Don Boudreaux, over at CafeHayek, has been posting about his debate with Branko Milanovic over whether middle class stagnation is a myth. I have some thoughts to add to that debate. I’ll do so at the end. But reading Milanovic’s comments reminded me of something he wrote in 1996 that I challenged in an article co-authored with my then colleague Robert McNab and my former student from Hungary Tamas Rozsas. The article is “The Hidden Inequality in Socialism,” The Independent Review, Winter 2005.

Here’s that segment of the article:

Referring to the effect of subsidies on essential items and to the quality of vacation homes for the top party brass, Milanovic argues that these privileges would not have altered measured income inequality to a great extent. He claims that others exaggerated the value of these privileges: “Elite privileges were exaggerated both by |the] indigenous population, because of the secrecy in which privileges were held, and by overly credulous Western analysts. In effect [as] anybody who has visited vacation homes previously kept strictly off-limits for all but the top Party brass can testify, their level of comfort and service is below that of an average Holiday Inn” (1996, 200). However, in dismissing the value of a vacation home by comparing it unfavorably to a Holiday Inn, Milanovic is, wittingly or not, implicitly appealing to Western standards. Although few middle-class Americans will regard a Holiday Inn as a luxury hotel, Americans are not the relevant group here; eastern Europeans are. In the mid-1970s, living space per person in the Soviet Union was approximately only 120 square feet (W. S. Smith 1973, 405, as cited in Pejovich 1979, 55). Every room at a Holiday Inn has its own bathroom, whereas in the early 1970s approximately half of all Soviet housing lacked running water or plumbing, and much of the other half shared bathroom facilities with other families (Pejovich 1979, 55-56). For almost anyone in the eastern European socialist countries, a Holiday Inn would have been the height of luxury. To “translate” Milanovic’s statement for Western ears, it would need to read something like this: “In effect, as anybody who has visited vacation homes previously kept strictly off-limits for all but the top Party brass can testify, their level of comfort and service is below that of an average Hyatt.” In other words, access to a vacation home of high quality by socialist standards for a couple weeks each year free of charge constituted a substantial perquisite for those with political connections, and it would have been widely envied by those without it.

Milanovic might have made the following more telling and accurate criticism of the idea that vacation homes increased the inequality of incomes. Not just the politically connected had access to such vacation homes. Rather, even those not so connected could get such access if they did what the authorities wanted them to do: refrained from criticizing communism, refused to support many of the victims of communism, showed up at work, and so forth. In other words, access to vacation homes, like so many other perquisites under communism, was a means of creating loyalty to the regime and cementing workers into the system.

Now to my criticism some of Milanovic’s recent claims. Don Boudreaux has handled it well but I want to add my own thought.

Here’s Milanovic’s statement:

But has real income of the American middle class gone down? Professor Boudreaux thinks it has not. To prove that, he engages into a doubly absurd exercise. He compares income of today’s middle-class Americans with income of the middle-class Americans 40 years ago using the goods that were inexistent 40 years ago. Indeed by that peculiar metric they are better off. It would suffice that one American out of perhaps 100 million middle-class American owns a smart phone today to obtain Professor Boudreaux’s result: since nobody had it 40 years ago, the growth rate of such income would be infinite.
If we equally one-sidedly, but somewhat less absurdly, make a comparison in terms of individual goods or services that existed then and now, like health care, education, and housing, we would find the very opposite result. This is why, if we want to engage into such comparisons, we use Consumer Price Index which includes all goods and services. When we do so and compare real incomes at different percentiles of U.S. income distribution, we get the result shown below. Over the thirty-year period U.S. middle class income has cumulatively increased by between 20 and 30 percent, which on an annual basis gives a rate of growth between 0.6% and 0.9%. This is hardly satisfactory and is especially galling when we compare middle-class growth with that of the top 5 percent, or even better with the top 1-percenters, whose incomes have increased by more than 80% (cumulatively).

Notice the first two sentences:

But has real income of the American middle class gone down? Professor Boudreaux thinks it has not.

That’s correct. That’s what Boudreaux does think. But here’s  something else interesting: Milanovic agrees with him. In the second paragraph, Milanovic writes:

Over the thirty-year period U.S. middle class income has cumulatively increased by between 20 and 30 percent, which on an annual basis gives a rate of growth between 0.6% and 0.9%.

In short, the real income of the middle class has risen, not fallen.

What about Milanovic’s point that looking today at only smart phones gives a biased viewpoint? He’s right. And contrary to what I had thought when I first read Milanovic’s statement too quickly, he admits that having access to a smart phone now does, all other things equal, increase one’s real income, possibly by a lot. But Milanovic’s point is that all else is not equal. In particular, health care, education, and housing have all become more expensive. Of course, in each case, other than education (hmmm: which entity provides most of that?) quality has clearly gone up. So then we have to dig into details.

That’s why Milanovic wants to use the CPI, because it includes a wide spectrum of goods.

Don Boudreaux catches the problem, citing the Boskin Commission report of 1996. The CPI overstates inflation.

But here’s something more. As the Report points out, and as Boskin has laid out in “Consumer Price Indexes,” his entry in The Concise Encyclopedia of Economics, there’s a particular problem with new items. Coincidentally, Boskin mentions the cell phone. He writes:

Finally, an additional bias results from the difficulty of adjusting fully for quality change and the introduction of new products. In the U.S. CPI, for example, VCRs, microwave ovens, and personal computers were included a decade or more after they had penetrated the market, by which time their prices had already fallen 80 percent or more. Cellular telephones were not included in the U.S. CPI until 1998.

In short, the introduction of new products does contribute to the CPI’s overstating of inflation. And it contributes substantially. Boskin writes:

The CPI currently overstates inflation by 0.8–0.9 percentage points: 0.3–0.4 points are attributable to failing to account for substitution among goods; 0.1 for failing to account for substitution among retail outlets; and 0.4 for failing to account for new products. Thus, the first 0.8 or 0.9 percentage points of measured CPI inflation is not really inflation at all. This may seem small, but the bias, if left uncorrected for, say, twenty years, would cause the change in the cost of living to be overstated by 22 percent.

Therefore, middle-class income has increased even more than Milanovic admits.

UPDATE:

Russ Roberts covers the issue of price indices overstating inflation in a very entertaining way here. The video is only 6 minutes long.

His thought experiment reminds me of a question I asked my students when we covered the economics of health care. We had looked at all the distortions due to subsidies, the tax code, restrictions on provider supply, etc., all of which made the cost of health care higher. But I wanted to put it in perspective. I gave examples of procedures docs and hospitals can do now that they didn’t do in 1965. I then asked: “Which would you prefer: 1965 health care at 1965 prices or 2015 health care at 2015 prices?” Over 80 percent of the class preferred the latter.