Fighting Over Numbers in the Public Debate
A pair of lunchtime-reading items are unified by their disputing of some of the studies (or, “facts”) being batted about in the public debate. They also share income as their primary subject matter.
First, James Pethokoukis summarizes the findings of Cornell University researchers that, from 1979-2007, median household income rose 36.7%, not the 3.2% number underlying President Obama’s claims of a growing economic disparity:
See, Piketty and Saez made lots of odd choices about what to measure and how to measure it [in order to achieve the 3.2% number]. They chose to measure something called “tax units” rather than households, a move which ignores the statistical impact — including economies of scale — of couples who cohabitate, kids who move back in with their parents after college, and senior parents who live with their adult children.
They chose to ignore the value of all government transfers — including welfare, Social Security, and other government provided cash assistance — received by the household.
They chose to ignore the role of taxes and tax credits.
They chose to ignore the value of healthcare benefits. In short, Piketty and Saez ignored a lot of stuff.
Basically, what the researchers argue is what a lot of people suspect to be true just by basic experience of the past few decades: Those at the top have gotten richer at a faster rate, but it’s not accurate to claim that the middle stagnated and the poor got poorer.
Another group that has been doing very well is that of public-sector workers. In an article in Reason magazine, Tim Cavanaugh disputes studies that have claimed to find public-private sector pay not out of whack:
These studies are false on several levels. The most basic of the misdirections is that they treat education — rather than the actual work you do — as determinative. Public school teachers are more likely to hold state-approved credentials than private school teachers (and they make 37 percent more, according to the U.S. Department of Education). Would anybody claim they do a better job of teaching children?
Most of these studies, though not all, also ignore the value of job security: The layoff rate of public workers is about one-third that of private-sector workers. And virtually none of the studies accounts for the easier and lighter schedules of government stiffs: Public workers work 1,825 hours a year vs. 2,050 hours for private workers, according to the Cato Institute’s Chris Edwards.
Even the central claim about higher rates of education may be bogus. In January the Congressional Budget Office, comparing the compensation of federal and private-sector employees, found it is actually the least educated public workers who get the biggest pay bump. Public workers with a high school degree or lower make 21 percent more in wages than equivalent private workers; those with less than a bachelor’s degree earn 15 percent more; and those with a bachelor’s degree receive 2 percent more. Only at the level of master’s degrees and higher do private wages outpace public. (Benefits are much higher for public workers at all education levels.) In federal work, it’s lower educational attainment that gets rewarded most.
To be fair, as with all batches of statistics, conclusions depend a great deal on the way one cuts the categories. Cavanaugh’s opposition would likely argue that the point that he makes in the first paragraph contradicts the point that he makes in the third. That is, public-sector workers with higher degrees are closer to parity with their private-sector peers, which (they’ll cede) is reasonable. What throws the comparison off, they’ll say, is that a higher percentage of public-sector work requires people with advanced education, so the workforce’s average salary goes up.
That is why Cavanaugh should have expounded upon his first point. Looking specifically at public-school teachers, anybody who has perused a few of their contracts will likely have noted the strong incentive built in to their collectively bargained pay structures to continue taking classes. Often, the degrees don’t even have to relate to the subject matter that they teach, and sometimes, they’ll get tuition reimbursement or even years of “sabbatical” during which to attend classes full time. (As the Current’s analysis of Woonsocket teachers’ contract highlights, teachers receive partial pay while studying and then return to work not only with at higher pay scale based on their degrees, but also at the higher pay scale of annual steps as if they had never left.)
The trick with numbers in public policy is never to lose sight of the underlying principles or human questions. Do we care whether one group achieved a higher percentage point, or do we care whether the society has improved conditions for all? Do we care that public-sector workers have more advanced degrees, or do we care that they make more for doing the same work?
One suspects that the big-government types sense that public opinion is against them when questions are phrased thus, so they use numbers to prove that people’s eyes are lying.