Last week I wrote about the forces shaping our economy, and said in passing that “factory jobs have disappeared everywhere”. Adrian Wood, economics professor emeritus at Oxford university, takes me to task in an email. I had in mind the developed world, for which the claim is indeed true, but Wood rightly points out that manufacturing employment has risen so much in emerging economies in the past three decades that the worldwide total is higher, too.
Apologies for being imprecise — and thanks to Wood for sending his data, which I have charted below. As they show, the nearly 100m factory jobs added between 1985 and 2014 in what used to be known as the Third World more than outnumber the just under 50m lost in the old west and the communist bloc.
But populations and economies have grown, so the relative weight of manufacturing employment looks somewhat different. The next chart illustrates the changing employment shares of manufacturing. These have fallen dramatically in the old industrialised economies, have increased somewhat in developing countries (but outside of China the share has been almost constant), and gone down by nearly 2 percentage points on a worldwide basis.
Wood uses these data in a recent article revisiting the debate in the 1990s among economists about the causes of increasing inequality between low and highly educated workers. As he points out, that debate was won by those who thought globalisation was at most a minor part of the explanation.
Wood himself argues that globalisation had a large effect on employment of different types of workers and suggests: “Although there were surely other causes of the long deterioration in the economic position of ordinary people in the North, we cannot be confident that globalisation made a minor contribution . . . a more tenable position would be that we don’t know how big the contribution has been.”
He intriguingly speculates that policymakers may have been more alert to those left behind by developments if the economics profession had not exonerated trade so overwhelmingly.
I lean towards the consensus view, but here I want to ask the slightly different question than the effect of wages. Following Wood’s instructive focus on quantities rather than prices of labour (numbers of jobs rather than wages), should we think of these jobs having “moved” from industrialised to developing countries? Put differently, if trade liberalisation had been limited, would we expect only the composition of factory jobs (or employment shares) across regions to be different, but the number for the world as a whole to have behaved as it actually did (an absolute rise but a moderate fall in the share)?
There are two reasons to think not. Recall that even as the old industrial countries’ manufacturing employment has fallen, their output of goods has not. That is because rich countries have had high and rising (despite recent stagnation) productivity in manufacturing. Suppose that total worldwide production of manufactures had followed its actual development over the past few decades: our best guess must be that if more of that had taken place in the old industrial world, it would have employed many fewer workers. Indeed it may be that the availability of cheap labour in emerging economies delayed further job-saving automation, and so its unavailability would have accelerated technology-driven job-shedding in rich countries. Total world factory employment would have risen less — it may not have risen at all — and its share of all jobs would have fallen by more.
That assumes that total production would have grown as much as it actually did. But — and this is the second reason — that is implausible. Some of the new factory jobs created in the past three decades have produced industrial goods for developing country markets. In this counterfactual history these economies would have been poorer and presented less demand for manufactures. Conceivably, if industrialised countries had “kept the factory jobs” they would largely have produced goods for themselves. So overall output would probably have been smaller, reinforcing the pressure on the number of jobs and the employment share in manufacturing.
We may not know well enough how big the contribution of globalisation to the predicament on the left-behind has been. It may in any case be hard to isolate because of the interaction between trade liberalisation and other forces — such as the speed of automation and the demand for manufactures. But those very interactions give us little reason to think many of the factory jobs created in the emerging world could have been created in rich countries instead.
- Three economists use leaked data to derive state of the art estimates of tax-evading behaviour by the very rich. Their findings are shocking: half of all offshore assets belong to the wealthiest 0.01 per cent — and for this group illegal tax evasion (not legal avoidance) amounts to 25-30 per cent of their wealth.
- The yield on US 10-year bonds has reached a seven-year high.
- In recent quarters, a decelerating eurozone has relied increasingly on export-driven demand growth. Brad Setser finds that Ireland accounts for one-half of that net export contribution to growth. Not so much an export powerhouse as an accounting powerhouse; as Setser suggests this anomaly has to do with the tax advantages for intangible services industries in booking profits in Ireland.