The Collapse of Work in the Second Gilded Age
How is the U.S. economy doing these days? How are Americans themselves faring economically? These two closely related questions are central to any assessment of the well-being of our society and the health of our body politic. But these questions are more difficult to answer today than at any time in living memory.
This is not because our information-saturated era lacks facts and figures to take our nation's economic measure. Rather, it is because fundamental indicators of our country's economic outlook are far out of alignment with one another. Since the end of the twentieth century, the United States has witnessed an ominous and growing divergence among three trends that should ordinarily move together: wealth, output, and employment.
In terms of wealth creation, the twenty-first century appears to be off to a roaring start. It may look as if Americans have never had it so good and that the future is full of promise. Between early 2001 and late 2015, the net worth of U.S. households and nonprofit institutions almost doubled, rising to $87 trillion (see figure 1.1). In 2015, net worth averaged $270,000 per American — well over a million dollars per family of four. And this upsurge of wealth took place despite the terrible 2008 crash. In 2007, at the pre-crash apogee of estimated U.S. private wealth, total net worth of U.S. households and nonprofit institutions approached $68 trillion. Eight years later it was reportedly almost $20 trillion higher.
The U.S. economy also still looks like the world's unrivaled engine of wealth generation, notwithstanding the vaunted "rise of China." The Credit Suisse Global Wealth Report, for example, estimated that as of mid-year 2015, the United States possessed 34 percent of the entire world's personal ("household") wealth. China ran a distant second at 9 percent. U.S. wealth holdings also exceeded those of Europe in spite of the fact that Europe's population is well over twice as large.
The value of U.S. real estate assets is at or near all-time highs today, and U.S. businesses and corporations appear to be thriving. In the summer of 2016, the Wilshire 5000 Full Cap Price Index set a new record, with a total calculated capitalization of over $22.5 trillion. Since stock prices are strongly shaped by expectations of future profits, it appears investors are counting on the happy days continuing for some time to come.
Impressive as this upswing in measured wealth appears on paper, though, there is also an element of artificiality to it. From the 2008 crash to this day, the Federal Reserve has deliberately inflated U.S. asset values through its unprecedented and prolonged "zero interest rate" policies, interventions that are, unsurprisingly, proving difficult to unwind.
A less cheerful picture emerges if we look at macroeconomic trends. Here, U.S. economic performance since the start of the century might best be described as mediocre and its future prospects no better than guarded.
The 2008 crash brought a severe recession — the worst since the Great Depression — and the recovery has been painfully slow and unusually weak. According to the Bureau of Economic Analysis, it took nearly four years for U.S. gross domestic product (GDP) to regain its late 2007 level. By contrast, in the sharp Reagan-era slump, the recovery took just twenty-one months. Our "Great Recession" was somewhat more akin to the Great Depression, when it took seven years to get back to 1929 levels. As of early 2016, the total value added for the U.S. economy was barely 10 percent higher than before the 2008 crash (see figure 1.2).
The situation is even more sobering with respect to real per capita output. It took the United States until mid-2014 to return to its late 2007 per capita production levels. As of the first quarter of 2016, U.S. per capita output was barely 3 percent higher than it had been eight years earlier. America, it seems, has suffered something close to a "lost decade." And the snapback in per capita GDP since its mid-2009 low has averaged only 1.1 percent a year, barely half of our long-run annual per capita growth rate of 2.2 percent for 1947–2007 or 2.0 percent for 1987–2007. In other words, the U.S. economy currently is not nearly on track to return to its historic growth patterns.
Why is this recovery so much more fitful than other postwar recoveries? Some economists suggest the reason has to do with the unusual nature of the Great Recession. Downturns born of major financial crises intrinsically require longer correction periods than business cycle downturns. Others theorize that the scale of recent technological innovation is unrepeatable or that we have entered into an age of "secular stagnation" with low "natural real interest rates" consistent with significantly reduced investment demand.
What is incontestable is that the ten-year moving average for U.S. per capita economic growth is lower today than at any time since the Korean War and that this slowdown commenced in the decade before the 2008 crash. As a result, a consensus among economists has developed in recent years redefining the growth potential of the U.S. economy downward. The U.S. Congressional Budget Office, for example, suggests that the "potential growth" rate for the U.S. economy at full employment of production factors has now dropped below 2 percent a year, implying a sustainable long-term annual per capita economic growth rate of 1 percent or less.
