, because it might be a little confusing to your listeners. How is it possible that output per capita is rising and output per worker is stagnant? And the answer is quite straightforward.

You have more workers. So you'll have more workers. If you have more workers producing the same amount of output, you're going to have higher output per capita, but the labour productivity is going to stall.

That's really the kind of macro variable that we will see and which Amit has alluded to, the mechanisms which are leading to it. So we have a lot more people, quote unquote, being a part of the labour force and being employed, but who are working at much lower output per worker than you would have imagined and would have wanted?

Puja Mehra: Typically, we see this on the farms. When more and more people from a family begins to spend time on the farm, they're not necessarily working because they're not altogether producing much more than what they were earlier, but just more people go and spend time there.

Arjun Jayadev: I mean, that's there, but I also think there's this very large surplus sector in the urban economy as well. So people are entering employment, being paid very little, still being employed and therefore producing very little as well. So I think it's not just a story about agriculture.

Puja Mehra: In fact, you say that we are seeing this in all sectors, including manufacturing and services.

Amit Basole: That's right, exactly. So we can talk about a couple of different things. One is not a puzzle.

It is not a puzzle to find productivity stagnating in surplus labour sectors. That can happen when the output of that sector does not grow, commensurate with employment in that sector. And employment in those surplus labour sectors, as you said, is an easy entry thing.

Simply speaking, anybody can open up a street shop selling mobile covers or a fruit cart or even farm work, any detail sector, agricultural sector, even places like construction. Many of these sectors of the Indian economy are surplus labour sectors, meaning there's easy entry for work. That doesn't mean that output will grow to keep pace with the people working in it.

And if that happens, if output doesn't keep space, then productivity, of course, falls.

Puja Mehra: And a lot of this kind of employment is what is showing up in the bump up in employment numbers that we are seeing, that gets reported.

Amit Basole: In particular for women, and we show this in the paper too, gender disaggregated, that for women, there has been a huge increase in surplus labour type of employment, particularly agriculture, but also other sectors. For men, there has been that, but also some good jobs. So it's a more complicated story.

And we get into the details in the paper. We don't want to paint a picture that the last few years have created no productive jobs. They have.

But the aggregate story is that the growth of surplus sector employment is faster than the growth of productive employment, such that overall, the Indian economy has seen a stagnation in productivity. Now, one other thing to see is a lot of listeners might be thinking, you know, people were criticising jobless growth for many years. And now we have growth full of jobs.

What's the problem? Seems like a good situation to be in. So two things to point out.

One, these are not jobs in the sense that nobody's employing you for productive work. It's mostly self-employment. But more importantly, this kind of extensive margin type of growth, you know, what economists would call, which is additional workers, as Arjun was saying, has limits.

Long term sustainable growth is productivity driven growth. This can get you only so far, you know, the additional worker effect. So if we want sustainable growth, there is really no option but to grow productivity.

One last point is that that does not mean wages will rise. I think there's a lot of discourse in India on why wages aren't rising and so forth. Now, of course, our paper says that productivity itself has not risen.

So that's a further problem to wages rising. But in a dual economy, even if productivity were to rise, wages may not rise because there's surplus labour. But it is still good for the economy because you can get profits reinvested, more jobs created.

And eventually, like in the Chinese case and other cases, you will exhaust your surplus labour and then wages will start rising. But for all of that, the first thing is sustained productivity growth, which is the problem.

Puja Mehra: But I still find it a little puzzling. How is it that in the manufacturing sector, labour productivity is not increased?

Arjun Jayadev: I have to admit we don't have a very good answer to this either. There are, of course, the standard kind of things, you know, regulations, this thing, that thing. I don't know if there is a very clear answer because it's not just in particular parts of manufacturing and so on.

It's across the board. And the interesting thing is if you look at the ASI, you actually find that productivity growth is slowing down even among the very large firms, over 100 workers. And, you know, it shouldn't really be something that's possible or really likely by standard theory.

