Last spring I sat on a panel convened to discuss the publication of The Occupy Handbook, a collection of essays in which a number of prominent economists addressed issues raised by the Occupy movement, in particular the stagnating wages and rising inequality that have defined our economy for the past generation. On these points the panelists shared the occupiers’ concerns, and our conversation soon turned to how such problems might be remedied. One of the panelists, Raghuram Rajan, of the University of Chicago, suggested that the way to return America to prosperity is to educate more of its citizens. His message to those in the audience was: Go to college.
The crowd could have been excused for responding with some skepticism. Anyone who spent much time around Occupy heard about the poor job prospects of debt-ridden recent graduates. Yet when mainstream economists seek to explain the plight of American workers, they blame lack of education to the exclusion of nearly everything else. As technologies advance, the orthodoxy tells us, our economy requires workers with more sophisticated training (this development is termed “skill-biased technological change”), a need that must be met with better education. “The rise and decline of unions plays a supporting role in the story,” write the Harvard economists Claudia Goldin and Lawrence Katz in a typical example of such thinking, “as do immigration and outsourcing. But not much of a role. Stripped to essentials, the ebb and flow of wage inequality is all about education and technology.”
Rajan is no ordinary economist. As far back as 2005, he argued that Wall Street was so large, so interconnected, and so heavily indebted that its failure could bring down the entire economy — an idea at which the likes of Alan Greenspan and Larry Summers scoffed. But when asked now about solutions to the crisis he predicted, Rajan sounds like any other economist. Growing inequality, he wrote in his lauded 2010 book Fault Lines, “stems primarily from the gap between the demand for the highly educated and their supply.”
Economists love these single-cause theories too well, especially when they reaffirm the core beliefs that free markets work and that government policies are not the answer. But practical realities have seriously challenged the pro-education argument time and again. This spring, the economics-data firm Sentier Research showed that the typical American household has seen its income fall every year since the Great Recession ended in 2009. That just isn’t supposed to happen during an economic recovery. Moreover, in the post-tech-bubble expansion that preceded the recession, household incomes never regained the high mark they set in 2000. The percentage of American adults with college degrees, meanwhile, is greater than ever, having grown in the past decade from 26 to 30 percent. Every year our country is better educated, while wages remain stagnant or fall.
I don’t mean to suggest that young Americans should steer clear of college. The odds they will eventually get a job are much higher if they first get a degree. But skill-biased technological change is not the main driver of inequality or wage stagnation in the United States, and education will not be the main solution.
Economists have long pointed to the earnings gap between the average worker with a college degree and the average worker with only a high school diploma as evidence that advanced technology requires an advanced labor force. This might have been a persuasive argument in the 1980s, when that gap grew rapidly, but since then it has grown much less quickly. As noted, those who lack a college degree are likelier to be unemployed than those who possess one — this is a natural outcome of increasing educational attainment: college graduates make more attractive job candidates no matter the skills required in a particular job. But in the absence of sufficient high-skill positions, the number of college graduates doing work once done by those with only a high school education has risen dramatically. “The college degree is becoming the new high school diploma,” the New York Times reported in February, meaning that it was now the “minimum requirement . . . for getting even the lowest-level job.” Not surprisingly, average income among those with degrees has remained more or less flat in that time, and income inequality continues to be high — some get good jobs, but many don’t. This “bumping down” has helped keep unemployment among college graduates from rising, but it has also reduced their wages: hourly earnings for the majority of college graduates declined in the 2000s. If the United States were to produce the additional graduates economists are demanding, average wages among them would likely fall even lower.
Faced with this evidence, orthodox economists have switched gears — but only slightly. Now their favored term is “job polarization.” At one end of the modern employment market, they argue, are well-trained specialists whose jobs are both “nonroutine” and “cognitive” in nature — lawyers, doctors, financiers. At the other end are unskilled workers, often without high school diplomas, whose jobs are also nonroutine but are manual in nature — waiters, hospital orderlies, security guards. Wages are up at both ends, because such tasks can’t be easily automated or shipped overseas. But between these two ends are many, if not most, high school and college graduates, in positions as middle managers, bank clerks, and the like, positions that are easily automated and easily outsourced.
