Bogeyman Economics
January 27, 2012
In this moment of economic challenge, it can be difficult to keep our problems in perspective. The scale of the financial crisis and the subsequent recession, the weakness of the recovery, the persistence of high unemployment, and the possibility of yet another shock — this time originating in Europe — have left Americans feeling deeply insecure about their economic prospects. Unfortunately, too many politicians, activists, analysts, and journalists (largely, but not exclusively, on the left) seem determined to feed that insecurity in order to advance an economic agenda badly suited to our actual circumstances. They argue not that a financial crisis pulled the rug out from under our enviably comfortable lives, but rather that our lives were not all that comfortable to begin with. A signal feature of our economy in recent decades, they contend, has been pervasive economic risk — a function not of the ups and downs of the business cycle, but of the very structure of our economic system. According to this view, no American is immune to dreadful economic calamities like income loss, chronic joblessness, unaffordable medical bills, inadequate retirement savings, or crippling debt. Most of us — “the 99%,” to borrow the slogan of the Occupy Wall Street protestors — cannot escape the insecurity fomented by an economy geared to the needs of the wealthy few. Misery is not a marginal risk on the horizon: It is an ever-present danger, and was even before the recession. But compelling though this narrative may be to headline writers, it is fundamentally wrong as a description of America’s economy both before and after the recession. When analyzed correctly, the available data belie the notions that this degree of economic risk pervades American life and that our circumstances today are significantly more precarious than they were in the past. Even as we slog through what are likely to be years of lower-than-normal growth and higher-than-normal unemployment, most Americans will be only marginally worse off than they were in past downturns. The story of pervasive and overwhelming risk is not just inaccurate, it is dangerous to our actual economic prospects. This systematic exaggeration of our economic insecurity saps the confidence of consumers, businesses, and investors — hindering an already sluggish recovery from the Great Recession. It also leads to misdirected policies that are too zealous and too broad, overextending our political and economic systems. The result is that it has become much more difficult to solve the specific problems that do cry out for resolution, and to help those Americans who really have fallen behind. Only by moving beyond this misleading exaggeration, carefully reviewing the realities of economic risk in America, and restoring a sense of calm and perspective to our approach to economic policymaking can we find constructive solutions to our real economic problems. INCOME AND EMPLOYMENTPerhaps the broadest measure of economic insecurity is the risk of losing a job or experiencing a significant drop in income. And the idea that this risk has been increasing dramatically in America over the past few decades has been absolutely central to the narrative of insecurity. It has fed into a false nostalgia for a bygone age of stability, one allegedly supplanted (since at least the 1980s) by an era of uncertainty and displacement. President Obama offered a version of this story in his 2011 State of the Union address:
Many people watching tonight can probably remember a time when finding a good job meant showing up at a nearby factory or a business downtown. You didn’t always need a degree, and your competition was pretty much limited to your neighbors. If you worked hard, chances are you’d have a job for life, with a decent paycheck and good benefits and the occasional promotion. Maybe you’d even have the pride of seeing your kids work at the same company. That world has changed. And for many, the change has been painful. I’ve seen it in the shuttered windows of once booming factories, and the vacant storefronts on once busy Main Streets. I’ve heard it in the frustrations of Americans who’ve seen their paychecks dwindle or their jobs disappear — proud men and women who feel like the rules have been changed in the middle of the game.
The tales of both this fabled golden age and the dramatic rise in the risk of declining incomes and job loss are, to put it mildly, overstated. But this popular story did not originate with President Obama: It has been a common theme of left-leaning scholars and activists for many years. The clearest recent example of this trope may be found in the popular 2006 book The Great Risk Shift, by Yale political scientist Jacob Hacker. Hacker’s fundamental argument was that economic uncertainty has been growing dramatically since the 1970s, leaving America’s broad middle class subject to enormous risk. Using models based on income data, he argued that income volatility tripled between 1974 and 2002; the rise, he claimed, was particularly dramatic during the early part of this period, as volatility in the early 1990s was 3.5 times higher than it had been in the early 1970s. Hacker’s conclusion — that the middle class has, in recent decades, been subjected to horrendous risks and pressures — quickly became the conventional wisdom among many politicians, activists, and commentators. As a result, it has come to define the way many people understand the American economy. But that conclusion turned out to be the product of a serious technical error. Attempting to replicate Hacker’s work in the course of my own research, I discovered that his initial results were highly sensitive to year-to-year changes in the small number of families reporting very low incomes (annual incomes of under $1,000, which must be considered highly suspect). Hacker was forced to revise the figures in the paperback edition of his book. Nevertheless, he again overstated the increase over time by reporting his results as a percentage change in dollars squared (that is, raised to the second power) rather than in dollars and by displaying his results in a chart that stretched out the rise over time. While he still showed an increase in volatility of 95%, my results using the same basic methodology indicated an increase of about 10%. In July 2010, a group led by Hacker published new estimates purporting to show that the fraction of Americans experiencing a large drop in income rose from about 10% in 1985 to 18% in 2009. But the 2009 estimate was a rough projection, and would have been a large increase — up from less than 12% in 2007. Then, in November 2011, an updated report (produced by a reconfigured team of researchers led by Hacker) abandoned the previous year’s estimates and argued that the risk of a large income shock rose from 13% or 14% in 1986 to 19% in 2009. Where did these assertions come from? The team’s 2010 claim was again based on a failure to adequately address the problem of unreported income in the data. Their November 2011 claim, meanwhile, used a different data set that was much less appropriate for looking at income loss (because it does not identify the same person in different years, does not follow people who move from their homes, and suffers greatly from the problem of unreported income). In fact, the most reliable data regarding income volatility in recent decades (including the data used by Hacker in his early work and in 2010) suggest a great deal of stability when analyzed correctly. The chart below shows the portion of working-age adults who, in any given year, experienced a 25% decline in inflation-adjusted household income (a common definition of a large income drop, and the basis for Hacker’s recent estimates). Viewed over the past four decades, this portion has increased only slightly, even though it has risen and fallen within that period in response to the business cycle…
