I share Robert Pollin’s view that the United States should strive for full employment—by which I mean, following his lead, an unemployment rate below 4 percent.

Can we do it? Pollin points to two historical precedents as grounds for optimism. The first is Sweden from 1960 to 1989. Sweden succeeded in keeping unemployment below 4 percent throughout those three decades by coupling employment-oriented monetary and fiscal policy with wage restraint. But Sweden’s central bank at that time was subordinate to the government. Ours, the Federal Reserve, is independent. Since the late 1970s, independent central banks such as the Fed have almost always prioritized low inflation, rendering low unemployment difficult to achieve. If the Fed isn’t onboard, even a workable plan for full employment supported by the American public and elected officials probably won’t be enough.

What about Pollin’s second precedent, the United States in the late 1990s? During those years the Fed, under Alan Greenspan, did keep interest rates low enough for the unemployment rate to drop below 4 percent. But Greenspan held rates low despite opposition from other Fed board members, who were concerned about potential inflationary consequences—particularly given the Internet-driven stock market bubble. Greenspan took this stance in part because his belief in the self-correcting nature of markets led him to worry less than others about the bubble. In light of the painful consequences of the 2000s real estate bubble, I doubt we’ll see the Fed take that approach again for some time.

Even if sub–4 percent unemployment is possible, do we need it? Here a cross-national perspective is instructive. The following charts show indicators of Pollin’s desired outcomes—decent pay, low poverty, good working conditions—in twenty rich democratic nations. Each outcome is plotted against the number of years from 1979 to 2007 in which each country had sub–4 percent unemployment.

(Click image to view larger in a new window or tab.)

Sources: Eurofund’s European Working Conditions Survey, Luxemburg Income Study, OECD, U.S. Bureau of Labor Statistics. Dollar amounts are in 2007 U.S. Dollars. Graphic by George Restrepo.

These charts tell us that while full employment may contribute to good outcomes, it isn’t a necessary condition. In each case some countries have done well despite seldom or never reaching sub–4 percent unemployment during the measurement period. In certain instances this is a function of strong unions or “production regimes” (think German manufacturing) that are unlikely to be relevant in the American context. In others, though, successful outcomes have owed much to government action.

This is good news because Americans have more influence on the policy choices of the government than on those of the Fed. Whether or not we return to full employment, we can reach important economic and social goals.

Yet I fear this conclusion is too optimistic. I’m confident that the United States could achieve satisfactory economic growth, a reasonably high employment rate, decent wages, poverty reduction, good working conditions, and less discrimination even in the absence of full employment. I’m less convinced that the country can manage sustained wage growth for those in the bottom half of the distribution.

The post–World War II experiences of the rich democracies suggest three routes to rising working- and middle-class wages. One is an environment in which firms face only moderate competition in product markets and limited pressure from shareholders, allowing them to pass on a significant share of growth to their employees. This characterized the period from the late 1940s through the mid-1970s, but it’s now long gone. The second is strong unions. I see little hope of that in America’s future. The third is full employment.

Is there an alternative? One possibility is to use the Earned Income Tax Credit to subsidize wages. Congress could extend it higher in the income distribution (currently it phases out at about $45,000), reduce its connection to children (currently it’s minuscule for households with no kids), and index it to average wages (it’s now indexed to inflation).

I would prefer the full-employment path that Pollin envisions, in which wage growth comes from firms rather than from taxpayers. But we ought to have a backup plan.