Economic Outlook n° 1187 | Special Report | The Reindustrialization of the United States
Euler Hermes
2000-10: momentum lost followed by massive restructuring
▶ Limited role in the U.S. economy? During the previous decade, the manufacturing weight in the U.S. economy remained fairly steady, at 12.4% (real value added) of GDP in 2000 and 2010, with limited fluctuations over the period (high of 12.9% in 2007 / low of 11.5% in 2009). The growth years witnessed noticeable contributions from the manufacturing sector to the improvement of the economy - a bit higher than would have been expec- ted given its weight in the U.S. economy - but they were always less than the contributions from the ser- vices sector. Conversely, the significant drops in manufacturing value-add, which occurred during the recessions, were a significant drag on the U.S. eco- nomy and, generally speaking, were major down- ward economic drivers.
▶ Sharp restructuring in manufacturing employment Between 2000 and 2010, the manufacturing sector lost 5.7 million jobs (-33%). In Figure 6, we compare the actual employment in manufacturing from 2000-2010 and two scenarios estimated: > the Productivity 2000 scenario shows the number of jobs needed to produce the actual output with pro- ductivity held constant; > the Output 2000 scenario shows the number of jobs it would have taken to create that output with the actual productivity gains. This scenario more clo- sely matches what actually happened, suggesting that any drop in output during a recession had a smaller effect on employment.
The analysis shows that manufacturers did not lay off workers because the economy was so bad, but because productivity was so good that they needed fewer workers. The main driver for manufacturing job losses during the last decade was certainly pro- ductivity gains. The output decrease resulting from the recession, shipping jobs overseas, and the lower
competitiveness of U.S. manufacturing in the global economy contributed much less to the decline in jobs. The impact of productivity improvement on job losses was more than three times as high as the impact of economic contraction.
▶ Different goods, different stories? We carried out the same analysis for durable goods manufacturing, which posted a -35% decrease in employment during the last decade – or 3.8 million jobs. The impact on employment that productivity improvements had, compared to the impact that the recession had, was only about two and a half times greater, as opposed to three times greater for all manufacturing. The difference suggests that durable goods manu- facturing was more sensitive to economic conditions than overall manufacturing. Durable goods meant to last three years or more - such as appliances, cars and planes - represent discretionary purchases which can be more easily delayed than those for non-durables such as food. The computer and electronic manufacturing sector witnessed a sharper drop in employment (-40% or loss of 735,000 jobs). The dramatic decline in jobs was a result of the adverse effects of strong produc- tivity improvements (+79% over the period), which were only partially offset by jobs created due to increasing activity. Between 2000 and 2010, the machinery sector demonstrated better resistance to erosion as its employment dropped by a relatively smaller -32% (- 467,000 jobs). Lower productivity enhancement (+27% during the decade) led to a lower adverse impact on job losses. In 2010, non-durable goods manufacturing jobs dropped -30% from 2000 (-1.9 million jobs). The impact of productivity improvements was four times as large as the impact of activity fluctuations during the last decade.