Nisha Vernekar on Night-shift Reforms and Women's Work in India

SHRUTI RAJAGOPALAN: Welcome to Ideas of India, a podcast where we examine academic ideas that can propel India forward. My name is Shruti Rajagopalan, and this is the 2026 job market series, where I speak with young scholars entering the academic job market about their latest research on India. 

I spoke with Nisha Vernekar who is a PhD Candidate in Economics at University of Duisburg-Essen, Germany. Her research focus is at the intersection of gender, education, labour economics and development economics. We discussed her job market paper, (coauthored with Karan Singhal) titled, “Breaking the Night Barrier: Night-shift Reforms and Women's Work in India.”

We talked about the regulatory frictions in the labor market in India, in particular the gendered restrictions imposed by the Shops and Establishments act, how labor demand changes after deregulation, how it affects overall labor force participation, India’s structural transformation, and much more. 

For a full transcript of this conversation, including helpful links of all the references mentioned, click the link in the show notes or visit mercatus.org/podcasts.

Hi, Nisha. It is so great to have you on the podcast. Thank you so much for joining us.

NISHA VERNEKAR: Thank you. Thank you so much, Shruti. I’m really excited about this.

India's Labor-Market Puzzle

RAJAGOPALAN: So, your paper begins with this really striking labor-market puzzle that’s very unique to India. Women’s education has risen as India has grown richer. Urbanization has increased. Maternal mortality rates have gown down. Infant mortality rates have gone down. The service sector has expanded. 

So, normally, all the things that you expect to relate to an increase in female labor-force participation have all increased. However, in India, female labor-force participation still remains exceptionally low. You ask whether part of this willingness to join the workforce is not just about women’s willingness to work, but also in firms’ willingness to hire.

And you examine a particularly revealing set of provisions and I love the moment crazy regulation pops into all this. You look at the Shops and Establishments Acts. These are state-level acts that a lot of Indian states have, and they have historically restricted women from working at certain hours, especially at night. In businesses such as retail, hospitality, IT services, finance, where employees really need to take different shifts or have to have rotating shifts, often they have to work overnight, and where these are very gendered restrictions when it comes to women. 

So, that’s really what you’re studying: the impact of these gendered restrictions from these kinds of statutory provisions and what kind of impact they have on all these services for a category of women who are otherwise very similar to men. Then you use the Periodic Labour Force Survey and Time Use Survey data, and you exploit this staggered change in regulation across different states. Different states started reforming some of these rules. Because now this is staggered, you’re able to look at how a change in regulation actually impacts women’s ability to work or willingness to work or firms’ willingness to hire.

And your headline finding is something that’s very Econ 101 intuitive result. You find that relaxing these restrictions actually expands women’s access, not only to night shifts, but over all to service-sector employment more broadly, and permitting women to work at night increases their urban service-sector employment by 5.8 percentage points and their probability of holding a contract by 6.3 percentage points. These are nontrivial effects. And the second effect is also about the nature of the employment, which is much more formalized and contractual than working in the informal sector.

And you find that these gains are concentrated more among educated women, you know, wage employees, relatively more skilled occupations and there’s very little gain outside of services. But the most interesting result is also that it has very little effect on employment of men, right? So, this is not really a displacement effect. This is very clearly an effect of removing gendered regulation and then finding the impact on female labor-force employment. Is this a good summary of your job market paper, and how did you get started on this work?

VERNEKAR: Yes, thank you so much for that, Shruti. That is, yes, a very good summary of the paper. I guess, I started from exactly where you started with the summary, which was this puzzle that we know of: In India, female labor-force participation has been woefully low for a very long period of time. We had a period of declining trends, then a slight uptick. But if you look at it compared to so many other countries, even on similar growth trajectories, we’re far behind. We’re at about 30 percent on average across the country, right?

And what we’ve known from the literature is that a lot of this can be attributed to supply-side constraints: things like high care work burdens; in general, social stigma; and norms associated with what kind of work women should do, and what jobs are desirable, whether women should work at all. All of these kinds of things. But the demand-side literature, also studying what kind of demand-side constraints exist in the context of India, is far more limited. 

But what we also know, as you said, is that women do want to work. There is also evidence that women are willing to work when the appropriate, desirable job comes along. What that means is, again, [chuckles] a matter of contention and study in and of itself. What we also know is, even when on paper jobs are undesirable, for example, with the Rural Employment Scheme in India, when we did MNREGA, the government provided these jobs to everyone. These were not targeted to women. And yet, if you look at 2018/19 stats on this, about, I think more than 50 percent of the take-up was actually from women. So, it seems like when jobs are provided, women will take this up, right?

