The Unemployment Puzzle: Where Have All the Workers Gone?

The U.S. unemployment rate is down, but rising numbers of Americans have dropped out of the labor force entirely

By Glenn Hubbard

April 4, 2014 7:10 p.m. ET - Wall Street Journal

The problem is not just a cyclical downturn. We need to tackle deep structural issues in the U.S. economy. Bloomberg News

A big puzzle looms over the U.S. economy: Friday's jobs report tells us that the unemployment rate has fallen to 6.7% from a peak of 10% at the height of the Great Recession. But at the same time, only 63.2% of Americans 16 or older are participating in the labor force, which, while up a bit in March, is down substantially since 2000. As recently as the late 1990s, the U.S. was a nation in which employment, job creation and labor force participation went hand in hand. That is no longer the case.

What's going on? Think of the labor market as a spring bash you've been throwing with great success for many years. You've sent out the invitations again, but this time the response is much less enthusiastic than at the same point in previous years.

One possibility is that you just need to beat the bushes more, using reminders of past fun as "stimulus" to get people's attention. Another possibility is that interest has shifted away from your big party to other activities.

Economists are sorting out which of these scenarios best explains the slack numbers on labor-force participation—and offers the best hope of reversing them. Is the problem cyclical, so that, if we push for faster growth, workers will come back, as they have in the past with upturns in the business cycle? Or do deeper structural problems in the economy have to be fixed before we can expect any real progress? To the extent that problems are related to retirement or work disincentives that are either hard to change or created by policy, familiar monetary or fiscal policies may have little effect—a point getting too little attention in Washington.

The unemployment rate, the figure that dominates reporting on the economy, is the fraction of the labor force (those working or seeking work) that is unemployed. This rate has declined slowly since the end of the Great Recession. What hasn't recovered over that same period is the labor force participation rate, which today stands roughly where it did in 1977.

Labor force participation rates increased from the mid-1960s through the 1990s, driven by more women entering the workforce, baby boomers entering prime working years in the 1970s and 1980s, and increasing pay for skilled laborers. But over the past decade, these trends have leveled off. At the same time, the participation rate has fallen, particularly in the aftermath of the recession.

In one view, this decline is just a temporary, cyclical result of the Great Recession. If so, we should expect workers to come back as the economy continues to expand. Some research supports this view. A 2013 study by economists at the Federal Reserve Bank of San Francisco found that states with bigger declines in employment saw bigger declines in labor-force participation. It also found a positive relationship between these variables in past recessions and recoveries.

But structural changes are plainly at work too, based in part on slower-moving demographic factors. A 2012 study by economists at the Federal Reserve Bank of Chicago estimated that about one-quarter of the decline in labor-force participation since the start of the Great Recession can be traced to retirements. Other economists have attributed about half of the drop to the aging of baby boomers.

Baby boomers can't be the whole story, though, since the participation rate has declined for younger workers too. This part of the drop is a function of various factors, including simple discouragement, poor work incentives created by public policies, inadequate schooling and training, and a greater propensity to seek disability insurance. Globalization and technological change have also reduced employment and wage growth for low-skilled workers—which raises questions about whether current policy is focused enough on helping workers to achieve the skills necessary to work productively and earn decent incomes.

Figuring out which explanation best fits today's labor market is important because the different narratives point to different possible solutions. To the extent that labor-force participation and job creation have a cyclical element, activist demand policies by the federal government may make sense. Does this mean that the Obama administration's "targeted, timely and temporary" stimulus package was the right approach? Actually, no. Increasingly, it appears to have been a poor match for the severity of the downturn and the magnitude of the required boost.

After the Great Recession's sharp decline in investment and employment, U.S. business probably needed a more curative jolt to restore confidence. A sustained infrastructure program, rather than a temporary one for "shovel-ready" projects, would have provided more reassurance of longer-term demand. And far-reaching tax reform could have provided both a near-term fillip from front-loaded business tax cuts and a credible prospect for future growth.

What we don't know is whether the Obama's administration's activist policies failed to draw more Americans back to work because they were poorly executed or because they didn't do enough to raise aggregate demand. A better designed activist fiscal policy would have made more headway in encouraging growth, but deeper factors behind the downward shift in labor force participation still remain.

The Federal Reserve also has used monetary policy, through aggressive "quantitative easing," to combat the shock from the financial crisis. In assessing this move's effect on the labor force, a key question again is whether the problem is best seen as cyclical or structural. If labor-force participation is down because of cyclical factors, keeping interest rates low has been a smart policy, even as unemployment falls—in fact, even if it continues to fall to very low levels to draw nonparticipants back into the labor force.

Research by economists at the Federal Reserve Board published in 2013 suggests that bringing Americans back to work in this way might succeed without sparking inflation—if low labor-force participation is largely a result of a conventional downturn in business activity. If the real problem lies in the rules of the game—that is, structural factors accounting for labor force participation—such a highly expansionary monetary policy ultimately runs the risk of igniting inflation.

