Obligatory Nonsense on Inequality at the NYT
Every year or so, the NYT feels obligated to print a piece of nonsense masquerading as economics from W. Michael Cox, the senior Vice-President of the Dallas Federal Reserve Bank. The first item in this series that I recall was a piece that argued that there was great mobility in the United States because those in the bottom quintile at any point in time were likely to move up to higher quintiles, including even the top quintile, in future years.
This looked very impressive until you found our that Cox used all adults in his sample, not just prime age people, as serious economists would do. This means that the law students and medical students, who are likely to be low income this year, are the basis for much of Cox's upward mobility story, since they will have relatively high incomes when they are lawyers and doctors.
This year's nonsense concerns consumption. Cox tells us that there is much less inequality in consumption than income, so therefore we should not really be concerned about inequality. I won't go through all the problems in Cox's analysis (there are many). I will just point out that the data set that he uses, the consumer expenditure survey (CEX) is not very well-suited for this sort of analysis.
The CEX misses a great deal of consumption. This can readily be seen by simply looking at the aggregate statistics. The average after-tax income reported in the survey is $58,101. Average consumption expenditures are $48,398. This implies a savings rate of 16.7 percent. The National Income and Product Accounts data show a savings rate of less than 1 percent. This suggests that the CEX is missing a great deal of consumer expenditures, which makes this sort of analysis very dubious.
You may wonder why the NYT would print columns from someone with such a consistent reputation for getting things wrong. I guess that is the price that we pay for having a regular column from Paul Krugman. Too bad they can't find a conservative who could at least make an honest argument.
--Dean Baker
Feeds: 


COMMENTS (16)
Because of wage inequality, working people go into more and more debt. This consumption analysis treats debt binges like living off your wages. Only a conservative would confuse these. Not only that, but our debt bubble is popping, and foreclosures and bankruptcies are up. So much for treating debt like income.
Posted by: dissent | February 11, 2008 1:43 AM
I think you write well and your very logical presentation significantly devalued the proposition that Mr Cox presents. But why refer to his work as nonsense and why use such a derisive tone. Make your point in a civil fashion. It is a matter of simple courtesy which is too often forgotten on the internet.
Posted by: John Jansen | February 11, 2008 4:50 AM
Greg Mankiw must have liked the Cox oped as he posted it sine comment. Mark Thoma posted it too but followed this with a stinging rebuttal by Paul Krugman - in the NY Times!
Posted by: pgl | February 11, 2008 5:21 AM
Dean's tone is remarkably constrained, considering the pernicious ideology he is debunking.
Posted by: Aunt Deb | February 11, 2008 6:50 AM
To reinforce the point you made look at the numbers for top 20 percent. According to Cox and Alm they make $149,963. The spend $69,863 and pay $23,376 in taxes. This implies savings of $56,724. This would imply that people in the top 20 percent save nearly 38 percent of their income. That is hard to believe.
In addition, if there is such a significant difference in savings this will allow people with higher incomes to retire earlier. Since life expectancy is greater for people with higher incomes they will spend more time in retirement and consume more over their life time (especially if you count consumption per year worked).
Finally, since Cox and Alm suggest that income doesn't really matter I guess the rich won't mind paying more in taxes so the rest of us can have national health insurance, more money for public education and more spending to improve our infrastructure and protect the environment.
Posted by: Rudy Fichtenbaum | February 11, 2008 12:17 PM
I agree with Aunt Deb that Dean is showing restraint in his debunking. Mr. Jansen's plea for civility would be understandable if the "mistakes" that Cox made were really honest. To have both the Times and Mr. Cox overlook such glaring misrepresentations is simply incredible. This piece shows an effort to promote a falsehood that perpetuates the further concentration of wealth and income. Rudy was spot on. If the rich are only spending such a small portion of their income, they won't mind paying more income tax. Anyone for an 80% tax bracket?
Posted by: Jim Hannley RIA | February 11, 2008 12:43 PM
NIPA data can also be misleading. Recall that you are looking at an income measure that excludes capital gains, includes income accruing in the pensions, but excludes pension benefits. Low saving rate measures reflect this.
To be sure, one-year consumption patterns are also misleading, demonstrqting the importance of looking at more homogeneous demographic groups and, perhaps, considering longer periods of time.
Posted by: Mike McCracken | February 11, 2008 3:27 PM
That is also one poorly designed graph, too.
Posted by: mcdruid | February 11, 2008 4:04 PM
One intersting aspect, commented on at Krugman's blog, is what the rich, not so rich, and poor spend income on. I have seen some changes over my past 198 years.
A few laughed that we shouldn't rejoice over cheap DVD players. But the issue is deeper than that.
