Other than vague generalities about turning corners, you don't hear a lot these days from the Republicans on how middle-income American families are doing. The strategy of the Bush campaign is to blame most of our economic woes on September 11, which brings the electorate right back to fear and terror, not changing horses, and painting the other guy as too indecisive to take it to the enemy.
Thus, when conservative economist Kevin Hassett of the American Enterprise Institute writes an article dismissing the notion of a middle-class squeeze, it's worth taking notice.
His article, “The ‘Squeeze' Play,” marshals an array of statistics to argue that middle-class families are doing fine and will thus support George W. Bush on November 2. To get there, though, Hassett omits some very inconvenient facts.
He makes two basic points. First, the economy is expanding, so middle-income families must be getting ahead. He asks, “If the economy on average is getting richer and richer, how exactly can the middle class be squeezed at the same time?”
Second, he points out that consumption is up, even for middle-income families, and therefore all must be well.
On the first point, Hassett makes the mistake of assuming that middle incomes move with the overall average. In fact, over the past few decades, rising inequality has often served as a wedge between overall growth and the living standards of middle-income families. As a recent Economic Policy Institute report shows, household incomes fell for middle-income households from 2000 to 2003 by 3.4 percent, or $1,500 in 2003 dollars, even though the gross domestic product increased by 5.7 percent over these years.
Note that this three-year decline in real median incomes was front-page news in both The New York Times and the Washington Post just a few weeks ago, yet it does not rate a single mention in Hassett's piece.
In fact, Hassett argues that income is not the best way to measure how middle-income families are doing. He points out that consumption grew in real terms by 2 percent for such families from 2001 to 2002. But this argument ignores a huge stumbling block for his story: the increase in indebtedness. If households are buying more while their incomes are falling, they must be taking on more debt. And, in fact, the data reveal historically high debt burdens over this period. What's more, rising interest rates and weak income growth will make those debts harder to pay off.
Though the most recent data are not broken down by income class, we know that household debt burdens as a share of disposable income are at their highest levels on record (going back to 1980). The most recent data that do break such things out by income level show, not surprisingly, that middle-income families are eight times more likely than high-income families to have high debt burdens (i.e., debt service payments equal to more than 40 percent of household income).
By ignoring both real middle-income trends and expanding household debt, Hassett undermines his case that there is no middle-income squeeze.
Hassett also tries to put a negative spin on the Clinton years by pointing out that real wages grew more slowly over Bill Clinton's first term than under that of Bush Junior. But that's just a function of how Hassett averages over the business cycle. Real wages were stagnant coming out of the early '90s jobless recovery, and they began to turn upward when Clinton ran for his second term.
The wage momentum from the latter '90s boom kept real wages rising through the 2001 recession and ensuing jobless recovery, but by mid-2003, the persistently weak job market took its toll on wage growth, and real wages started falling. (Over the past two months they've ticked up again, but they're still lower than they were a year ago.)
But if you really want to understand the difference between Clinton 1996 and Bush 2004 with regard to middle-income families, you can't ignore what is arguably the most relevant indicator: the real income of the median household. By the time Clinton ran in 1996, median household income was rising solidly in real terms, up 5 percent from 1992. During the Bush years, in contrast, such households have lost ground every year.
It's also the case that productivity has been rising much more quickly now than it was a decade ago. Thus, the wages and incomes of middle-income households are not only stagnant or falling in real terms (relative to inflation); they're falling much more relative to the growth in productivity. Between 1992 and 1996, both productivity and real median family income were up by 5 percent; from 2000 to 2003, productivity's up a stellar 12 percent, but median household income is down 3 percent.
This widening chasm between overall economic growth and the fortunes of middle-income families means that one cannot assume, as Hassett does, that such families' fortunes must be rising along with the general tide.
Given these facts, it's no surprise that some middle-income families feel squeezed, especially when the costs of college and health care are rising so quickly. Here again, Hassett is off point. He notes that financial aid is up but fails to point out how quickly tuitions are rising. Since 2000, the price index for college tuition is up by 38 percent, compared with 12 percent for the overall price index. A group called the Higher Education Research Institute has shown that this trend is leading to an ever-higher concentration of students from high-income families (with income above $100,000) at select universities.
Hassett's reading of carefully selected facts leads him to proclaim, “The consumption data and the topline GDP numbers suggest that the middle class has been doing fine” -- and to conclude, therefore, that, “If GDP is growing, voters are happy.”
Assessing how middle-income families are faring is just not that simple, and there are clearly voters who aren't happy about the fact that their incomes are down and their debts are up.
As long as we're inveighing against those who get such matters wrong, Slate's Jack Shafer also deserves a swipe for his critique of a recent Washington Post piece on middle-class economic issues.
Shafer says that the Post piece presents some data that seem to contradict its own very detailed and eloquent story of the middle-class squeeze in action. A figure accompanying the article points out that the share of households with incomes between $35,000 and $50,000 a year (in 2003 dollars) has shrunk from 22 percent in 1967 to 15 percent in 2003, suggesting a contracting middle class. As Shafer correctly notes, the figure shows that over time, families are shifting from lower- to higher-income brackets, suggesting uniformly improving living standards. In fact, according to Shafer, the data reveal that the nation is “ … in the count-down stage of an unprecedented economic boom with great jobs and material success for almost everybody.”
The problem here is that the shift of households to higher real-income levels over the long term is about the lowest imaginable benchmark of success. As the economy grows, wouldn't we expect families to be better off in absolute terms, i.e., wouldn't we expect there to be fewer families in lower real-income brackets that are fixed in real terms? In fact, unless all the income growth were wholly concentrated among the wealthy, it is inevitable that families will move as shown in the figure. To applaud that as success is like saying that the environment must be fine because the sun comes up every day.
Since, as noted above, the growth of income inequality diverts income away from the bottom half, the question is, how quickly are families passing through these real brackets in different time periods? (The Post article is quite clear on this point, acknowledging absolute gains but noting that the growth has disproportionately accrued to the wealthy.)
The data only go back to the latter 1960s, but between 1969 and 1979, when inequality growth was just getting underway, 6.5 percent of households shifted from less than $50,000 a year to more than $50,000. Over the next 16 years, 1979 to 1995, as inequality took off, that shift slowed to 4.1 percent of households. In annualized terms, households moved up the income scale 40 percent as quickly in the latter period before picking up pace again, from 1995 to 2000, as tight labor markets dampened the growth of inequality. Over the last three years, 2000 to 2003, the shift has reversed, and there's been a 1.5- percent shift from households earning more the $25,000 a year to those earning less. Does that sound like “great jobs and material success” for all?
The punch line for all of the above: This has been a highly unbalanced recovery thus far, and you will learn little about how middle-income families are doing by looking at the GDP and consumption growth. Middle-incomes are down in real terms, and they've diverged widely from productivity growth, households are heavily indebted, and key prices -- college, health care, energy -- have been rising quickly.
Of course, it's far from a sure bet that these households will vote their pocketbooks on November 2, but if they do, even the most careful cherry-picking won't help the incumbent.
Jared Bernstein is a senior economist at the Economic Policy Institute in Washington, D.C.