Why would you employ people? Now, one hypothesis, which our colleague Zico Dasgupta has mentioned, we still have to try to assess is whether, you know, it doesn't matter if labour productivity is not rising fast, if wages are actually rising even slower, in which case there's still enough profitability to be made. So these are the kind of things we have to look at.

But I don't think we have an answer, a very clear answer.

Puja Mehra: Yeah. Amit, do you want to try explain how is it that the large firms you're saying, that the firms that have more than 100 employees, they're also not seeing labour productivity growth. This is the formal sector, right?

Amit Basole: Right. So I think what Arjun was alluding to was, and this is very much a hypothesis that needs to be tested. And I want to stress that this is not in the paper, but in the discussions around the paper, this point has come up that from the point of view of an employer, the critical thing is unit labour costs.

So when you produce something, I mean, unit cost of anything, but since we're talking about labour productivity, it is unit labour costs. How much am I able to produce in value per rupee that I'm spending on workers? Now, here there are two factors.

There's the number of workers, but also what I'm paying each worker and then what each worker is producing. So if wages stagnate or wages don't rise, and one of the things that we saw around the recent discourse, for example, around the labour protests and so on, was that we found examples. Now, of course, this is anecdotal that nominal wages had also not risen for many years in many places.

So which means in real terms, wages are falling. So when you have very weak wage growth or even negative wage growth, real wage growth, then what Arjun was saying earlier, that even if productivity growth is not strong, from an employer's point of view, the unit labour costs are still in check or they are still competitive. Another way to look at it is, think of it from the point of view of an entrepreneur who is facing several challenges, bottlenecks, uncertainties and so on.

So they produce some amount of value. To remain competitive, what they do is suppress wages, because that's the easiest thing available to you to remain competitive with unit labour costs. In the counterfactual, if you had a productive firm, less uncertainty, better infrastructure, whatever other many other factors people talk about, then you can afford to have higher wages and still have low unit costs, because your productivity is commensurately higher.

So we may be stuck right now in some sort of a low productivity, low wage type of an equilibrium, out of which we need to break. And notice none of this is really explaining why any of it happened. This is all mostly hand-waving around the phenomenon.

We need a lot more work.

Puja Mehra: But what is the link between labour productivity and wages?

Arjun Jayadev: Okay, so typically in an economy, let's say a single sector economy, you should expect that as labour productivity rises, wages rise. Of course, it depends on the bargaining power between capital and labour. But there is a belief that at some point or the other, you will run out of people who are willing to work at lower wages and wages will go up.

And that's actually been the case in country after country. Lewis called this the turning point in a separate discussion about structural transformation. But this point where, you know, actually at some point or the other, productivity gains will go to workers is something that you see in many places.

Not always, some place, actually often there's a wage repression in the first part of growth. Now, as Amit was saying earlier, it's not necessary for in the aggregate, if you have labour productivity growth to have wages grow, because you might have still a very large surplus that you have to get through. It's a necessary condition, not a sufficient condition for wages to grow in the medium to long term.

Puja Mehra: The rise in women's participation seems more like this just work than the great cause for celebration that we've been seeing past few years about the increase.

Amit Basole: I suppose from one point of view, one could say that it's good to see more economic participation of women if that results in more incomes generated by women under the control of women. There are many potential benefits over there on their own terms. But what we are discussing here today, yes, it's not the kind of increase in employment that you would want to see.

So just so that readers or listeners have a clear picture in their mind, you think about the electronics industry or the smartphone industry where the production incentive scheme has been putting in a lot of resources. And we've seen some pretty big increases in output and employment in those sectors, including a lot of women working in those sectors because they tend to be women intensive. So those are high productivity, regular wage jobs for women.