As I see it, the policy response to our disturbing doldrums in the labor market has indeed struck the wrong balance. Whatever can be said for shorter-term measures to jump-start job creation and business activity, it seems clear by this late date that our problems are in no small part structural. What we need most urgently is to rethink the federal government's wider role in the labor market. The importance of structural problems doesn't imply that policy can play no role beyond conventional fiscal or monetary policy.

The fierce debate now going on in Washington about extending unemployment insurance and raising the minimum wage largely ignores these issues. Such policies may affect the incomes of some Americans, but they won't do much to expand opportunity and bring more people back into the labor force. Sparking a broad-based return to the labor force demands a more ambitious agenda.

In the first place, we need to encourage low-wage workers and remove barriers to their lasting participation in the labor force. This encouragement is particularly important given the downward pressure on wages encountered by many low-skilled employees in the face of globalization and technological change. The Earned Income Tax Credit, which supplements the income of low-wage workers as they earn more, is supported by many conservatives and liberals alike. Expanding this program's payments for single workers (that is, beyond workers with families)—or using an alternative low-wage subsidy—would create more powerful work incentives. Phasing out the support over a longer income range, so that it provides more help to those who succeed and advance and reduces the marginal tax rate on work as the support phases out, also makes sense. These changes would cost money, but they could easily be accommodated in a broad tax-reform package.

Another priority for bringing low-wage workers back into the labor force is reforming disability insurance, which is part of the Social Security system. Since changes in qualifications in the 1980s made it much easier to receive federal disability payments, the percentage of individuals reporting disabilities who are still working has dropped by half. For some, disability insurance has become an incentive to give up on work—but it doesn't have to be this way. The program could be restructured to instead provide the employers of disabled employees with tax advantages for retraining them to remain on the job.

The Affordable Care Act, while giving some Americans access to health insurance for the first time, also creates certain disincentives for work. The law's generous private insurance subsidies phase out as income rises. In a recent study that I co-wrote with John Cogan and Daniel Kessler of Stanford University, we estimate that the amount of the federal subsidy can decline by as much as 50 cents for each dollar of additional earnings. This implicit tax comes on top of existing income and payroll taxes, raising the effective marginal tax rate on earnings to as much as 80% to 100% for some middle-income families. A broader tax reform that gives a more uniform subsidy for health insurance and health spending would reduce this problem.

A second broad area of policy in need of structural reform is unemployment insurance. Unemployment insurance was originally designed to provide income to workers during temporary spells of joblessness. Longer spells of unemployment during the Great Recession have led to continued calls to extend these benefits. Such extensions certainly keep income support in place longer, but they also lengthen spells of unemployment, potentially making workers less attractive to employers going forward.

A better approach would be a policy pivot toward easing the return to work. A first step would be to complement traditional unemployment insurance with block grants to states to support training and workforce development through community colleges and vocational education. Congress could also create Personal Re-employment Accounts for individuals. These accounts would give lump sums to individuals who lose their jobs and make it likelier that they receive some support and training during long periods without unemployment. If such individuals find a job quickly, they could keep some of the money as a re-employment bonus. Advancing and updating skills are also important: Funds currently in other federal training programs could be repurposed to provide this pro-work support.

Finally, in response to the profound change in the demographics of today's workforce, we really must consider eliminating the Social Security payroll tax on older workers. Today, older workers who delay retirement must keep paying Social Security taxes while receiving virtually no extra benefits—a strong incentive to stop working early. Getting rid of the payroll tax (currently 12.4% for Social Security) for older workers would remove this disincentive and increase employers' demand for older workers because employers pay half the employees' payroll tax.

In addition, Social Security reform should dispose of the "retirement earnings test," which reduces benefits by about 50 cents on the dollar on earnings above $15,000 for individuals aged 62 to 65. This heavy tax directly discourages work. These pro-work reforms to Social Security wouldn't be budget busters. Additional work by older Americans would produce income and Medicare taxes to offset much of the budget cost—while also slowing down the exit of workers from the economy.

None of the supply-side changes I've proposed would be easy to enact. They would require Democrats in Washington to confront the inadequacy of their stimulus policies to raise employment. And they would challenge many Republicans, who have focused their attention on economic growth, pure and simple, rather than on much-needed changes in federal labor policies. They will need to face up to the need for a more opportunity-oriented agenda for work, as Rep. Paul Ryan and Sen. Marco Rubio have argued, rather than simply opposing the extension of unemployment insurance or raising the minimum wage.

John Maynard Keynes once famously declared his fear that, at some point, much of humankind would have to cope with the problems of abundant leisure and little work. Perhaps. But we can no longer sit back and watch as growing numbers of Americans—not just the wealthy or the elderly—exit the labor force. This trend spells trouble for the nation's economic and fiscal future. It is a bigger and less understood problem than we think, and it requires bolder policy action than we have contemplated so far.

Mr. Hubbard, the dean of Columbia Business School, was chairman of the Council of Economic Advisers under President George W. Bush and an economic adviser to Mitt Romney's presidential campaign. The paperback edition of his book "Balance: The Economics of Great Powers From Ancient Rome to Modern America" (with Tim Kane ) will be released by Simon & Schuster in May.