Preferred Citation: Paul Krugman, "The Rich, the Right, and the Facts," The American Prospect no. 11 (Fall 1992): 19-31 (http://epn.org/prospect/11/11krug.html). The Hubbard Study In June the Treasury's Office of Tax Analysis, under the direction of Glen Hubbard, an economist on leave from Columbia, released a report claiming that there is actually huge upward mobility in the U.S. In particular, it claimed that 86 percent of individuals who started in the bottom quintile in 1979 had moved out by 1988, and indeed that an individual who started in the bottom quintile was more likely to end up in the top quintile than to stay where he was. But this report was based on what we may charitably call a strange procedure. Here's what Hubbard's report did: it tracked a group of individuals who paid income taxes in all ten years from 1979 to 1988, and compared their incomes not with each other but with those of the population at large. COMMENT: Krugman is incorrect about this. The Hubbard study tracked a group of individuals who FILED income tax forms in both 1979 and 1988. They are included in the study even if they did not file in some of the years in between, and even if they did not PAY taxes, but filed to get back money that was withheld. See the letter from Glen Hubbard at the end of this file-----jeb BACK TO Krugman "Rich Right & Facts" The restriction to individuals who paid taxes in all years immediately introduced a strong bias toward including only the economically successful; only about half of families paid income taxes in all ten years. This bias toward the successful was apparent in the fact that by the end of the sample period the group contained very few poor people and a lot of affluent ones: indeed, only 7 percent of the sample were in the bottom quintile by the sample's end, while 28 percent were in the top quintile. More important, by comparing the sample with the population at large rather than with each other, the report essentially treated the normal tendency of earnings to rise with age as representing social mobility. The median age of those whom the study classified as being in the bottom quintile in 1979 was only twenty-two. Kevin Murphy, a labor economist at the University of Chicago, neatly summed up what the Treasury study had found: "This isn't your classic income mobility. This is the guy who works in the college bookstore and has a real job by his early thirties." --------------------------------------------------------------------- Thank you, Ronnie N. Carpio rncarpio@kudonet.com Hi, I got your letter and read the analysis by Krugman. I thought that you must have mis-quoted him. But later, I was able to connect to the URL and, no he did say what you said he said. It contains many strange ways of looking at data, but those seem to have come from Krugman and not you. RR&F Now for the Krugman essay "The Rich, the Right and the Facts" First, I don't doubt that most people in the USA have become richer (in terms of money) during the past 10 or 20 or 30 years. Or any 10 year interval since WW II. And those who are VERY rich are even richer than the rest of us. Even more so than in the past several decades. Probably more so than any time since before WW I. I can see this in anecdotal evidence with professional athletes, movie and rock stars and, and corporate CEO's. But the gap may not be as much as was the case during the last century with the great "captains of industry". He seems to think that instead of MEASURING a reality, his figures ARE the reality. For example he claims in the "Krugman Calculation" that since the US growth rate was fixed at a certain percent, the "rich" "siphoned off" 70% of the gain, the implication being that this gain would have been distributed to the others had the "rich" not taken it. The number of children per family is, say 2.2. Suppose we found that many couples have no children, while a few families had 4 or 5 kids. Would he then claim that the large families had "siphoned off" a large number of kids that might otherwise have gone to the couples with none? Or would the average just have been different if the large families had fewer kids? If Bill Gates or Henry Ford had never existed, all the money they "siphoned off" would have been distributed to the rest of us? Or is it as likely that without them, our productivity would be lower now? So much for Krugman, up to the section on INCOME MOBILITY. There he repeats his claim that the movement of individual people upwards in income level "unfortunately...[does not] actually characterize the US economy." As evidence, he cites the Census data (given also in PEDDLING PROSPERITY--see my book review) that between 1985 and 1986 "only" 18.4% of those people in the bottom income quintile moved up, and "only" 23.7% of those in the top quintile dropped down. He seems to see this a proof of, or at least evidence that, there is little income mobility. But a 20% turnover at both the top and the bottom (rounding off a little) would be consistent with a complete turnover in about 5 years! That would correspond to an incredibly high level of income mobility. (Not that I think this actually happens but his own data would not rule it out.) He cites some study by Greg Duncan of Michigan U about transitions into and out of some undefined "middle class", but I would need a lot more information to understand what this is even about. And another study showing that rich people got rich because they had a high income. Rather like proving that only the "rich" won the lottery last year. (I surveyed the winners, and they are all rich (NOW!). The real core of this centers on the Hubbard Study. It is summarized in the INCOME MOBILITY TABLE on my web page. The HUBBARD SUBJECTS: Are They Typical?? Unfortunately, Krugman does not cite a reference to the complete study: it was released in June of ?? by the US Treasury's Office of Tax Analysis, and was done by Glen Hubbard, then on leave from Columbia University. I guess the year to be 1992, since it was printed in the Wall Street Journal on June 2, 1992. The study tracked a relatively large number of individual taxpayers from 1979 to 1988. For why individual people and not "group statistics" are needed, see the section on