RALEIGH – If you read only one academic study on income inequality this year, let it be Robert Gordon’s new working paper for the National Bureau of Economic Research.
Though I can’t imagine why you would want to read just one paper on this fascinating and important subject.
Gordon, an economist at Northwestern University, has been studying income trends and related topics for decades. His latest work explores three topics that ought to be of interest to anyone who wants to gain a better understanding of issues such as economic change, tax policy, and the development of human capital via education and health care.
The first topic is the proper measurement of income trends. Gordon shows that some apparent disparities in income are simply due to the composition of the denominator of the fraction. If you divide income by household, you have to keep in mind that the size of households has changed over time. If the average household in a given region received $45,000 in income 20 years ago and $43,000 last year, that doesn’t signify a decline in income if the household had four people in it 20 years ago and three people last year. You should at least supplement your household data with a per-capita measurement.
The second topic is the proper measurement of real income. At the John Locke Foundation, we’ve been stressing this point for virtually the entire 20 years of our existence: simply adjusting every figure in sight by the Consumer Price Index is a big mistake. CPI overstates inflation in a variety of now-recognized ways. Using the Gross Domestic Product price index is far better, and shows greater real increases in personal income over time.
Even more importantly, national price indexes do not begin to help put real income trends in perspective when you consider that neither living costs nor Americans are equally distributed across the 50 states and thousands of local communities. People with $100,000 incomes in New York City live very different lives, and paying much higher prices for a range of basic goods and services, than do people with $100,000 incomes in, say, Asheboro.
While this is obviously an important factor to consider when making interstate comparisons, that’s not the only implication. In his research, Gordon found that people who obtain higher education tend to end up living in higher-cost jurisdictions. They’re certainly free to do that, and presumably they know what occupational and residential arrangements will best meet their needs. But studies that fail to adjust for cost-of-living variances can seriously mislead policymakers and the public.
In Gordon’s paper, he cites research showing that as much as two-thirds of the apparent “wage premium” of higher education – the additional income that, it is argued, is caused by getting a college education – disappears when the income data are properly adjusted for living costs. That’s huge.
“Newly released cross-state cost of living data,” Gordon continues, “yield the complementary conclusion that the excess of per-capita income in New York state over the national average vanishes when translated from nominal to real terms and that in Massachusetts falls by half.”
Finally, the third topic Gordon explores in his paper is perhaps the most important of all: causality. The political rhetoric about the “rich getting richer and the poor getting poorer” is usually based on the assumption that certain policy decisions in the past – whether to cut taxes, whether to raise the minimum wage, etc. – have had a significant effect on subsequent income trends. But before you ever get to questions of what caused what, you have to at least show that the timing of events makes sense. Unless you are into some pretty weird quantum physics, cause must precede effect.
Gordon has found that, for many of the most popular explanations of income trends and inequality, the timing doesn’t really work when you examine the data rigorously. In sum, he writes, “The rise of American inequality has been exaggerated in magnitude, and its impact is now largely in the past.”
If you don’t want to subscribe to NBER, you can read the paper in pdf form here. Gordon is by no means a free-market purist and offers recommendations that some of us won’t like. But his discussion of the data is fascinating and highly recommended.
Hood is president of the John Locke Foundation