The Roots and Rollback of India's Gendered Labor Restrictions

VERNEKAR: So what we wanted to study here was essentially what are these demand-side constraints that could potentially exist in India, right? When we look at the global literature, a lot of it is related to statistical discrimination where firms say, “OK, we don’t want to hire women because maybe they’re less productive than men,” or it’s preference-based discrimination. But actually, historically, there are also these institutional restrictions. They were rooted in ILO conventions from 1919 that historically wanted to treat women as a protected class of workers alongside children, in fact.

The idea was to say that, “OK, women are maybe physically more vulnerable than men. The working conditions in these industries where they were going to be employed are poor. And therefore, we need to protect these women by preventing them from working in hazardous jobs.” 

What makes this very interesting in the Indian context is that the convention came in in 1919. About 34 out of 54 countries that had ratified the convention actually denounced it by sometime around 1990. Then it was only abrogated in 2017. So, India is one of 20 countries that continue to place night-work restrictions on women when most of the world has moved away from it, including the ILO itself.

RAJAGOPALAN: Yes.

VERNEKAR: This is what makes it a really interesting both time and place to study it. Then I think the other thing that makes it very interesting that we really wanted to highlight was, in India, these night-work restrictions are both in the services sector and the manufacturing sector.

RAJAGOPLAN: Yes.

VERNEKAR: We specifically choose to look at the services sector. The reason for that—there are a couple of different ones, but it’s, again, one of the things you said—a lot of these service-sector jobs are public-facing. They are shift-based jobs.

RAJAGOPALAN: Yes.

VERNEKAR: If you look at why these restrictions were removed in the context of India, there seems to be a demand from employers to hire women in these jobs. That’s extremely important, right?

RAJAGOPALAN: Yes.

VERNEKAR: Then alongside that, there is also the fact that, as the landscape changed, one of the reasons was actually women going to courts and saying, “You know what? These restrictions are discriminatory and prevent us from exercising our right to equal opportunity and access to work.”

Then, initially, we had high courts, of course, holding their ground and saying, “No, these restrictions are important and necessary.” But then over time, this precedent was set that, “Yes, this is in fact discriminatory.” 

So you have these multiple stakeholders, and arguably the most important ones, which are the workers themselves, the women themselves, willing to take up these jobs, and then firms willing to hire them in, again, the service sector, which is also the sector that has led a large part of India’s growth narrative over the last 20-odd years. All of this together made it very exciting for us to study this, specifically in the services sector.

And so what we see happening is from about the 2000s onwards, these Shops and Establishments Acts, which at the level of the state govern the services sector, started to be reformed initially by employers or groups of employers going to states and saying, “Can we have exemptions to this restriction on women working night shifts?” So, for example, 2002, Karnataka had a large number of BPOs and call centers come in, and they got this exemption. Then in Delhi and Rajasthan, it was luxury hospitality chains that got these exemptions, right? So, very specific places.

And then around 2015 and 2017, around the actual abrogation of the ILO convention, we have this more intensive deregulation or removal of this restriction at the state level across the entire services sector. This is essentially what we leverage: the state-wise variation in the timing of these reforms to study, potentially, the causal impact on women’s labor-market outcomes.

The Chicken and the Egg: Disentangling Night-Work Restrictions and Women's Willingness to Work

RAJAGOPALAN: Yes, because this is really hard to disentangle, right? It’s hard to know if women are willing to work and firms are not offering the jobs or if firms are unwilling to set up in a particular place because they don’t expect the right kind of human capital to be forthcoming. The other part of it is, we normally talk about people’s willingness to do something very differently than how economists mean it. So, when we think about women’s willingness to work and fight for their rights, you’re talking about them going to the high court or lobby their politician.

But at the end of the day, when the reform comes through, the only way we’re able to measure willingness to work is if they actually show up at work at a given wage and, similarly, for employers, actually willing to employ them at a given wage. Right? So, to disentangle this, you set it up really well, that you looked at different states that staggered these reforms. You solve the chicken-and-egg problem, so you know which one came first.

Now, given that the reform happened, can you walk us through what was in these laws to begin with? How were they specifically discriminatory against women? And, more specifically, what was it that was changed and not just exempted? Are we talking about minor relaxation in hours? Are we talking about major exemptions for entire IT and hospitality services? And what does the landscape look like across different states?

VERNEKAR: Yeah, with the Shops and Establishments Acts, as you mentioned, these are state-level labor laws that govern the entire services sector. And as you know, in India, labor laws come under the concurrent list. So, we have some laws that are sort of a central-level legislation, and then states have slightly different implementing rules. That’s what is the case with the Factories Act, which governs manufacturing. In the services sector, these are all state-level laws to begin with from their inception.