Hunger has greatly decreased in the U.S., (a major improvement over Britain when I was at the helm all four times). So while being poor is still a hardship, the worst element has been removed for most.
Imagine if using the internet cost $300/month, so that only the well off would have easy access where half the population would be at libraries or internet cafes. This was almost the situation when PCs spread in the early 1980s and connectiveness meant sharing a floppy disk.
Or consider health care where for fairly common but deadly serious problems like heart disease and cancer. Bill Gates isn't much better off than the average worker.
And health care will become more equal over time as new drugs and therepies significantly decrease risk for everyone.
It is election time, so improvements which will soon help the poor and middle class are the last thing Krugman and Baker will point out. Nor do they have a science background to see this. But the public is about to find out where this is heading very soon.
(And Rudy is right.)
Posted by: gladstone | February 11, 2008 7:49 PM
gladstone wrote, Nor do they have a science background to see this.
LOL!
This coming from the guy who can't understand the reason why improvements in 5-year cancer survival rates are deceptive without appropriate controls?
Posted by: liberal | February 12, 2008 1:57 AM
I saw the argument. That isn't the only reason cancer death rates are dropping 2% a year. See, liberal, at times multiple causes are in play.
Neither Baker nor Krugman have taken science beyond high school beyond a required class.
Posted by: gladstone | February 12, 2008 8:46 AM
Are 38% of American families living below a more accurately drawn poverty line?
50 percentile (technically, mean third-quintile) family income for 2005 was $56,277.
A plausible poverty line for a family of three (on the "minimum needs" table on p.44 of the 2001 book Raise the Floor) is $33,345 in 2008 dollars -- if health care is otherwise covered. Add $10,000 for a 2008 family health policy and we get a plausible poverty line (as opposed to the implausible federal line computed at 3X the price of a super cheap diet) for a family of three of $43,345.
(Raise' supplies extensive explanations for its minimum needs numbers in Appendix B -- its budget tables cite Solutions for Progress as their source.)
The difference between second and third quintile mean family income ($35,000, $56,000) is about $1,000 per percentile. That computes ($35,000 is 30 percentile + $8,000 more is 8 more percentile) to 38% of American families living below a believable poverty line (at least without food stamps and other helps -- average family size is 3.13 persons).
I have been touting 25% as a realistic poverty line -- just doubled the 12.5% official line -- made a conservative fit with Raise the Floor's doubling of official criteria (was more than doubling -- $17,000 being the official criteria for a family of 3) -- conservative because, if anything, I expected the low income curve to be flatter than 45 degrees. It seems it was flatter than I even guessed.
Assuming I can't find something wrong with 38% in poverty, I will email this startling stat to media all over the country. The media continuing to report 12.5% poverty without qualification is like reporting in Columbus' time that the world is flat -- it makes no waves; but informed people know otherwise.
Posted by: Denis Drew | February 12, 2008 9:59 AM
Just an extra thought on 38% of American families living below a believable minimum needs (a.k.a.) line: I hope it give some perspective the $500/wk being our 40 percentile wage (State of Working America).
It should wake us up to the fact the the 40 percentile wage in America in 2008 should be a middle class wage. Who could live a middle class live on $500/wk before taxes. Most of it is gone for rent (no utilities) for one person.
It doesn't seem to wake anybody up that $400/wk was the MINIMUM WAGE in 1968 -- when AVERAGE INCOME WAS HALF WHAT IT IS TODAY. Hope the 38% stat can finally make the 40 percentile stat look like what it really represents: a working America that is basically being taken totally to the cleaners (mostly by their own ignorance and complacency in the labor market I hasten to add -- but maybe the "38%" will wake up the wonks to wake up the workers).
Posted by: Denis Drew | February 12, 2008 11:10 PM
Denis Drew wrote, Most of it is gone for rent (no utilities) for one person.
Ah, but that's the important point: it's not that wages are too low, it's that landowners get to collect rent for doing nothing.
Were land rent to be paid to the government, acting on behalf of the people, instead of parasitic landowners, the lot of the poor and working class would be much improved.
As it is, government shovels about 10--20% of GDP into the pockets of landowners in return for nothing.
Posted by: liberal | February 13, 2008 4:26 PM
Perhaps the reason that they publish such nonsense back to back with Krugman's column is so that intelligent readers can note the difference, and agree with the Times' own columnist.
But I guess this is wishful thinking.
Posted by: njf | February 14, 2008 4:45 PM
Commenting much later, but was this the piece where the guy argued that po' folks are just fine, because if their needs outstrip their wages, they can easily sell their car, home or "other assets" to maintain their consumption?
Truly hilarious. I agree that "nonsense" is a pretty civil characterization.
Posted by: Anonymous | April 10, 2008 1:22 AM