Jobs are not enough

American workers have had no news this good for years. In June employers added 288,000 jobs, bringing the total for the year to 1.4m, the best six-month stretch since 2006. Unemployment has sunk to 6.1%, the lowest rate in almost six years. It could hit levels long regarded as “full employment” within a year. Help-wanted signs are proliferating, with vacancies up by 20% since January.

Such an ebullient labour market is usually the token of a booming economy. Not now. In the first quarter gross domestic product fell by 2.9% at an annual rate, the worst showing since the recession. This was a result in part of bad weather. Yet the second quarter will only be strong enough to make up the ground lost in the first. Economists had thought 2014 would be the best year since the recession; with growth in the first half of around zero, it is shaping up to be the worst.

Economic growth over the business cycle is driven mostly by swings in demand, and in recent years demand has been held back: households have been repaying their debts; the government has restrained its spending and raised taxes; and interest rates, having reached zero, are unable to fall further. Over the long run, however, a country’s potential growth depends on supply: how many workers it has and how productive they are. The recent divergence between America’s employment and output suggests the country faces not just deficient demand but also enfeebled supply, as more people working without more output means lower productivity. That is bad news for all Americans since their standard of living depends on productivity. It is also a headache for the Federal Reserve, since inflation emerges more quickly when economic capacity is expanding more slowly. Thus it could mean interest rates rising sooner than might otherwise be expected. If so, though, it would also mean they might not rise that high; in a slower-growing economy, there is less demand for capital.

In the 1990s America boasted one of the rich world’s highest potential growth rates, of more than 3%, thanks to a labour force that was expanding by more than 1% a year and productivity, fuelled by the spread of information technology, growing at around 3% a year. By 2007 the Congressional Budget Office (CBO) had trimmed its estimate of potential growth to a still respectable 2.6%. It now thinks it may be just 2.1%. The Fed has lowered its projections of long-term growth by almost as much.

Even that may be optimistic. The recent spell of strong jobs growth and feeble output means that productivity declined by 0.4% over the past year, JPMorgan calculates. The labour force did not grow at all. Economic theory holds that unemployment declines when the economy grows faster than its potential on the upswing of the business cycle. If the slow growth of the past year was above the long-term potential, as the rapid drop in unemployment suggests, it would seem to imply that the long-term potential was actually negative. Things are almost certainly not that bad. Still, JPMorgan reckons America’s potential growth is just 1.75%—about half the rate it enjoyed from 1947 to 2007.

Measuring potential growth is notoriously difficult. Productivity is volatile, making underlying trends hard to discern. Disentangling short-term demand from long-term supply is complicated by the fact that the former has a direct effect on the latter. When the economy is booming, businesses invest and innovate more, which raises productivity, and people who might have stayed at home, retired or remained in school join the labour force. That is what happened in the 1990s: as the economic boom continued with no uptick in inflation, economists concluded that potential growth had risen.

The great reversion

The optimistic way to read the current situation is as the same thing happening in reverse: potential growth may be being depressed by the hangover of weak demand from the Great Recession, rather than by underlying structural forces. For example, the labour force has grown by just 0.3% per year so far this decade, compared with 0.8% in the previous decade, and the participation rate—the share of the working-age population either working or looking for work—has fallen from 65.9% at the end of 2007 to 62.8%. Some of that is structural: of particular note is the fact that the first baby boomers qualified for Social Security (the public pension) in 2008. Some is cyclical: those who have not found work since the recession are quitting the jobs market. But which effect is bigger?

A new report by Barack Obama’s Council of Economic Advisers reckons 1.6 percentage points of the 3.1-point decline in participation can be explained by ageing alone. It reckons another half point is clearly cyclical. That leaves a gap of roughly one percentage point requiring explanation. One factor is that 16- to 24-year-olds are staying in education longer, and are less likely to work while learning. But participation among those aged from 25 and 54, the biggest and most active portion of the workforce, has also fallen—and it was doing so before the recession hit.

This fall has been most striking for those with less education: participation has dropped by four percentage points for those with only a high-school diploma, according to Judd Cramer, a doctoral student at Princeton University. These are the workers most likely to be displaced by technology or foreign competition. But this long-standing trend was made worse by the recession; participation in states hit harder by the recession fell more than it did in those less afflicted, according to Christopher Erceg and Andrew Levin of the International Monetary Fund.

In theory, a hotter economy should draw some of these workers back into the labour market. In practice, the impact is likely to be small. Many dropouts have retired or begun collecting disability benefits, a decision that is “more or less permanent”, according to Shigeru Fujita of the Federal Reserve Bank of Philadelphia. And the structural problem will get worse; the baby-boomers will continue to retire, even as the supply of new workers shrivels. The Census Bureau reckons America’s working-age population will grow by just 0.3% a year from 2010 to 2030, less than a third of the rate of the past two decades. Ageing is not the only reason: falling fertility rates and declining immigration also play a role.

Like labour, productivity is growing more slowly, averaging a little over 1% since the recovery began, about half the average of 2.3% from 1947 to 2007. This might be partly cyclical: weak sales and financial crises have discouraged investment in recent years. But productivity growth had begun to slow even before the recession, from around 2005. John Fernald of the Federal Reserve Bank of San Francisco attributes this to the waning of the IT revolution. Led by the likes of Walmart, a fiercely efficient retailer, businesses began using IT in the late 1990s to better manage supply chains, deploy workers and design products. By 2005 they had reaped most of the benefits, the theory runs, and the pace of innovation in semiconductors had slowed.

The spread of social media which allow new forms of working, of automation which increases an individual’s output and of many other technological innovations which, like those of the previous wave, are taking their time to show up in the productivity figures may yet improve the outlook. But such a pay-off could be many years away. As Michael Feroli of JPMorgan notes, the share of GDP devoted to investments in IT plunged during the recession and has continued to fall, even as investment of other sorts has recovered. The Bank Credit Analyst, an investment journal, notes that lower potential growth means business needs less capital to meet future sales. That would explain why investment, at 12% of GDP, remains below its pre-recession peak.

Even if potential growth picks up a bit, America will increasingly resemble the ageing slow-growth economies on which it used to look down. To improve potential growth policymakers can take various steps, such as raising the age at which the elderly receive government benefits, lowering the top corporate-tax rate and reforming support for the disabled. But such steps would take years to bear fruit. In the meantime the Fed has held interest rates at zero out of a belief that the economy is loaded with spare capacity which is holding down inflation.

Recent data have prompted a reappraisal. Not only has unemployment fallen rapidly, broader measures of underemployment which include the unemployed who have given up looking for work have fallen even further. Yet participation has not risen. Meanwhile, employers are having more trouble filling jobs: in May 3.2% of all jobs went vacant, close to a seven-year high, suggesting the jobless lack the skills that employers are looking for.

All this indicates that the economy is closer to full employment than the Fed had expected just a year ago. Given how quiescent wages and prices remain, rate rises seem still at least a year away. But as Janet Yellen, the Fed chair, noted on July 15th, that date will come sooner if unemployment keeps falling so quickly.