Now, the specific provision that restricted women’s work, essentially, in our laws, we have a definition of what day shift and night shift is, to begin with. And again, each state defines this slightly differently. Majority of the states use, I think, 6:00 a.m. to about 7:00 p.m. as the day shift

RAJAGOPALAN: —And 7:00 p.m. to 6:00 a.m. is the overnight shift. Yes.

VERNEKAR: Yes, exactly. So, 7:00 p.m. to 6:00 a.m. is the night shift. But then, some states, I think, like Maharashtra, for example, actually starts night shift only from about 10:00 p.m. onwards. So, it’s 10:00 p.m. to 6:00 a.m., something like that. What we’re looking at in terms of the time period of what a night shift is is quite a considerable amount of time, especially in occupations or industries where shift-based work does happen, right?

Now, the exact provision that was historically included, where it was restricted, was that women and children below the age of 18, these two categories, are not allowed to work in night shifts, as the state defines the night shift, right? That is, as you rightly pointed out, not a trivial restriction. It is quite a blanket restriction that has been imposed.

And so, what the changes were when these laws were reformed, or when these specific provisions were reformed, was essentially to remove the word “women,” in most part, from that specific provision. Now, it only says that children under the age of 18 years cannot work night shifts. And then, of course, there are child-regulation laws which are separate, for the larger thing. This is the big change that has happened.

Now, alongside that, what states have done is that they’ve included a sub-provision within this, either within the law itself or within the implementing rules, where they’ve added certain prerequisite, broadly safety considerations, as well as infrastructural requirements, to say that if you want to hire these women in a night shift, the employer needs to provide XYZ, right?

RAJAGOPALAN: Yes.

VERNEKAR: What that includes, one—and this is pretty much blanket across most states—is that women must give consent to working night shifts. Some states are much more particular, where they say women must provide written consent. Employees must prove that written consent has been given on a case-by-case basis. This is an example of a highly restrictive state. Other states just say women must have given consent. 

Then, there is, “Employers must provide transportation facilities.” Again, there’s a range of restrictiveness there and prescriptiveness there. Maybe there’s lighting, installation of infrastructure to ensure there’s adequate lighting, there are adequate toilets for women, things like this. So, here’s a range of different additional restrictions or compliance costs that are included within the provisions themselves. This is the way in which the change has been made.

Formality and the Impact of Reforms and Women's Labor-Force Participation

RAJAGOPALAN: Yes, and it helps you explore two slightly different things, right? One is the nature of the formalization because a lot of these requirements of explicit written consent, requiring contracts that include the night shift in them, inclusion of provision of transportation or other kinds of infrastructure, all those things are associated with more formal labor. Because women also overwhelmingly in India work in the informal sector.

VERNEKAR: Yes.

RAJAGOPALAN: Right? Either in agriculture or very, very micro workshops, doing sewing work or tailoring work and things like that. So, this is one more interesting change that this kind of law allows you to exploit in very urban areas, especially in large-scale services because IT, hospitality, all of those tend to be fairly large-scale.

VERNEKAR: Yes. Absolutely.

RAJAGOPALAN: These are not your two-, three-people micro workshops that you’re talking about anymore.

VERNEKAR: Yes, yes, absolutely. What we do in our paper, again, because of exactly this point that you mentioned—a large part of India’s services sector, approximately 80 percent, is informal and unorganized, so, these are unregistered firms, right?

And so, what we do to try to capture where we think the bite of this reform might be the largest is actually restrict our sample to urban services sectors, where we expect there to be at least a higher concentration of the formal service-sector firms and jobs. And then, we also try to introduce these additional proxies for formality. We try to look at, say, small cities, big cities.

That’s, again, a little bit rougher, but then one of our outcomes is formality itself, having a formal contract. Studying that in and of itself helps us get a little bit closer to where we think this bite might actually be. Then we also look at subgroup analysis using wage-employed versus self-employed women. Here we see very cleanly that we find our effects only amongst women who are wage-employed, whereas the self-employed are really not where you would expect the bite to be, right?

RAJAGOPALAN: Yes.

VERNEKAR: Yes. Then we also look at jobs separately in terms of the level of skill required. A large part of the informal and unorganized sector could fall into mom-and-pop shops as opposed to, like you said, ITES and hospitality, financial services, etc., where we know night shifts are used, where we know these are much larger firms requiring higher skill levels. We try to proxy this formality. Exactly right, I think that this becomes a thing with anything we study in the Indian context, right?

Even if female labor-force participation is, say, 30 percent or, say, 20-something percent in the urban space, the absolute numbers are so huge. So, actually, the effect that we’re looking at, we’re looking at it at a baseline for about 25 million women, which is not at all a small number. I think that’s really important as well.

