Reform in Practice: Kadambari Shah (Mercatus Center) on Labor Reforms

If you would ask me this question before November 21, 2025, I would have said labor. Today, I'll still say labor, but in a slightly different form. So, we want our firms to grow, we want them to scale, and a large part of our firm's scaling depends on who they can employ and how they can employ people.

Many labor laws draw hard lines, that is, thresholds on the number of people that they can employ. And once a firm crosses these lines, compliance and regulatory burdens kick in. So, while the new codes are a huge win and will make things significantly easier for firms, we've still not necessarily fixed the thresholds problem.

Take the case of Mumbai's mills. In the 1970s and 80s, the textile mills employed over 100,000 workers. And then came an era of slow death, where industrial licensing wouldn't let them live, and labor laws wouldn't let them die.

Compliance burden has also come down. Before the recent reforms, India's compliance burden was massive, what Manish Sabharwal has termed regulatory cholesterol, and a lot of the burden was labor-related. So now, a lot of that has come down in terms of micromanagement, so firms no longer have to maintain several different types of registers.

They don't have to have spittoons at different points in the workplaces. They don't have to follow font sizes. A lot of the micromanagement has come down now.

And what accompanies this is also decriminalization. So earlier, if you didn't clean the factory floor at least once a week, you could go to prison. Now, things like that have been decriminalized.

There's been modernization as well. So, for example, the night shift was not an option for women earlier, and now it is. So the night ban has been lifted, and women can now work in different sectors and in night shifts as well.

But the contractual costs and the compliance costs, they don't exist in a vacuum. They depend on firm size. And we've still not fixed who these regulations apply to.

So the Industrial Relations Code is, of course, a huge win in terms of 100 to 300, but the other regulations have not necessarily changed, or even if they've changed, it's just marginal. So, for example, the Factories Act, which had 10 workers with power and 20 workers without, now the applicability is 20 workers with power and 40 workers without. So it's a marginal increase.

For India's firms to really scale, we need to allow them to grow, and that means not having regulation too early in a firm's cycle. 20 and 40 is higher than 10 and 20, but it's still not enough. Let's take an example.

There is a Banarasi embroidery firm in Uttar Pradesh. It may be able to grow and absorb this regulation at, say, 36 workers. The very same firm in Gujarat may be able to grow and absorb the regulation at 85 workers.

These are just hypotheticals. But if the regulation kicks in at 10 workers, both firms will stop growing at 9 and will not be able to scale beyond that. As a result, we've had the economy dominated by tiny enterprises.

So firms stay small to escape regulation, and that means either staying small intentionally or breaking up into multiple smaller units or hiring informally to just stay under the radar and escape regulation. Amirapu and Gechter show that when a firm increases beyond the 10th worker, the costs of the firm increase by 35 percent, and this is the all-India average. In Bihar it's 69 percent, and in Kerala it's 14 percent. So the move from the 9th to the 10th worker does have very real costs for the firms. We have an economy that is dominated by tiny enterprises. 96 percent of our firms are MSMEs, and 99 percent of all MSMEs are microenterprises, which employ less than 10 workers.

So we have several small firms, but our labor laws are written as if all our firms are large firms. We'd want the bound to be higher and more inclusive so that more firms are under formal legislation. While the new labor codes are, of course, a huge win and will be extremely beneficial, we've still not fixed the thresholds problem entirely.

And in terms of deregulation, I would say my two cents are to take the reforms further, and it will rely on the states. So, for example, back in 2014, this is still under the old labor regime, Rajasthan had actually increased the threshold of the Factories Act from 10 to 20 workers with power and 20 to 40 without power, and for industrial disputes from 100 to 300, which is what the new codes are now. And the economic survey, the 2018-19 economic survey, found that this increase actually led to higher productivity both across the state and across factories.

It also saw a surge in firms growing beyond 100 workers and improvements in efficiency as well. So we've seen that higher thresholds can be beneficial. We would need to increase them.

And with the Rajasthan example, even though we increased from 10 to 20, 20 to 40, and 100 to 300, and saw great improvements, imagine if the threshold was 500 or 1,000. This is not a one-off. Evidence has shown that states with more flexible labor laws actually have higher improvements in productivity, in job creation, in output.

And states with more rigid laws rarely see firms cross the threshold because the moment regulation kicks in, firms stop to expand. While we've done a great job with new labor codes, we need to increase the bounds even further. We need states to increase the bounds even further and allow our firms to scale.

So to conclude, the thresholds problem remains largely unsolved with the new codes. We want our firms to scale, and my one-line solution is to increase thresholds.