While Donald Trump plans to fix America’s economic malaise by blocking imports from Mexico and China, the latest IMF World Economic Outlook laments the slow growth of global trade. They can’t both be right.

It is almost exactly 200 years since David Ricardo set out the advantages of countries trading with each other. This is one of the few things economists agree on, and the two centuries since then have provided ample evidence that international trade boosts productivity and provides greater choice for consumers.

The fact that trade has grown twice as fast as GDP since WWII is central to the rise in living standards in this period. The benefit is seen not only in the industrial countries, but in those previously-poor countries which have participated in the Great Convergence, which lifted a billion people out of poverty.

World Real Trade and GDP Growth in Historical Perspective

Source: IMF

Here is the two-fold concern: international trade is now growing more slowly than GDP, with adverse implications for living standards. And if Trump becomes president, it will grow even more slowly.

The IMF sees the feeble recovery from the 2008 crisis as the main reason for slower trade expansion. The two interact: when economic growth is slow, investment is low and international trade stagnates.

Trump and Brexit have both demonstrated once again that, however confidently the case is made for free trade, the public remains unconvinced. Even the strongest advocates of international trade accept that it is disruptive – after all, the productivity enhancements come largely from shifting resources (including people) into different – and more useful – tasks. Ricardo didn’t promise that there would be no losers; only that the overall benefit would be large enough to compensate them and still leave everyone better off.

There are, it is true, transitional costs involved in this resource shift that economists tend to downplay. The public, for its part, tends to assume that displaced workers lose their jobs permanently. The reality is that if the economy is growing at a good pace, workers are adaptable. On the surface, US jobless data since 2008 appears to support this: while the recovery has been feeble, unemployment is now back to pre-crisis levels.  However, other related statistics reveal a harmful legacy of the slow recovery: displaced workers have become discouraged and stopped looking for a job. In America many ‘prime-age’ males (25-54 year-olds) have dropped out of the labour force.

If Trump’s main appeal is to blue-collar workers, let’s look more closely at the experience in their core industry: manufacturing. Adaptation and change have been the norm in manufacturing. Contrary to general perceptions, manufacturing output continued to grow strongly following what another one-time presidential candidate,  Ross Perot, described as the ‘great sucking sound’ of NAFTA in 1994.  Even when China’s entry into the World Trade Organisation in 2001 opened up the US market, US manufacturing output continued to expand.

The shock of intense competition from China, however, bought a change in manufacturing employment.  After plateauing between 1980 and 2000, employment fell sharply.  China was not the only reason: the ‘tech-wreck’ recession of 2001 and the Great Recession of 2008 also brought sharp cut-backs in manufacturing employment. But the competition from China provoked a burst of innovation which transformed manufacturing, shifting labour-intensive processes offshore, splitting production through the use of offshore supply chains, and computerising the processes that remained onshore. Output has now reached record levels, but it is produced by far fewer people.

Studies have identified communities where Chinese import competition can be specifically identified as the cause of persistent unemployment and social disruption. The ability of the labour market to find alternative employment was severely hampered by the weak economy. Government transition schemes are inadequate and largely irrelevant. The traditional readiness of the American worker to relocate seems to have weakened, and the higher technical requirements of the current manufacturing jobs leave many former workers under-skilled.

Computer technology may have played a new role in this recent experience, eliminating semi-skilled clerical jobs. This may also help explain the anxieties of Trump’s middle-class supporters.

Seen in a longer perspective, the decline in manufacturing employment is just a continuation of a transformation that goes back many decades. In the early years after WWII, manufacturing employed nearly half of the American work force. As a percentage of the workforce, manufacturing has fallen fairly steadily, with the post-2000 experience hardly deviating from the trend. Manufacturing is now a modest employer, with employment growth in the now-dominant services sector.

 Private Employment Shares in the Service, Manufacturing and Construction SectorsSource: Federal Reserve Bank of St Louis Review Jan/Feb 2013

No doubt governments could have done more, to soften and assist the adaptation to the twin drivers of globalisation and technology.  But trying to slow globalisation or the advance of technology makes no sense. Those wondering what or who is to blame for the anxiety and dislocation that has prompted many to support Trump should instead focus on the fiscal policies that failed to deliver an adequate recovery after the 2008 crisis.