The situation in the nation's labor force, for its part, is plainly awful (see fig 1.3). Between the start of the century and early 2016, the employment-to-population ratio ("work rate") for Americans ages twenty and older declined by over four percentage points. Postwar America has never experienced anything like this. From peak to trough, the collapse in work rates for U.S. adults in the Bush-Obama years was roughly twice what had been the country's previous worst postwar recession in the 1980s. At that time, it took America five years to regain the adult work rates recorded at the start of 1980. This time, over a decade and a half into our new century, the U.S. job market has scarcely begun to claw its way back to the 2007 work rates. As can be seen in figure 1.3, U.S. adult work rates never recovered entirely after the 2001 recession.
The country's work rates virtually flatlined in the four years after the Great Recession (late 2009 to early 2014). So far as can be determined, this is the only "recovery" in U.S. history in which this basic labor market indicator almost completely failed to respond.
The work rate has improved since 2014, but it would be unwise to exaggerate that turnaround. As of early 2016, our adult work rate was still at its lowest level in three decades. If our nation's work rate today were back to its start-of-the-century highs, approximately 10 million more Americans would currently have paying jobs.
Here, then, is the underlying contradiction of economic life in America's second Gilded Age: A period of what might at best be described as indifferent economic growth has somehow produced markedly more wealth for its wealth-holders and markedly less work for its workers. This paradox may help explain a number of otherwise perplexing features of our time, such as the steep drop in popular satisfaction with the direction of the country, the increasing attraction of extremist voices in electoral politics, and why overwhelming majorities continue to tell public opinion pollsters, year after year, that our ever-richer America is still stuck in a recession.
However bad our new employment profile may appear to the untutored eye, another facet looks even more dismal: employment trends for America's men. Male work rates have been in almost relentless decline, and not just since the dawn of the new century. Work rates for adult men have been falling for most of the post-World War II era. Between the early 1950s and today, the work rate for adult men has plummeted by more than eighteen percentage points. The drop since the Great Recession accounts for less than a quarter of the total long-term decline of twenty-plus employment-to-population ratios for U.S. men in the postwar era.
Many will find all this astounding. Others might object that I'm comparing apples and oranges here. After all, postwar America was an aging society, and older people tend to be out of the workforce. Would not a long-term fall in work rates exactly be expected in a prosperous and graying nation?
Alas, adjustments for changes in the postwar population structure do not come close to "correcting away" the collapse in male work rates. Even after appropriate corrections, work rates for U.S. men have still undergone a stunning decline. I shall detail the particulars of this sad saga in the following pages.CHAPTER 2
Hiding in Plain Sight: An Army of Jobless Men, Lost in an Overlooked Depression
Much current analysis of labor market conditions paints a cautiously optimistic — even unabashedly positive — picture of job trends. But easily accessible data demonstrate that we are, in reality, living through an extended period of extraordinary, Great Depression-scale underutilization of male manpower, and this severe "work deficit" for men has gradually worsened over time.
Expert opinions on U.S. labor market performance have been increasingly sanguine over the past year or so. A few select media headlines and quotations illustrate the emerging consensus:
"The Jobless Numbers Aren't Just Good, They're Great" (August 2015, Bloomberg)
"The Jobs Report Is Even Better Than It Looks" (November 2015, FiveThirtyEight)
"Healthy Job Market at Odds with Global Gloom" (March 2016, Wall Street Journal 3)
An excerpt from "Two Sides to Economic Recovery: Growth Stalls, While Jobs Soar" stated: "The job market, according to Labor Department figures released in recent months, is at its healthiest point since the boom of the late 1990s." (April 2016, International New York Times)
"June's Super Jobs Report (July 2016, Atlantic Monthly)
In addition, U.S. economists and policymakers who have served under Republican and Democratic presidents maintain that today's U.S. economy is either near or at "full employment":
"It is encouraging to see that the U.S. economy is approaching full employment with low inflation." (Ben Bernanke, former chairman of the Federal Reserve Board, October 2015)
"The American economy is in good shape ... we are essentially at full employment ... tight labor markets are leading to increases in hourly earnings and in the producer prices of services." (Martin Feldstein, former chair of the President's Council of Economic Advisers and longtime director of the National Bureau of Economic Research, February 2016)
"We are coming close to [the Federal Reserve's] assigned congressional goal of full employment. [Many measures of unemployment] really suggest a labor market that is vastly improved." (Janet Yellen, chairman of the Federal Reserve, April 2016)
All of these assessments draw upon data on labor market dynamics: job openings, new hires, "quit ratios," unemployment filings and the like. And all those data are informative — as far as they go. But they miss also something, a big something: the deterioration of work rates for American men.
The pronouncements above stand in stark contrast to the trends illustrated in figure 2.1, which track officially estimated work rates for U.S. men over the postwar era (see figure 2.1).