Measuring Women's Work

RAJAGOPALAN: I had an important question here on how do you actually measure all of this? Indian employment data has never been the greatest. We’ve always had issues with it because of all kinds of historical reasons: seasonality, how we look at agricultural versus nonseasonal work, and things like that. We are even worse at clocking in women’s work [chuckles] because, again, a lot of it happens within the household, on the farm. We’re not able to easily capture how much of that is paid work versus unpaid work and so on.

We are also not great at having very granular data, which is exactly what hours do people work, what timing they work, and so on and so forth. How were you able to put this data together where you measure employment, job formality, hours, wages, the nature of the contract fundamentally that women are engaging in?

VERNEKAR: Right. So, the primary dataset that we use is the Periodic Labour Force Survey of India. This started in 2017/18 financial year. It is actually a rotating panel, but we use it as a repeated cross-section. Essentially, we’re not tracking the same women from the first year until the latest available data, which is for the financial year ’23/24. Now, what this dataset gives us is quarterly employment statistics for a nationally representative sample. Already, this is huge and relatively more granular than what we had available prior to the PLFS becoming available.

What the PLFS does is it captures labor-force participation as well as employment. These are tagged along with the occupation and industry codes consistent with the national industry codes. What we’re able to do using that is actually identify which exact industry women are working in. Therefore, we can separate or isolate what is services sector, or rather, what is governed by the Shops and Establishments Act, and what is nonservices, in that sense. That’s one big advantage of this.

We use that in terms of employment. Then the PLFS specifically asks a question about whether you hold a formal contract or not. Here, again, this will, of course, only be asked to women who are already employed or people who are already employed. What we’re doing is essentially creating a binary variable saying if you have a formal contract in the urban services sector, that is your one, and then everyone else is your zero. Unconditional on being employed in the services sector.

Then similarly, it records weekly wages as well as weekly hours worked. We use a log for both these variables, but we’re able to calculate both weekly hours, weekly wages, and hourly wages, importantly, to test whether what we’re seeing in terms of effects is the extensive margin, the intensive margin, whether it’s a combination of wages increasing versus just women working more hours, things like this. It allows us to do this.

What PLFS Data Can't Tell Us

So this is what we’re able to do with the PLFS data. One major limitation of the PLFS data is that it doesn’t capture actual take-up of night shift versus day shift. This then becomes a problem because what we’re able to say with our primary results and primary specification is, “OK, sure, employment increased or is likely to be higher for the treated cohort’s aggregate in the services sector,” but we can’t say whether this is specifically led by women taking up night shifts.

In order to try to capture this on some level, what we do is that we complement our primary results or analysis with time-use data.India’s time-use data, we have it for two waves, or two rounds, which is 2019 and 2024. What it captures is essentially a 24-hour diary of time allocation across a range of activities, both paid work as well as unpaid work. Again, with the paid work, we have the benefit of not just it being coded alongside the occupation and industry codes that we’ve used earlier with the PLFS as well, but it also then gives us the exact time spent in night shift versus day shift.

What we also do with this is we actually match this to what the state rules are in terms of night shift and day shift. Where a state says 7:00 p.m. to 6:00 a.m., that’s night shift for that state. Another state says 10:00 p.m. to 6:00 a.m.; we use that as night shift. What we’re really capturing is within a state, what is the actual night shift take-up, according to its specific rules.

With the PLFS data, we’re able to come closer to this question of, is it actually night shift that is changing, rather than just the aggregate effects across the services sector. And here, what we find is when we look at a simple difference-in-difference model comparing women in treated states to women in controlled states, pre- and post-reform, we don’t actually find any significant effects. There are no changes in the paid-work variables that we look at or the unpaid-work variables that we look at. Things like leisure time and care work within the household, domestic work, etc.

But, when we use household-fixed effects, what we find is that women, compared to men within the same household, actually increase their take-up of night shifts, as well as the time spent in night-shift work. We have some evidence that leisure time for women increases in the initial difference-in-difference model, but this disappears. It’s less precise when we look at it with household fixed effects. But this tells us something then about, again, what is the margin at which we’re seeing our effects happen.

Maybe with the PLFS data, we’re able to say more precisely that, yes, there is this aggregate effect that we see across the services sector. And at least some part of it is being explained, as per the time-use data, by women actually taking up night shifts as well.

RAJAGOPALAN: Yes.This is super helpful. I think probably putting these two things together and then looking at this at the state level, given the regulatory changes, gives us perhaps the most granular version of what has changed in terms of labor demand for female workers specifically, or at least the potential for labor demand for female workers? 

