Throwing facts aside, a recent New York Times editorial, “Consumption Gap,” nonetheless began by saying, “conservative economists often argue that wage stagnation and income inequality are not as big a threat to Americans’ standard of living as they’ve been made out to be. In their view, how much one buys — rather than how much one makes — is a better measure of economic well- being.”
No economist who hopes to avoid professional ridicule would try to deny that consumption is a better measure of long-term living standards than the most widely cited income distribution figures, which do not even add transfer payments or subtract taxes.
A 2005 study by Dirk Krueger and Fabrizio Perri concluded that any increase in income inequality from 1980 to 2003 “has not been accompanied by a corresponding rise in consumption inequality.” In my book “Income and Wealth,” I show that inequality of consumption is about 40 percent lower than inequality of income (before taxes and transfers) and that consumption inequality declined slightly between 1986 and 1999–2001.
The New York Times editorial takes sides in a little spat between American Enterprise Institute economists Aparna Mathur and Kevin Hassett on the right, and Jason Furman of the Center on Budget and Policy Priorities and Jared Bernstein of the Economic Policy Institute on the left.
Calculating the annual growth of personal consumption spending per capita for the middle fifth of households, Mathur and Hassett wrote, “The average annual consumption growth for the middle class was less than 1 percent in the period from 1990 to 1994, rose to 1.5 percent in the period from 1995 to 1999, and jumped to more than 2 percent in the period from 2000 to 2005.”
The editorial explains that Furman and Bernstein “reworked the figures [by] including newly available spending data for 2005.” But changing one year just affected the average for 2000 to 2005, leaving the slower growth of the Nineties unchanged. The updated figure for 2000–2005 is now 1.8 percent, not 2.1 percent. So what?
To get the 1.8 percent figure down, Furman and Bernstein leave out the gains of 2000 and 2001, because they claim to be most interested in a cyclical question — comparing the past four years with comparable early periods of previous recoveries.
The New York Times thus concludes “there is no question that spending by the middle class has been weaker in the current economic expansion than in previous recoveries.” The editorial also claims, “there is no dispute among the various researchers over the new findings.” Perhaps not, but there should some dispute about the dates.
The statistic “beyond question” rose by 1.3 percent a year during the first four years of this recovery, 2002 to 2005, and by 1.3 percent a year during President Clinton’s first term, 1993 to 1996. See the difference?
To make 1.3 percent look worse for Mr. Bush than it did for Mr. Clinton requires boosting Mr. Clinton’s record with a lot of help from Ronald Reagan. Bernstein and Furman compare the 2002–2005 period with annual consumption growth from 1985 to 2000. From 1985 to 1988, real consumption rose by 4.1 percent a year, which makes the entire 1985–2000 period look better than the first four years under either Bush or Clinton.
The editorial changes the subject from growth to distribution: “Consumer spending by low-income households is way down since 2001. Over the same period, spending by high-income Americans has been robust.” The phrase “since 2001” carefully excludes 2001 for another too-clever reason.
In the bipartisan years of 2000 and 2001, growth of consumption by the top fifth was only 0.3 percent a year. The stock market, after all, collapsed. Yet growth of consumption by the bottom fifth grew by 7 percent in 2000, and by another 3.1 percent in 2001. They consumed much more than their apparent money income, which is always the case.
Because energy costs can be a particularly large part of low-income budgets, 2001 was not an entirely bad year. It was tough if you lost your job, but a majority of the bottom fifth is not employed. In most cases, cheaper energy freed up their cash to spend on other things. The price of West Texas crude reached $34 in November 2000, before recession yanked it down to $19 in December 2001.
For the same reasons, it seems entirely likely that low-income consumers were disproportionately injured by soaring energy prices after 2001, particularly in 2005, and had to cut back on spending on gasoline and other things. The oil price reached $32 in December 2002, then $43 a year later, then $59 at the end of 2005, which is about where it is now (after topping $74 in July).
It is not surprising that consumption of low-income seniors, for example, was weak during a period when energy prices spiked so much. It would be partisan nonsense, however, to blame the world price of crude oil on whoever happens to be sitting in the White House. Any policy that helped get U.S. industry back on its feet after Sept. 11, 2001, was going to contribute to the escalating global demand for industrial materials, including energy.
If we add figures for 2000 and 2001 to those of 2002–2005, consumption still grew a percentage point faster in the top group (2.2 percent) than the bottom (1.2 percent) and slightly faster than the middle (1.8 percent). But six years is much too short a period to detect a new trend, particularly during a period when the price of oil nearly quadrupled.
When editorial writers proclaim, “there is no dispute” and “no question” about politically charged statistics, you can be sure they are blowing enough snow to create a White Christmas.
