Is free trade efficient and/or good?

Greg Mankiw's recent column in the NYT praising international trade, where he urged Americans to see emerging economies as trade partners and not competitors, has generated an interesting debate about the costs and benefits of international trade.

Uwe Reinhardt highlighted the difference between the way non-nationalistic economists and nationalistic citizens view the issue of gains and losses from trade,

"Many Americans might balk at the lower-priced scarf if it were offered not by an American but by a low-cost manufacturer in Shanghai or Bangladesh... Whether a fellow American gains from a trade or someone in Shanghai does not make any difference to most economists, nor does it matter to them where the losers from global competition live, in America or elsewhere."

He also points to the controversial work of Alan Blinder (see pdf here and here), where he questioned the wisdom of the theory that "every country gains by unfettered international trade" in the context of the threat to American jobs posed by off-shoring of services to emerging economies. Prof Blinder had estimated that 30-40 million jobs in the US are potentially off-shorable and this "may pose major problems for tens of millions of Americans over the coming decades". He also raised the concerns about how America would be able to cope up with its impact, especially given its tattered social safety net.

In this context, David Autor, David Dorn and Gordon H. Hanson (via Mike Konczal) explored the impact of import competition on US local labour markets that were differentially exposed to the rise of trade with China in the 1990-2007 period due to differences in their initial patterns of industry specialization. Controlling for various external factors, they found,

"Increased exposure of local labor markets to Chinese imports leads to higher unemployment, lower labor force participation, and reduced wages. While the employment reduction is concentrated in manufacturing, wage declines occur in the broader local labor market, and are most pronounced outside of manufacturing. Growing import exposure spurs a substantial increase in transfer payments to individuals and households in the form of unemployment insurance benefits, disability benefits, income support payments, and in-kind medical benefits. These transfer payments are two orders of magnitude larger than the corresponding rise in Trade Adjustment Assistance benefits. Nevertheless, transfers falls far short of off-setting the large decline in average household incomes found in local labor markets that are most heavily exposed to China trade."

They also write about the distributional consequences of trade - while it lowers incomes for workers in industries or regions exposed to import competition, gainers include consumers who have increased product variety and industries and regions with expanding exports have higher income growth. They find that the gains from trade with China are between $32 and $61 per person, whereas the deadweight losses are estimated at $52 from the transfer mechanisms in place.

This analysis does not fully capture the benefits. As the article itself mentions, while the benefits of trade are permanent, the deadweight losses of transfers are temporary. More importantly, it does not include the benefits of trade by way of exports (after all, we cannot assess the impact of trade by merely valuing only one side of the equation!).

The critical issue here is not whether free trade is beneficial or not. On the net, it is hard to argue that free-trade is not beneficial and harder still to prove it. But, as the aforementioned study acknowledges and we all know, the benefits and costs of free-trade are not evenly distributed. It is natural that the gainers are happy and the losers are disgruntled and oppose free-trade. The important issue therefore is how the gains and costs are distributed among different people and regions and whether governments can have any role in that.

This raises questions about the relative gains and losses, both in terms of amounts lost and gained and the numbers of people on both sides of the equation. It is here that there is a growing belief, among a larger number of people and regions, that they are being forced to take a disproportionate share of the costs of trade. They accuse a few of disproportionately benefiting from trade. Economics has little to teach us about how to resolve such distributional issues. That is the realm of politics.

The politics of fairness and the sustainability of Pareto optimizing trade demands that the losers be appropriately compensated for their losses. The compensation is usually in the form of some type of transfers from the government to these losers. However, this raises the issue of where the government can find the resources to finance such transfers.

Since the gainers benefited disproportionately, it is only natural that some portion of their excess gains be appropriated by way of say, taxation. In other words, re-distribute a share of the (disproportionate) benefits obtained by the gainers from trade to the losers. This will buy the support of the losers and ensure the sustainability of the Pareto improving trade.

Now free-market economists would argue that this raises questions of how much to re-distribute and to whom. How do we redistribute? Who are the gainers and how much did they gain? Who are the losers, and how much should they be compensated? Yes, the answers to these and more are not simple. But they are not insurmountable.

But the alternative is to accept the losers argument that they do not deserve to lose because of trade and therefore all trade should be stopped (in any case, once you oppose some types of trade, a slippery slope appears, where trade in general gets questioned). After all, and economists understand it better than anyone else, any transaction is economically efficient (and Pareto optimal) only if it makes people better off without making anyone worse off. Free-trade, minus re-distribution, clearly leaves the losers worse off and is both inefficient and not a Pareto improvement.

Such re-distribution can be made less distortionary and with minimal deadweight losses if it is structured so as align the incentives of all the parties. The losers, getting the transfers, should have incentives in place to search for newer jobs and acquire new skills that can help them find a job quickly or recover their income losses. Similarly, the gainers should be taxed for the disproportionate gains, arising from easy arbitrage opportunities, so that their incentives are not distorted badly.

The important take-away from all this is simple. Free trade is good, but only if it is supplemented with transfers that cushion the losers. Put differently, sustenance of free-trade makes the strongest case for social safety nets. In fact, losers from free trade are just another example of market failures (in so far as an "efficient outcome generates "bad" results). A social safety net can address all these problems, not just those arising from trade failures. And Econ 101 teaches us that social safety nets are in place precisely to resolve such failures. William Polley is spot on

"You see, it is the potential for a Pareto improvement that makes free trade desirable. There are winners and losers. But the winners gain more than the losers lose. So effect a transfer from the winners to the losers that still allows the winners to gain but compensates the losers for what they lost. Only then can you really say that free trade (with the compensating side payment) benefits everyone. If the compensation is not there, then I cannot unconditionally advocate free trade. I must call attention to the fact that some will lose."

See also Mark Thoma (also here), Tim Worstall, and Free Exchange debate the issue.