Samuelson: Protecting the Jobless

It's an election year, and partisan acrimony has escalated. Democrats and Republicans portray themselves as the nation's saviors and protectors against all the programmatic atrocities of the other side. Can we find a refuge of common-sense agreement amid this self-serving political din? Well, here's a modest proposal for the economy: enact a temporary extension of unemployment insurance from the standard 26 weeks to 39 weeks. There's nothing original about this; benefits have been extended in every recession except one since the 1950s. The proposal's virtue is precisely its modesty. It would directly aid obvious victims of the downturn: people who lost their jobs through no fault of their own. Although most unemployed usually find new jobs within the normal six months, the task becomes harder in a slump. Perhaps 3 million people will exhaust their benefits this year, estimates the Congressional Budget Office. The cost of added protection is also modest: about $13 billion for a proposal that recently passed the House Ways and Means Committee.

True, it's not yet clear that we're even in a recession. In the first quarter of 2008, the preliminary estimates showed that the economy's output of goods and services (gross domestic product) increased at a 0.6 percent annual rate. That's not much, but it's still growth and suggests the economy doesn't meet one basic test for a recession—two consecutive quarters of negative GDP. But that verdict comes with a big caveat: the job market is already in retreat.

Look at the numbers. Though the April unemployment rate of 5 percent is not historically high (the average for the 1990s was 5.8 percent), it's way up from the recent low of 4.4 percent in October 2006. The difference reflects an additional 900,000 unemployed out of a total of 7.6 million. For much of 2007, the number of new jobs was actually increasing, though not fast enough to absorb all the new entrants into the labor force. Even this growth has halted, and the number of jobs has begun to decline. Since December, payroll employment has fallen by 260,000.

How does employment decline in an economy that's expanding, even if feebly? Easy. Weak companies fail or shrink; the survivors don't hire more people to handle the extra business. In technical terms, productivity (old-fashioned efficiency) improves. Fewer workers do more work. By many indicators, the job situation may get worse before it gets better. For example, the Conference Board's index of online job vacancies is now dropping and is 16.4 percent below what it was a year ago; declines occurred in 44 of 50 states.

The great danger of unemployment insurance is that it worsens the very problem it's supposed to relieve. People are paid to be jobless. Benefits that are too generous or that last too long can raise unemployment. This is a problem in Europe, where benefits are relatively lavish. But it's a smaller issue here; some academic studies find that extending unemployment benefits by 13 weeks might slightly slow the flow of workers back into jobs. People don't start looking so quickly or are more picky. But the effects aren't large, because the benefits are fairly stingy.

For starters, only workers who are laid off can get them. People who quit or who are entering or re-entering the job market aren't eligible. Altogether, only about 36 percent of today's unemployed receive benefits, says Maurice Emsellem of the National Employment Law Project, a research and advocacy group. Nor are benefits particularly high. They now average about $300 a week; that's $7.50 an hour for a 40-hour week. (States set the benefit levels. Average weekly benefits range from $179 in Mississippi to $408 in Hawaii. The payroll taxes that pay for the program also vary by state.)

Congressional Democrats—and some Republicans—have supported an extension of benefits. The Bush administration has resisted, arguing that Congress has never before lengthened the benefits with such a low overall unemployment. True. When benefits were extended in early 2002, the unemployment rate was 5.7 percent. In 1991 the extension occurred at 7 percent, and in 1982 it was 10.1 percent. But so what?

What's wrong with this argument is that it ignores basic changes in U.S. labor markets. Over the past two decades, American businesses have gradually toughened their hiring and firing policies. In recessions, they resort more to permanent dismissals as opposed to temporary layoffs; in recoveries, they're more cautious in adding new workers. After both the 1990–91 and 2001 recessions, job creation didn't resume for many months; indeed, after the 2001 slump, payroll employment didn't reach its prerecession peak for more than three years.

It's harder to find a new job. Average spells of unemployment have slowly lengthened. The increase since 1960 has been about six weeks, estimates economist Gary Burtless of the Brookings Institution. "It's more likely you'll exhaust your benefits today than in the 1950s and '60s," he says. That's especially true in a slump, when the share of those unemployed for more than six months rises to a fifth or more.

Congress ought to send the president a stand-alone extension of unemployment benefits. It would be hard to veto. Compared with the $152 billion price tag on the economic stimulus program earlier this year, the cost is slight, and the added protections would go to innocent victims of the downturn.

But this may be a fantasy. It seems equally plausible that the extension would be added to an expensive extravaganza of other spending increases (construction projects, grants to states, energy subsidies) and tax breaks labeled "Stimulus II." The whole package would be a partisan sledgehammer designed to show that Democrats care about the economy and Republicans don't. It could then become easily mired in partisan politics, going nowhere and demonstrating again the long odds against common sense.