We might have some criticisms of their working conditions but in the limited context that we are talking about, these are good, high quality, high productivity jobs. The counterfactual is the woman who is taking in tailoring work at home or tuitions at home, any number of such activities to bolster household incomes, which haven't gone up in a while. So the husband's earnings haven't risen but costs have risen, etc.

So I'm taking on more work to supplement household incomes. Again, nothing wrong with it on its own terms. But as a structural issue, it is pointing to the fact that those incomes that were earlier enough are not enough anymore.

And there are no other jobs that potentially the woman can do. Now that could be a factor of not those jobs not being created. That's a labour demand issue and an investment issue.

But of course, we haven't talked here and the paper doesn't talk about this because that's not the main point. You could also have supply side problems in women's work, which is that the jobs exist but women are not mobile, they can't get out, norms are in the way, etc. So there is big literature on that too.

For whatever reason, we've seen an increase in women's employment of the kind which is self-employment and appears to be a distress type of employment.

Arjun Jayadev: So I want to say a little coda to this. As you noted, India still has a large increase since 2019 of the fraction of its workforce which is in agriculture. That's driven mostly by women.

Some of the men who used to, there was actually distress and men went back into agriculture. Some of that has reversed, not entirely, but it's reversed. So we like to think of this not simply as say, you know, failure of structural transformation, but what we sometimes call an intensification of dualism, which is that something is happening in the productive modern sectors, as it were, but also there's a large amount of activity in the informal sector and agricultural sector and so on.

So you see these two poles actually intensifying rather than a transfer that we would expect.

Puja Mehra: So there is a narrative out there that says that structural reforms are taking place and that formalisation of the economy is taking place. You're saying that we have to look at all of that narrative a lot more closely and the official, this is all of your studies based on official government data. All of this data is not supporting that narrative.

Am I right? Have I got it correct?

Arjun Jayadev: No. Well, let me nuance that. I think there is some formalisation.

There's, as we said, some growth in some of these sectors. And clearly, if you look at things like EPFO and others, there is some formalisation going on. So that's what we said, you know, there's an intensification dualism.

But it's a small proportion of the overall changes which are happening in the employment. Because a large fraction is this rise in female participation in agriculture and some of the more surplus sectors.

Amit Basole: I just want to add that, you know, Puja, the question you're asking really demands a counterfactual which we don't have. So what we know is that in the last few years, the central government and many state governments have really tried a whole range of approaches to deal with this problem that we're talking about, even if they have not understood it in exactly these terms. You look at tax incentives, look at subsidies, you look at infrastructural spending, industrial policy of various kinds, protection and trade policy, deregulation and decriminalisation.

Puja Mehra: There is a scheme called the productivity link scheme, where a lot of money from the budget goes. And therefore, you know, it's a bit surprising that productivity is not increasing.

Amit Basole: So a bit of that is because the scheme is called that, but it's really sales. So I think sales linked incentive would not have been a very good name. So they went with productivity.

So the way PLI works is actually a very interesting scheme, maybe if time for a separate conversation. I'd use this as an example in my classes on industrial policy. You know, it does have things like sunset clauses and many other interesting features that our industrial policy literature says policy should have, you know, to avoid the old industrial policy mistakes.

So PLI is an interesting scheme. It has had some successes. Yeah, mobile phones, I think have done quite well in increasing the domestic value.

So, you know, let me also point out there that the critics often point to the fact that, well, it's only assembly and so on. Domestic value added has gone up. I've seen that.

And in any case, without doing assembly, you can't do anything else. So, you know, you have to enter somewhere. But my larger point is that all of these things are being tried by the government all at the same time.

But what we don't know is what would have happened if they hadn't done it. That counterfactual is missing. Would it have been worse?

We just don't know. Or is it that there is not enough coordination between the things going on? You know, there could be many.

Of course, PLI for all its virtues, the amount that has been spent is actually a very small fraction of the amount that was budgeted. I think about 10 percent or less. So it's not like all of these would also have the bite that they seem to have on paper.