how "Mexicans Get Younger" in the USE OF STATISTICS file on my web page. The first objection raised by Krugman is that since someone had to be a taxpayer during all of there 10 years, the sample is not representative of the general population. He claims that only "only about half of all families" fit this criteria. He is wrong about this. See the letter from Hubbard below. Also is there a shift here from "individual" to "family"? Which did the study follow? What about individuals who were single in 1979 but married sometime before 1988? And how does that effect the degree to which the sample is representative? We will need the details of the study to answer these questions. Well, yes. If the result is to be extrapolated to the US, the sample should be "typical". But no study can be perfect. I have heard census data criticized because "people lie on census forms", and the IRS is more likely to get the truth. But to track people, they must be there during the time of the study, or least at the start and finish. So what has kept the others out? And are the excluded likely to be richer or poorer, and have a greater or lesser change in their income than those included? The first things that come to mind are these: people who died during the time interval, or were just kids in school (not earning enough income to file and dependents of their parents) when the study started. And people whose income was too low to pay income tax during the study. While this last group would exclude the mostly the "upwardly mobile" (unless their income dropped down below the tax level and then shot way up--poor lottery winners for example), those who died are likely to be older and richer than average. And kids leaving school early in the study are likely to have a bigger jump in income than the average--more on that later. So it is not obvious to me which way this exclusion of "being there" will bias the result. COMPARED TO WHAT? When the study shows big gains in income for most of the people, K says this is just the "normal tendency of earnings to rise with age". (What, does he want them to make more money and NOT get older?? How can you do ANYTHING and not get older?) But in the 6th paragraph of the next section on INCOME GAINS, he says "most poor ... stay that way". Which implies to me that their income does not go up as a normal tendency of aging. But those in the Hubbard Study DID have increases in income. Not just in dollars, but RELATIVE to the entire population. (It would be instructive to see how the income of the subjects changed, not just relative to the general population but in actual dollars. Is that data included in the study??) Krugman objects that the subjects are compared to the US population in general, and not just to each other. But the POINT of the study is to see how individuals do in the US. Do they move up or down in income? I mean if you selected a random group and compared them only to each other, you would "discover" that on average, they did about average. For example, lets say you selected 10 poor new immigrants at random, and 10 years later they had all become millionaires: had $1 million in assets. I would say that this means that poor immigrants have a very good chance of becoming millionaires in a decade. But by Krugmans criteria, they would all just be "average". I suppose he would say this just reflects the normal tendency of poor immigrants to become millionaires as they get older. AN EXAMPLE Let me give an example of how the "rich get richer" can be misinterpreted because of income mobility. Professional Athletes have had one of the biggest gains in income during the past 30 years of anyone anywhere. (And while I have no objection to people getting rich in general, see the Baseball Players Syndrome, and Brewers files on my web page.) This sounds like a classic case or the "rich" getting "richer". Until you look at the people. When Bart Star was the Green Bay Packer quarterback, he was paid $15,000 a year. Corrected by the CPI to dollars today, about $70,000 (and I think CPI an estimate a best, but that is another story: see my "Inflation" & "CPI" files). That was probably enough to put him in the top quintile of income then, but not by a lot. Now Brett Favre makes I don't even know how many MILLIONS a year, and he is well toward the top of the top quintile. The "rich" became "super-rich"?? Of course not. A poor kid from Alabama became mildly rich then, and another poor kid from rural Mississippi has become VERY rich today. Krugman may see that as the "rich getting richer". To me THAT is INCOME MOBILITY. And finally notice that had a "Hubbard Study" been run to finish 1n 1994 Brett Favre could probably not be included: while a multi- millionaire when it ENDED, he would have been a 15 year old high school student when the 10 year interval started. ,,,,,,, _______________ooo___(_O O_)___ooo_______________ (_) jim blair (jeblair@facstaff.wisc.edu) Madison Wisconsin USA. This message was brought to you using biodegradable binary bits, and 100% recycled bandwidth. For a good time call: http://www.geocities.com/capitolhill/4834 ADDED NOTE: To clear up some of the questions about the study, I e-mailed this to Glen Hubbard at Columbia: Hi, I have a summary of your famous income mobility study on my web page at: http://www.geocities.com/capitolhill/4834/101.htm In discussions about it some questions have come up. Since it compares income status for only 2 years, were individuals excluded from the study if they failed to file a tax form ANY of the years in between 1979 and 1988? Or did they need have data for only those 2 years? Also, did they need to PAY taxes in those 2 years (or in all of the years in between?), or was it enough to simply file? (I recall that some people filed for rebates and to collect EITC). > September 18, 2000 > >The panel was not balanced, and you need to file in the year given to be >included. > >----Glenn Hubbard [rgh1@columbia.edu] A panel data set is one where you can study cross-sectional variation over time. A balanced panel is one where the elements of a given cross-section are there in every time period (no one drops out of the sample). His data set evidently has some people who dropped out for some years.