So, now, I actually want you to walk us through your results. This is both women’s overall urban service-sector employment, formal employment, and then other kinds of effects that you might see overall in the labor-force participation, not just in women.

VERNEKAR: Right, right. Absolutely. So with our primary results, what we see is that our treated women post the reform are almost 6 percentage points more likely to hold a job in the urban services sector, about 6 percentage points also more likely to be holding a formal contract within the urban services sector. And then we have also higher wages of about over 50 percent. As you mentioned earlier, these are not trivial impacts.

I would say, I think when you started with the summary, you said, Econ 101, you remove the restriction, people take up these jobs. I have to say this was a surprising result for me because when we started off, I almost—yes, it is Econ 101, but I almost hypothesized that given the nature of these jobs, maybe we would see no effect because these are not, on paper, very desirable jobs. There are a lot of concerns with them in some sense. 

Indian Women Workers: Exotic—or Homo Economicus?

RAJAGOPALAN: But you know, I feel like that’s partly the narrative. I think we’ve had too much supply-side narrative when it comes to female labor-force participation. I’m not saying any single one of those points is incorrect, but put them together, and we seem to have, as economists, internalized to some extent that Indian women are this exotic cultural artifact who come with all these issues and historical baggage. Maybe that is true, but underneath all of that exoticism that we’ve built in, they are Homo economicus. [chuckles]

Their demand curves and supply curves also slope downwards and upwards, respectively. [laughter] You give them a job that matches their skill level at an appropriate wage, which in this particular instance also brings in formalization plus safety, which is written into the contract, and guess what? They will act exactly like men, and they will act exactly like women in non-Indian markets and go to work.  That’s what I meant by that, that I think what I really like about your paper, it tells us how important demand-side factors are. 

VERNEKAR: Yes.

RAJAGOPALAN: And we rarely study demand-side factors because it’s hard to study firms that don’t exist or firms that are not putting out a call to hire. So, what you’ve managed to do is quite interesting. You’re able to take this and study it with a gendered provision and relaxation of a gendered provision and very, very specifically look at, “Now we’ve relaxed the constraint for potential for labor demand, but does that actually convert into labor demand and therefore then convert into willingness to work?” That’s why maybe I’m not surprised, because I’ve always been a little bit frustrated with this literature.

VERNEKAR: Right. A little bit, maybe even skeptical of that. No, I think genuinely this is what this paper has taught me, because you’re exactly right. I think coming into this space very early on in my PhD, what I was reading was, “OK, urban, formal jobs, women are so woefully underrepresented in this space.” We were calling it “alarming” and “puzzling.” Those were the words that were being used by our top economists. And for me, with my PhD, I started off saying, “OK, let me pick up this puzzle.”

From the beginning, it was like, the norms are too sticky. How much is it going to change? Where is the bite going to come? But this is exactly what this paper has, I think, taught me along the way, that this kind of margin, it’s a very explicit restriction. It should make complete sense. I think presenting it in a non-Indian audience also immediately, that was the question I got. As my results came out, I guess they were also not surprised. So, yes, I see where you’re coming from.

But I think what this also then tells me, going to your point of if the job is desirable and the right fit, as you mentioned. I think essentially what our results are pointing to is exactly that, where these urban services jobs that were previously restricted and have now become available to women maybe are the formal, higher-paying, arguably, jobs in formal firms that already existed, where these employers would have, maybe, if they had higher negotiating power with the state, already gotten exemptions for them previously.

Maybe that is why we’re also seeing those subgroup analysis effects that we see, where the concentration of the effect is amongst more educated women, forward-caste women in higher-skilled jobs. It feels like all of these pieces fit together in terms of that narrative. Right, so even with the formality thing then, maybe it’s just as simple—seeing that there is a higher likelihood of women holding formal contracts, maybe it’s just as simple as that. The other thing that we consider, though, is if firms really were wanting to hire these women, they now want to retain these women in these specific roles—we’re not going to make the case that firms are suddenly benevolent in some way, but maybe they have realized that, “We have to offer some sort of higher wage and this increased formality and this safety if we actually want to attract these workers that we are demanding.” There’s something in that piece also that I think is very interesting.

Relaxing Gender Restrictions Without Significant Displacement

RAJAGOPALAN: What I really like about your result, coming to the last part, is you find virtually no evidence that they are displacing men. This is not one of those examples of, “This was a very gendered profession, which couldn’t attract women or couldn’t exist in its full form because of this awful regulation, and the moment we remove the gendered restriction, this particular gendered occupation attracted lots of women.” It’s not one of those. It’s like, “Men used to do this job, these jobs existed, and the moment you relax the constraint for women, women also do this job, and they don’t really displace men doing similar jobs.”