The federal government did not begin releasing continuous monthly data on U.S. employment until after World War II. By any broad measure, U.S. employment-to-population rates for civilian, noninstitutionalized men in 2015 were close to their lowest levels on record — and vastly lower than levels in earlier postwar decades.
Between 1948 and 2015, the work rate for U.S. men twenty and older fell from 85.8 percent to 68.2 percent. Thus the proportion of American men twenty and older without paid work more than doubled, from 14 percent to almost 32 percent. Granted, the work rate for adult men in 2015 was over a percentage point higher than 2010 (its all-time low). But purportedly "near full employment" conditions notwithstanding, the work rate for the twenty-plus male was more than a fifth lower in 2015 than in 1948.
Of course, the twenty-plus work rate measure includes men sixty-five and older, men of classic retirement age. But when the sixty-five-plus population is excluded, work rates trace a long march downward here, too. By 2015, nearly 22 percent of U.S. men between the ages of twenty and sixty-five were not engaged in paid work of any kind, and the work rate for this grouping was nearly 12.5 percentage points below its 1948 level. In short, the fraction of U.S. men from ages twenty-to-sixty-four not at work in 2015 was 2.3 times higher than it had been in 1948.
As for "prime-age" men — the twenty-five-to-fifty-four group that historically always has the highest employment — work rates fell from 94.1 percent in 1948 to 84.3 percent in 2015. Under today's "near-full employment" norm, a monthly average of nearly one in six prime-age men had no paying job of any kind.
Though the work rate for prime-age men has recovered to some degree since 2010, the latest report as of this writing (July 2016) is barely on par with the lowest-ever Bureau of Labor Statistics (BLS) reading before the Crash of 2008 (the depths of the early 1980s recession). In 2015, the proportion of prime-age men without jobs was over 2.5 times higher than in 1948. Indeed, 1948 work rates for men in their late fifties and early sixties were slightly higher than for prime-age men today.
Even more shocking is the comparison of work rates for prime-age men today with those from the prewar Depression era.
During the Depression era, we did not possess our current official statistical apparatus for continuously monitoring employment conditions. Our postwar statistical apparatus for continuously monitoring employment conditions only came in response to the prewar employment crisis. Consequently, our main source of information on Depression-era employment comes from our decennial population censuses. As fate would have it, the Great Depression spanned two national censuses, the 1930 census, near the start of the Depression, and the 1940 census, near its end. We contrapose male employment patterns then and now in table 2.1.
According the 1940 census, the work rate for civilian non-institutional men twenty-to-sixty-four years old was 81.3 percent. In 2015, that rate was 78.4 percent. The work rate for prime-age males in 1940 was reported to be 86.5 percent, two points higher than in 2015 and about a point and a half higher than readings thus far for 2016. In other words, work rates for men appear to be lower today than they were late in the Great Depression when the civilian unemployment rate ran above 14 percent. Furthermore, the work rate for American men is manifestly lower today than it was in 1930, to judge by returns from the 1930 census.
Admittedly, the comparison is not straightforward, since the 1930 census used different questions about employment status than we use today and did not break out "civilian noninstitutional population" from the total adult population. Nonetheless, the Census Bureau has harmonized those 1930 employment figures with modern definitions of work and joblessness. By these reconstructions, the 1930 ratio for employment to total population for men twenty-to-sixty-four was over 88 percent. Among men twenty-five to forty-four (prime work ages for that era) the ratio for employment to total population was over 91 percent. In 2015, the official work rate for working-age men twenty-to-sixty-four was nearly ten percentage points below this 1930 figure (78.4 percent vs. 88.2 percent) and for men twenty-five to forty-four, the nominal gap was nearly six points (85.3 percent vs. 91.2 percent). These numerical differences, I should note, understate slightly the true work rate gap between adult men in 1930 and today, since the 1930 numbers do not exclude men in the armed forces, prisons, long-term hospitalization, etc., from the demographic denominator by which current work rates for the "civilian noninstitutional" population are calculated.
To be clear, the employment disaster in the depths of the Great Depression was unquestionably worse than it was in either 1930 or 1940. For better or worse, however, we only have these two census data points for that era's labor market conditions, and current data indicate that work rates for American men are lower today than in either of these years. It is thus meaningful to talk about work rates for American men today as being at Depression-era levels. In fact, they are more depressed than those recorded in particular years of the Great Depression.
Just how great is our current "work deficit" for American men? One reasonable benchmark for measuring that gap might be the mid-1960s. Then, the U.S. economy was strong and labor markets functioned at genuinely full employment levels.