Predicting the future of the job market is not a task for the faint of heart. New industries and products can create hundreds of thousands of jobs almost overnight, while competition or technology shifts can destroy jobs just as fast. What's worse, revisions to government data can alter a seemingly solid picture.
Yet peering through the fog, it's clear that something has changed. In the nearly three years since the recession started in March, 2001, the private sector has lost almost 3 million jobs. In contrast, almost three years after the 1990 recession started, all of the jobs lost during the downturn had come back, plus a few more.
The patterns of the past three years -- if they persist -- indicate that the U.S. is leaking jobs at a rate of 1 million per year. A closer look shows the main causes: Faster productivity growth in manufacturing, a three-year gain of only 2% in U.S. exports, and government attempts to control spending on home health care and nursing homes, industries that employ millions of low-wage Americans. In addition, college-educated workers face a threat -- now in its early stages -- of competition from better-educated workers around the world.
This downward drag on the U.S. job market need not be a disaster. History shows that job losses often can be overcome as existing industries expand and new ones are created. An innovative spurt, such as the boom of the 1990s, has the potential to power up demand quickly and turn a weak job market into one where workers are in short supply. Moreover, some of the job losses come from one-time events, such as the terrorist attacks of 2001 and the need to work off the excesses of the 1990s. And 2003 may look better when the Bureau of Labor Statistics issues revised data later this year.
Still, continued weakness in the labor market raises the question of whether the link between job growth and gross-domestic-product growth, which economists have long accepted, has been broken. Even the 8% growth the economy turned in for the third quarter generated hardly any new jobs. Instead, jobs, more than ever, depend on the creation of new products and new industries. The hurdles to job creation get even higher if corporate managers remain risk averse and reluctant to invest in new businesses.
The key factors that could continue to drag down jobs:
Historically, periods of high productivity growth, such as the 1960s and the '90s, also bring strong job markets. High productivity and lower costs help keep prices low, which generates more demand.
But if demand stays weak, the cycle is broken, and higher productivity translates into fewer jobs. That's the fate of manufacturing, where output is down 3% since the recession started. Over the same stretch, factory productivity has gone up by about 15%, vs. a 9% rise in the comparable period in the early 1990s. The faster rise in productivity accounts for roughly half the 1.5 million shortfall in manufacturing jobs. As long as manufacturing has overcapacity, that erosion is likely to continue.
Yet it's hard to find a big dampening effect of higher productivity on job growth in the rest of the economy. Look at wholesale and retail trade and financial services, three areas where information technology is boosting productivity. Employment growth in all three has behaved roughly the same way it did in the early '90s. Wholesale and retail trade show only a modest combined shortfall of 163,000 jobs relative to the previous business cycle, while financial-services jobs are growing faster than they did in the early '90s.
Many argue that U.S. workers are being buried under a mountain of imported goods. But according to the Jan. 14 trade statistics, imports have risen by only 6.6% since the recession started, slower than the gain in GDP over that stretch. Imports from China are soaring, but the growth of goods from Japan, Canada, and Mexico is slower than in the early '90s. Back then, imports rose at more than twice the rate of U.S. output.
It's the weakness in exports that poses the real problem. While exports have risen sharply in recent months, to a three-year high of $1.1 trillion, at annual rates, in November, that recovery has come far later and has been far smaller than the turnaround following the recession and recovery of the early 1990s. Exports today would be $170 billion higher if they had grown by about 18%, as they did then. Instead, they shrank through much of the period and are only now topping pre-recession levels. By themselves, those stronger exports would have added 850,000 jobs, using a conservative gauge that $1 billion in exports generates 5,000 positions in the U.S.
However, even as exports begin to recover as the global economy picks up speed and the dollar weakens, the move to global production means that more and more factories are no longer in the U.S. Much of the shortfall in export-related jobs may thus be permanent.
The full impact of globalization on skilled, educated workers in the U.S. is still unclear. For all the talk of moving high-end jobs overseas, college-educated workers are faring better than anyone else. From September, 2003, to December, 2003, for example, employment of adult Americans with a college degree rose by more than 700,000, while employment of adults lacking that level of education rose by less than 200,000. Even consulting employment in the U.S. has started to increase over the past few months, despite outsourcing.
The free ride is over for educated Americans, however. For years, the U.S. devoted more resources to college education than most of its rivals did. But the number of college grads is rising around the world, with China alone boasting about 2 million new college graduates in 2003. That's potent competition for the 1.3 million graduates coming out of U.S. colleges this year.
Federal Health-Care Spending
Weak job growth is not due solely to megatrends such as globalization. One important example: Legislative changes passed in 1997 cut government payments to nursing homes and home-health-care agencies -- which employ almost 4 million people -- and sharply slowed the growth of jobs in these industries. Partly as a consequence, employment in these industries is rising more slowly than it did in the early 1990s, for a combined shortfall of almost 400,000 jobs.
The same thing is happening at privately run social-assistance programs, such as food banks and individual and family services, which employ more than 2 million people. Such programs, which are dependent on government and charitable funding, boosted jobs by almost 20% in the early 1990s -- but by only 6% in the current business cycle. This translates into a difference of 250,000 jobs.
The prospects of more of these jobs being created are mixed. Medicare announced in July that it was augmenting payments to skilled nursing facilities, in part making up for years of underpayment. Home-health-care spending has also started to rebound. Nevertheless, the huge federal budget deficit makes big spending increases for such programs unlikely.
The ability of the U.S. to compete globally, and to generate jobs to replace those lost, depends on innovation. Innovation, in turn, depends on a willingness to take risks, which may be lacking after the past few years.
One sign of timidity is continued weakness in capital investment, which is barely higher than it was at the recession's end in November, 2001. Another sign is the stagnation of financing for startups. According to Dow Jones & Co.'s (DJ ) VentureWire, U.S. venture-capital outlays in the fourth quarter fell 16%, vs. a year earlier. That's despite the NASDAQ's sharp rise in 2003, which would ordinarily stimulate more investment in new companies.
Venture investment could bounce back, especially if a high-profile initial public offering -- such as Google Inc., due later this year -- does well. But if it stays flat, that's bad news for jobs.
The U.S. labor market has shown remarkable resilience over the past two decades, and there's a good chance it will demonstrate it again. But today, the downward drag on jobs feels like a firmly lodged anchor.
By Michael J. Mandel