That to me tells me that this is really a complete labor-demand structural transformation story more than this is about norms. This is not to say norms don’t matter, because of course they do. Providing safe transportation and a safe place, proper restrooms, maybe a female manager, all of those things may be very, very important, but now this is not women instead of men. Especially for me, that’s very nice to see in the hospitality sector.

VERNEKAR: Yes, yes, absolutely. I would take that particular result with a pinch of salt, right? Because definitely, there are some sectors in which we can very explicitly say that, “Yes, we know employers want women in these jobs.”

RAJAGOPALAN: Yes.

VERNEKAR: Certain client-facing positions, like you mentioned, hospitality, things like that. However, when you look at our results for the men, it’s not as robust. It’s not as precise. It’s slightly noisy. But there is this small decline in employment of men post the reform, as well as hours worked by men post the reform in the urban services sector. If we look at the baseline employment of men, even that small decline, maybe that is something. I don’t want to reject the null completely. I want to be a little bit—

RAJAGOPALAN: Yes. We need more evidence.

VERNEKAR: —cautious with that interpretation.

RAJAGOPALAN: Yes.We need more evidence.

VERNEKAR: Yes.We need more evidence, [chuckles] definitely.

RAJAGOPALAN: Yes, but we also don’t see any serious large-scale displacement.

VERNEKAR: No, no.

RAJAGOPALAN: That was my bigger point, right?

VERNEKAR: Agreed.

RAJAGOPALAN: Because this is you talking about the potential size of 25 million women who are affected by all this.

VERNEKAR: Yes.

RAJAGOPALAN: Right? That’s a pretty big chunk in the labor force potentially. And, so, now you’re thinking about, “Oh, is this going to have massive general effects and displacement effects?” And we don’t really see that.

VERNEKAR: No, absolutely. I think this is, again, a very important point that we also want to highlight in the paper while being cautious, which is to say because when we look at this policy—it, like I said, started in 2017. It’s been happening in services and manufacturing. It is not without contention. So, even when we can normatively make the argument that it’s discriminatory, even when there’s just no evidence that this has had a very positive effect for women, and you have labor unions for men specifically making the claim that this will displace male work, and given that in our Indian society, whether we think it’s right or wrong, men are primarily the breadwinners, this is then a bad thing for the household welfare.

This argument is being made. What’s also interesting, I think, is if you look historically at the discourse happening when these reforms happened in the 1960s in now high-income countries, it was essentially the same argument happening from male labor unions, right?

RAJAGOPALAN: Yes.

VERNEKAR: I think this evidence and exactly the point you’re making is very important from a policy context as well in terms of assuaging all the different stakeholders in the matter. Because that is important if we want this to work, if we want to push this ahead.

RAJAGOPALAN: Yes.Here I have a follow-up: What is a way of thinking about your results for women who were already in the workforce versus women who were now brought into the workforce because of this reform in a particular state?

VERNEKAR: Right. So, we do look at this. Basically, our primary results are looking at, “Are wages and hours and holding a formal contract unconditional on whether you were already employed in the services sector?” But we do also look at it conditional on already being employed. What we find there is actually a decline in the hours worked and no effect on the wages, right?

RAJAGOPALAN: Yes.

VERNEKAR: So, what that is then telling us is that both our wage results and our hours-worked results, it doesn’t seem to be being led by women suddenly working many more hours or more intensively, taking up potentially day-plus-night shifts or something like that. It does again seem to fit with the story that higher-skill, better-paying, better-quality jobs, where quality is proxied by formality, are the ones that have now become available to women. I think it’s all consistent with that narrative to some extent.

Implications for Gender Equality

RAJAGOPALAN: Now I want to understand actually what’s happening with these women, especially given all our discussions we’ve had about norms and so on. You said it’s not like they’re working double shifts, day and night shifts, so that’s useful to hear. Are they able to get more leisure time? What are they doing at home? Do the men end up doing more hours at home and more care work versus the women? I ask only because you have time-use data. That’s a very useful thing to understand when we’re talking about the larger picture of how this is changing us socially.

VERNEKAR: Yes.I think even situating it within the broader literature, a paper that looks at very similar gender-equalizing reforms in the context of 1960s USA, so, the Haddad and Kattan paper from ’25, essentially makes this argument that when you make the gender laws more equal, and women move into the workforce, it actually can shift norms in and of themselves. One piece of that in our context—I don’t think we can say it; it’s purely conjecture. But maybe having this law means that women are able to see these jobs as more desirable and the jobs themselves become more socially acceptable if they’re higher-wage.