So they need to carefully look at each of these things and see what effect they've had. We need really, honestly, a lot more careful empirical work on most of the things. The last seven or eight years in India, I've seen an amazing amount of policy dynamism on welfare, on industrial policy, on everything.

But because it's politics and it's policy, you know, we just don't have the time to do the counterfactuals and all of those checks, right? So academics should be doing this, going back and trying their best to disentangle these effects and see, well, what has worked, what hasn't worked. We need to understand that much better.

Puja Mehra: But I find it interesting that, you know, politicians, we've seen this whole spurt in cash transfers to women. Is that a response, you think, to this distress situation, the stress that women are going through? Or is that an incentive for more and more male workers, for instance, to stay back in these kind of jobs and not go and look out for more productive jobs?

But that would be assuming that there is demand for labour and productive jobs.

Arjun Jayadev: See, I think cash transfers are largely to do with the fact of insufficient structural transformation. I wouldn't say it's necessarily only about lower productivity. Obviously, they're linked.

If you had higher productivity, you'd have had, you know, more output and so on. But I think the cash transfer is, and the proliferation of cash transfer schemes is clearly an indication, in some sense, of a compensatory state or, you know, a state which is not able to, let's say, generate or policies which are not able to generate the kind of growth that you would see, which would allow people to have wages which are sufficient to, you know, to live decent lives. So certainly, I think that's all linked.

There's, of course, also the fact that it's much more politically expedient. It's linked to a particular person often when it's transferred. It's more visible.

Maybe there are transparency gains. So there are reasons for the cash transfer, which may be more, should we say, operational. But I think, at heart, the fact that a large swath of the Indian population is not seeing the kind of growth in wages and incomes that they would need to, to be happy, is very central to this.

Puja Mehra: And my last question is that, with so much spending on infrastructure happening, with so much digitalisation happening, with digital payments, the spread of it, the penetration of it through the economy, logically thinking a layperson would think that that should improve productivity of labour. How is even that not helpful?

Amit Basole: I guess there's a few different ways to think about it. When we talk about a lot of this digitalisation and so on, in fact, people often point to the surplus sector, ironically, to show evidence of formalisation, right? You'll see, you hear people talking about, oh, you know, even street vendors have QR codes or stuff like that.

That is okay. I mean, that just telling you that a previously informal income stream is now a formal income stream. But analytically, the question is, to put it simply, where there were 10 street vendors, are there now two larger shops?

That's structural transformation. If you have not changed that, then the same 10 vendors who were sharing in whatever income was coming earlier are doing so, it's just formally now shared, in the sense that income is formal. Or to make matters worse, where there were 10 vendors, now there are 15, because people have shown up having lost jobs in the surplus sector.

So, what output was being sold by those 10 people and demanded in that thing is now 15 people sharing in that same. So, per person, actually, it's fallen. But each person has a bank account linked to a QR code and so on.

So, that's okay for some kind of visibility on incomes, but it's really saying nothing at all about productivity. In fact, in my example, productivity fell as incomes became formal. So, we need to disentangle these two things.

On the counter side, you can have a job that is fairly informal, in the sense, maybe you get a slip or a chitty or something in a factory, but you get no PF or anything, but you're actually working in a modern factory and generating a lot of value, and you're getting cash in hand salary. It's a very informal kind of job, but much more productive. So, the productivity lens is really very different from the legal or the technical kind of formality lens.

And it's really the first one that is more important from a long-term sustainable growth perspective.

Puja Mehra: And it's something that doesn't come very naturally and easily to lay listeners. And I can see that there is so much that we still don't understand about this new phenomenon. But nevertheless, thank you to both of you for drawing our attention to it, for helping us understand what's going on in the economy and for sort of prodding us to look at more closely on some of the things that we celebrate about the economic performance in the last few years. Thank you so much.

Arjun Jayadev: Thank you, Puja.

Amit Basole: Thanks.