Trade Is Not Bad For US Jobs

This month Gregory Mankiw, chairman of the US administration's Council of Economic Advisers, remarked that higher productivity was "probably a plus for the economy in the long run". The reaction was immediate and hostile. Professor Mankiw's theory "fails a basic test of real economics", said Dennis Hastert, speaker of the House of Representatives. "We can't have a healthy economy unless we have more jobs here in America." Senate Democrats even signed a resolution urging President George W. Bush to denounce his top economist.

This, I hasten to state, did not happen. Prof Mankiw did not need to defend higher productivity. Nor would politicians have attacked him if he had. It was his views on trade that awoke this ire. Yet it would be just as logical to rail against productivity growth as against cheap imports. Indeed, it would be rather more so, because productivity growth has been a far more significant source of job losses. The only relevant difference between productivity and trade is the all too visible involvement of foreigners, who do not have votes. They make wonderful scapegoats for unscrupulous politicians.

The explanation for the wave of protectionist sentiment is not at all hard to find. The current recovery is remarkable for the weakness in employment (see chart). For this there are two explanations: exceptional productivity performance and the weakness of the recovery (partly explained by the shallowness of the previous recession). Output per hour in the non-farm business sector has risen at a rate of 4 per cent over the past three years. This is far higher than the growth of the economy, at a little over 2 per cent. The fall in employment has been the inevitable result.

The decline in manufacturing employment, at 2.63m between March 2001 and January 2004, was greater than that in the whole economy, at just 2.35m. By January 2004, employment in manufacturing was 17 per cent below its level in June 2000, the peak month for manufactured output in the last cycle. The proximate cause of the job losses was a 17 per cent increase in output per worker, while output fell by a mere 3 per cent (see chart).

Yet even politicians know that the ability of the US to produce twice as much manufactured output today as it did two decades ago, with even fewer workers, is a good thing. So why do they not celebrate increased trade, since it also permits a country to obtain goods and services more cheaply than it otherwise could, just as rising productivity does?

Trade does, as critics stress, mean painful adjustment for those affected, as well as shifts in the rewards for different workers. Yet this is just as true of productivity. Information technology destroyed the jobs of armies of clerks and raised the wages of educated workers relative to those of less educated ones. But it had no deleterious impact on employment. Between 7 and 8 per cent of US private jobs are lost every quarter. But employment has still increased. Neither rising productivity nor growing imports will undermine overall employment, provided the labour market is flexible.

By increasing competition and lowering the prices of inputs, trade can also generate higher productivity directly. This is particularly true for IT, a point made with great force by Catherine Mann, of the Washington-based Institute for International Economics, in her analysis of "offshoring" of IT services, which is, alas, the current focus of protectionist paranoia.*

In the late 1990s, argues Ms Mann, globalisation of IT hardware production made hardware about 10 to 30 per cent cheaper than it would otherwise have been. Real GDP growth might have been some 0.3 percentage points less a year if IT production had not been globalised. Now it is IT software where opportunities are largest. The potential for faster productivity growth in the economy, as cheap foreign skills are applied to this sector, are enormous. Some jobs will disappear in the software- producing sectors of the economy. But more activity will be generated elsewhere.

The same point is made, in a different way, by the McKinsey Global Institute's study of offshoring.** It argues, persuasively, that the US will capture economic gains through several channels: reduced costs (about $0.58 for each dollar spent overseas); new revenues, as overseas providers buy US goods and services (about $0.05 per dollar spent); repatriated profits (about $0.04 per dollar spent); and redeployment of domestic US labour (about $0.45-$0.47 per dollar spent). Overall, then, the US gains $1.12-$1.14, while the foreign country gains another $0.33 (see chart). Nobody should take the specific estimates seriously. But the broad point is powerful. As Prof Mankiw did, in fact, say, there is nothing exceptional about services. As he argued, "more things are tradable than were tradable in the past. And that's a good thing."

Suppose the politicians did succeed in halting offshoring. Would that save the jobs of programmers or call centre operators? In all probability, no. Both are vulnerable to technology in any case. All it would do is raise costs to users and slow economic advance.

What is depressing about the debate is not just the blaming of foreigners but also its irrelevance to the challenges confronting the US. The most immediate of these is to create sustained growth in demand. Equally, the US confronts significant structural challenges. If its people are to gain from the emerging division of labour, they need high-quality education, as Alan Greenspan, chairman of the Federal Reserve, argued last Friday. In addition, a case can be made for subsidising the wages of the unskilled.

What must be avoided are policies that undermine increases in living standards, threaten the US commitment to liberal trade and, not least, attack the nascent exports of a poor and gigantic democracy that is, at last, trying to participate in the global economy. US legislators need to take a grip of themselves. Attacking cheap imports of services is no more logical than bewailing rising productivity. The US, they should remember, benefits hugely from both.