We can’t test that unfortunately, but there is an argument to be made there. Then part and parcel with that is this: What happens to intra-household bargaining power and allocation of time use between unpaid work and paid work? I wish I could say with a lot more confidence that our time-use result lends itself to this reallocation. What we do see, if we were to just look at the direction of our coefficients, is that there is a marginal increase of men taking up domestic work.

There is a significant result in our difference-in-difference of women having slightly more leisure time. We also see, I think, a negative coefficient on women working day shifts, which, again, goes back to what I was saying before that women are not necessarily doing both. But again, with the time-use data, just a bit of caution because it’s not very precise. It’s not very accurate.

I think what I would hope for is, because, essentially, what we’re seeing with our results, whether with the primary PLFS data or with the time use, our effects are going to be a function of which states we have included within the treatment based on the data we have available, and then also the time since treatment. Maybe in a few more years, the next round of time use, we’ll actually be able to get closer to this answer. It is all very contemporary, especially for trying to do causal analysis in this way.

RAJAGOPALAN: Yes. And now, do you generally see there is just more acceptance of this night-shift work and more states that have not yet amended are willing to amend Shops and Establishments? You’ve staggered it because you are solving for an endogeneity problem, whereas the reform is staggered potentially because now there is just greater social and political acceptability for this kind of work by women. What do you see generally in this regulatory and sociopolitical landscape?

VERNEKAR: No, absolutely. I think, as I mentioned earlier, the landscape has been changing since the 2000s itself, where it’s increasingly become more commonplace for states to give out these exemptions to employers. And Nirmala Sitharaman actually put out a statement saying across the board, women should be allowed to work in night shifts as well. And that’s obviously a huge win. That’s not legally binding. It doesn’t mean anything. We know how this works. But that is a huge stamp of acceptance of the state to say, “Yes, night shifts should happen.” 

At the same time, I don’t know if you’re familiar with the R.G. Kar case of a female doctor that was murdered during a night shift. This again becomes a point of contention, where as soon as that happened, the then–chief minister of West Bengal’s first response was, “Oh, women shouldn’t work night shift.” Right? So, the landscape generally is definitely changing.

Again, coming back to your precise question, if you look at who has now reformed these laws or these provisions from the Shops and Establishments laws, it’s increasingly almost every state. There are very, very few—as of, I think, ’24 and 2025 especially, we see pretty much all states moving in the direction of saying women can work. Again, the main difference that now remains, which, if we had firm-level data, the things we would do would be to say, “What is the level of these restrictions? How restrictive or how prescriptive are we being in how we want to protect women now within these night shifts that we’re allowing them to work? And then, what does that do for the effects from the firms’ hiring-decision perspective?”

RAJAGOPALAN: Yes, and there’s been a lot of great work on this even at the civil-society level.

VERNEKAR: Yes.

RAJAGOPALAN:Prosperiti, which is one of the think tanks, has been talking about amending the Shops and Establishments Act and removing these gendered restrictions for a really long time. Their work made a big impact in trying to get different states to change all these laws, but that’s still restricted to the services sector.

VERNEKAR: Yes.

Opportunities for Improvement in the Manufacturing Sector

RAJAGOPALAN: So, I think the other interesting thing would be to see how this plays out in the manufacturing sector. Because there are still, as far as I understand, lots of restrictions, like women are not allowed to handle certain kinds of chemicals or certain kinds of what they call “hazardous activities” and so on. That would be great material for your next paper, hopefully. [laughter]

VERNEKAR: A couple of things here. Firstly, the Prosperiti reports that talk about these specific restrictions, again, were one of the first things I came across.

RAJAGOPALAN: Yes.

VERNEKAR: Before I even started this paper, I had a conversation with one of the coauthors, and they really were encouraging to take this up. So, yes, fantastic work from them. Secondly, with the manufacturing sector, reforms of this exact nature in the manufacturing sector for night shifts, as I mentioned, have already been happening. And, there is actually a paper that looks at that.

RAJAGOPALAN: OK.

VERNEKAR: There, too, they find very consistent results with what we’re finding, which is that large, export-oriented manufacturing firms are the ones that are being able to hire these women, and there they have the firm-level data. So, they don’t have the individual level— 

RAJAGOPALAN: Oh, nice.

VERNEKAR: —they have the firm. And, so, together, there’s also a very interesting—we’re trying to put the evidence together from both the sides and for both the sectors, right?

RAJAGOPALAN: Yes.

VERNEKAR: And then when it comes to this larger regulation landscape in India, that’s a problem in and of itself that maybe [chuckles] we can talk about in a bit. There are some jobs in which we can still make the argument that statistical discrimination could still exist, right? So, something like lifting of heavy things and this and that. Sure, there will be physical differences between what women can do, what men can do, etc., but— 

RAJAGOPALAN: Yes, or care work. So much care work is usually done by women.

