The onshoring myth

The decade that preceded the 2008 financial crisis was marked by massive global trade imbalances, as the United States ran large bilateral deficits, especially with China. Since the crisis reached its nadir, these imbalances have been partly reversed, with America’s trade deficit, as a share of GDP, declining from its 2006 peak of 5.5 per cent to 3.4 per cent in 2012, and China’s surplus shrinking from 7.7 per cent to 2.8 per cent over the same period. But is this a temporary adjustment, or is long-term rebalancing at hand?

Many have cited as evidence of more durable rebalancing the “onshoring” of US manufacturing that had previously relocated to emerging markets. Apple, for example, has established new plants in Texas and Arizona, and General Electric plans to move production of its washing machines and refrigerators to Kentucky.

Several indicators suggest that, after decades of secular decline, America’s manufacturing competitiveness is indeed on the rise. While labour costs have increased in developing countries, they have remained relatively stable in the US. In fact, the real effective exchange rate (REER), adjusted by US manufacturing unit labour costs, has depreciated by 30 per cent since 2001, and by 17 per cent since 2005, suggesting a rapid erosion of emerging markets’ low-cost advantage — and giving America’s competitiveness a substantial boost.

Moreover, the shale-gas revolution in the US that took off in 2007-2008 promises to reduce energy costs considerably. And America’s share of world manufacturing exports, which declined by 4.5 percentage points from 2000 to 2008, has stabilised — and even increased by 0.35 percentage points in 2012.

Upon closer inspection, however, the data for 1999-2012 present little evidence of significant onshoring of US manufacturing. For starters, the share of US domestic demand for manufactures that is met by imports has shown no sign of reversal. In fact, the offshoring of manufacturing increased by 9 per cent.

This trend holds even for those sectors dominated by imports from China, where labour costs are on the rise. Indeed, for sectors in which Chinese imports accounted for at least 40 per cent of demand in 2011, the import share has increased at a faster pace than it has for manufacturing overall.

Furthermore, if relative labour costs are an important driver of America’s terms of trade (the relative price of exports in terms of imports), more labour-intensive sectors should have experienced a larger decline. But the data provide little evidence of this.

The only solid evidence of an increase in US competitiveness stems from the sharp rise in output of shale gas. Industries with large energy requirements, like chemical manufacturing, have experienced a much smaller increase in import share than less energy-intensive industries like computers and electronic products. This suggests that energy-intensive sectors are more likely to experience onshoring.

More broadly, the data on US domestic production seem to be inconsistent with the behaviour of the REER and its suggestion of a significant increase in competitiveness. To a large extent, this discrepancy reflects a low and delayed exchange-rate pass-through into US import prices, linked to America’s unique advantage of having more than 90 per cent of its imported goods priced in its own currency, with dollar prices remaining unchanged for 10 months at a time. Even conditional on prices being renegotiated, the pass-through is quite low, with a 10 per cent depreciation of the dollar appearing as a cumulative 3 per cent increase in import prices after two years. The disconnect between America’s terms of trade and the far more volatile REER is also consistent with low and delayed exchange-rate pass-through.

The evidence is clear: Claims that manufacturing is returning to the US simply do not hold water. Of course, given that the increase in emerging economies’ labour costs and the decline in American energy prices are recent developments, import shares could begin to decline in a few years. But, with that outcome far from certain, the US cannot rely on a rapid increase in manufacturing competitiveness to underpin its economic recovery.

Federico Díez is an economist at the Federal Reserve Bank of Boston. Gita Gopinath is a professor of economics at Harvard University. ©Project Syndicate, 2014. www.project-syndicate.org