The polarization theory — up at each end, down in the middle — caught on quickly in the press. It seemed plausible, especially when advanced by well-regarded economists such as MIT’s David Autor. But the theory lacks substantiating evidence. Unfortunately for Autor and his colleagues, there simply wasn’t any job polarization to speak of in the 2000s — no increase in the demand for nonroutine workers, no decline in the demand for routine workers, no technological shift requiring more highly qualified job candidates. Yet wage inequality kept rising.
This is to say nothing of the top 1 percent of earners, whose incomes rose sharply, a trend even more pronounced among the top 0.1 percent. Privileged “cognitive workers” constitute the top 20 to 30 percent of the earners in Autor’s analysis. But over the past several decades, average CEO pay rose exponentially compared not only with that of the general population but also with that of nonroutine workers at both ends of the market. A tiny group of workers has enjoyed all the benefits of economic growth while everyone else treads water. This problem simply can’t be explained by invoking abstract thinking and nonroutine work, and so more education can’t be the way to fix it.
What is driving inequality? The labor economists David Card and John DiNardo have shown that in the postwar period inequality rose fastest between 1980 and 1986, before rapid advances in computer technology began to affect the job market. And as computer use exploded in the 1990s, inequality stopped widening. This suggests that technological change is less to blame than lost bargaining power.
Unemployment was high throughout the 1980s and much of the early 1990s, and has been again since 2008. It seems clear that, even among the well-educated, the fear of unemployment deters workers from demanding wage hikes, particularly when joblessness is pervasive. And now that unions represent no more than 7 percent of private-sector workers, there are even fewer bargaining protections in place than there once were. The relatively low minimum wage is another significant factor often dismissed by mainstream economists. Although the federal minimum wage was raised in 2009 to $7.25 an hour, it stands below its 1967 level in terms of purchasing power.
Then, of course, there is the loss of domestic manufacturing jobs, owing to both the construction by American companies of factories overseas and the preference of American consumers for cheap goods produced abroad. In 1979, the manufacturing sector employed nearly a quarter of the U.S. workforce, usually providing relatively well-paid jobs; by 2011, that proportion had fallen to 8.9 percent. Autor and his colleagues have themselves shown that the offshoring of routine work is a factor in overall wage stagnation, yet they recommend no ways of dealing with it.
As for the continued success of the One Percent, much of their ongoing gain can be attributed to financialization, as speculation and market-making have surpassed manufacturing as the engine of the American economy. Investment bankers continue to rake in huge bonuses, and CEOs are still plied with stock options, which have soared in value along with the stock market.
Bad economics has real consequences. As John Schmitt of the Center for Economic and Policy Research notes, the polarization argument provides zero room, beyond the expansion of education subsidies, for a government policy response. Autor doesn’t even mention unemployment levels in his attempt to explain inequality. Mainstream economists are disturbingly wedded to an ideology that fails to take into account the fact that labor markets can fail or that workers can be abused.
Several bills currently under consideration in Congress do seek to address these issues. Last year, Senator Tom Harkin (D., Iowa) introduced the Rebuild America Act, which funds greater investment in infrastructure and education training, closes loopholes that encourage businesses to invest abroad, raises fines on companies that violate labor law by paying less than minimum wage or denying employees the right to unionize, and enables more workers to qualify for overtime and paid sick leave. In the House of Representatives, the Illinois Democrat Jan Schakowsky is sponsoring a bill that very much resembles FDR’s Depression-era employment programs and would create more than 2 million jobs by providing funding for the rehabilitation of public schools, the improvement of parks, and the hiring of early-education teachers, police officers, and firefighters. The Congressional Progressive Caucus has in its new budget proposal put together perhaps the most ambitious of these programs, allocating federal money to education and infrastructure, state aid, and tax cuts for the working middle class.
Given the evidence that more education alone won’t fix our stagnant wages or rising inequality, one might expect at least some economists to support policies like the ones advocated by Harkin, Schakowsky, and the Congressional Progressives. But usually they do not, relying instead on the assumption that efficient labor markets will sort themselves out. Set aside the absence of evidence for this assumption — in my ideal world, mainstream economists would occasionally invoke economic justice as well as economic efficiency.