VERNEKAR: Exactly.

RAJAGOPALAN: We are not yet a country with mannies. We have lots of nannies. You see evidence of this gendered work pretty much everywhere.

VERNEKAR: Absolutely, right? In one way or the other, there is this—either it’s feminized work, exactly like you said, or it’s paternalistic discrimination, whether from the employer or the state, to say, “Let’s protect these women.”

But I think essentially the argument that I would make with this—and especially with the results of this paper and the other paper by Gupta and Mahajan on the Factories Act—would be essentially that these restrictions are not for us to make. At the end of the day, all of these jobs, women will either opt in or opt out. Removing that choice from them is then discrimination. It’s just as simple as that. It’s just an equality-of-opportunity argument, which is, I don’t think, a very hard argument to make. [chuckles]

RAJAGOPALAN: No, it’s not a hard argument to make, but it usually doesn’t persuade economists if they’re only focused on the norms side of the story, right? I think it’s just what you get out of that argument. I doubt most economists would say, “No, we must have gendered discrimination when it comes to this kind of regulation.” It’s usually the opposite. It’s like, “Sure, we should treat everyone equally, equal protection envelope, but we are not exactly sure how much bang for the buck you’re going to get,” which is why reforms are typically not pursued. There’s a flip side to this also.

If the entire focus is on norms or on women’s safety or women can’t do A, B, and C, or you need to have a women’s-only workshop or things like that, the trouble is it increases the cost for the firm in which case it’ll affect labor demand and therefore lead to lesser female labor-force participation if the binding constraint is on the demand side.

VERNEKAR: Absolutely.

RAJAGOPALAN: Which is why I think we just need to triangulate and get a better picture. And that’s why I’m so happy to see more recently, so much work on the demand side of the story, as opposed to what we have largely internalized, which is the supply side of the story.

VERNEKAR: Right, right. To your point on the nature of regulation, I think, again, in the Indian context, whether it’s with labor laws or even schooling, for example, we have this inspection culture. We go in knowing that we lack capacity, we say we are going to regulate to no end.

RAJAGOPALAN: Yes.

The Case for Gender-Agnostic Regulations

VERNEKAR: And so, [chuckles] it becomes one of these situations where—so, the normative argument that I was trying to make is, sure, on the side of, “OK, yes, women shouldn’t be discriminated against,” that is the obvious efficiency argument of these many skilled laborers who are not contributing to the paid-work productivity GDP piece. Sure, that’s one part of it.

But then, the other thing is, what is regulation supposed to do? Is it supposed to, essentially, regulate the extensive margin or the intensive margin? Because, again, if we look at it historically, these ILO conventions, etc., they were coming in at a time where women were being pushed into poor working conditions in an industrial sector post the World War. There they were saying, “When working conditions are so poor, then maybe women need extra protection,” etc., etc.

Now, what my argument would essentially be is, you don’t regulate the extensive margin. Don’t put these explicit restrictions. Regulate the working conditions, agnostic of gender, because that’s what we should really be moving towards. Then exactly like you said, the more you micromanage restrictions, what you’re doing is potentially disincentivizing firms from hiring these women who you are trying to target through this. We see that with maternity benefits.

RAJAGOPALAN: Exactly. 

VERNEKAR: We have more than enough evidence of this.

RAJAGOPALAN: Yes 

VERNEKAR: Then on the other hand, in our Inspector-Raj, Indian-labor-regulation way, we’re incentivizing rent-seeking behavior. That, I think, is the margin at which we should really be talking about regulation going forward, right? Yes, again, I wish I could test all of these things one by one. But, yes . . .

RAJAGOPALAN: Yes, but when you can’t test things, common sense and Econ 101 help. [chuckles]

VERNEKAR: Yes.

RAJAGOPALAN: Deregulation overall in labor law, I don’t think it’s a stretch to make a case for that. But thank you so much for doing this. This was such a pleasure. It was a pleasure to read your paper, and thank you for coming.

VERNEKAR: No, thank you. Thank you so much. I’m so happy with this opportunity. I’m glad you liked the paper.

RAJAGOPALAN: Ideas of India is produced by the Mercatus Center at George Mason University. If you enjoy this podcast, please subscribe on Apple Podcasts, Spotify, or your favorite podcast app. Help us grow by giving us a rating and leaving a review. Follow us on Twitter, @srajagopalan and @ideasofindia. Also check out our initiative commemorating 30 years of India’s market reforms at the1991project.com.