The cat is formally out of the bag. Trade reforms are bad for jobs, but usually good for productivity.

In a recent paper*, Rana Hasan, Devashish Mitra, and K V Ramaswamy have studied a lot of industry-level data for 1980 to 1997, disaggregated by states and come to this conclusion.

They find "a positive impact of trade liberalisation on labour-demand elasticities in the Indian manufacturing sector". This means that things turn nasty for the workers when trade reform takes place.

"These elasticities turn out to be negatively related to protection levels that vary across industries and over time." This means that higher the protection, the lower the elasticity. Naturally.

But the most interesting finding is that "these elasticities are higher for Indian states with more flexible labour regulations." That explains industrial progress in West Bengal, on the one hand, and Gujarat on the other. Investment will flow to states with more flexible labour markets.

It is interesting to find in this context that "the length of time it takes to move halfway to a new equilibrium in response to a (trade reform induced) shock works out to be roughly one-third of a year (0.30 years)." The authors say, rightly, that given how everyone goes on about how inflexible Indian labour markets are, this is surprising.

They also say that this is suggestive of the fact that "adjustment of employment is in fact less costly than the critics claim". Clearly, ways have been found around the labour laws.

Finally, it turns out that "after the reforms, volatility in productivity and output gets translated into larger wage and employment volatility". That is, fewer people get better paid jobs. Also, prices drop as monopolies get dismantled and technology levels improve.

There is a political hot potato here as well as an economic one. The political one is obvious: why would any political party support something that might cost it votes?

And this is true not just of India but all countries, except China, which simply expands its export markets to absorb its workforce.

India need not follow the Chinese model, which depends to an uncomfortably large degree on coercion. It can, however, take advantage of its own domestic market by dismantling the system of reservations for small-scale industries.

As so many studies have shown, more than anything else, it is this idiotic policy that has kept down employment growth because it prevents large-scale investment in precisely those areas where domestic demand is huge.

In view of this, the authors of this paper appear to have come to a slightly untempered conclusion. That increased competition engendered through trade reform will adversely affect employment in a flexible labour market is clear enough.

The policy issue, however, is to absorb the displaced labour. Their next effort should be on what has happened to employment in industries where there has been tariff reform and which have been de-reserved from the SSI list.

The larger economic issue is equally perplexing: if your labour market doesn't adjust fully, you handicap your exporting firms, so that even the labour that has been shed as a result of half-hearted trade reform doesn't get fully absorbed. This is what probably explains the declining trend in employment growth since reforms started.

This, to my mind, is the most important issue for policy: the pace and timing of trade reform. Some economist should now ask if there is an "optimal" level for these two as well. If nothing else, the exercise to find an answer will keep some of them in jobs for a while.