[NYT] AUG 31, 2001 Boom of 1990's Missed Many in Middle Class, Data Suggests By JANNY SCOTT
The booming late 1990's appear to have left the middle class in the New York region and California no better off than it was a decade before, an analysis of Census Bureau data suggests. The poor got a little poorer, the rich got a lot richer and the large group in the middle emerged slightly worse off than when the decade began.
The analysis, conducted for The New York Times, compared income data from the 1990 census with data from an experimental Census Bureau survey done in 2000. It found that the median family income clearly declined in New York, California, Connecticut and Washington, D.C. Andrew A. Beveridge, a professor of sociology at Queens College who conducted the analysis, said he also found that the gap between rich and poor throughout the country had inched wider during the 1990's. In Washington, D.C., for example, the average income of families in the wealthiest fifth of the population, once adjusted for inflation, grew to 24 times the average in the bottom fifth, up from 18 times.
"There is a worsening of income inequality," Professor Beveridge said. "And in New York State, the middle class has not kept up. The poor have treaded water, more or less. The well-off have made gains. So you have this squeeze in the middle - people like cops, firemen, people making under $80,000 a year."
Professor Beveridge's finding does not mean that particular individuals took home less money in 2000 than they had a decade earlier. It suggests a decline in the average income of families in the middle - the ever-changing group comprising the 60 percent of all people who at any given time fall in the center of the income spectrum.
Some economists were skeptical that the decline was real.
"I would challenge anybody to find a middle-class family in this region whose economic condition has declined," said Stephen Kagann, chief economist to Gov. George E. Pataki. He added, "Nobody's real income goes down during periods of prosperity - no group of people. Everybody rises, but they rise at different rates."
Mr. Kagann pointed out that many middle-class people left New York during the recession of the early 1990's, and many immigrants moved in, some earning less than the people who left. "That by itself would bring the median down," Mr. Kagann said. "But that does not mean that a middle-class person is less well-off than they were in 1990. That would simply be untrue."
Meanwhile, the Census Bureau yesterday released updated estimates of the poverty rate nationally and state by state, based on several sets of federal data from 1998. When compared with 1990 census numbers, the new estimates show that the percentage of people in poverty dropped slightly nationally but crept upward in New York, New Jersey, California, Connecticut and other states.
What is expected to be a more definitive picture of state-by-state trends in income will be available next year, when income statistics from the 2000 census are scheduled to be released. In the meantime, the results of the experimental Census Bureau survey, which Professor Beveridge analyzed, are thought by many to offer a preliminary glimpse at state-level trends in the 1990's.
The bureau has warned that some of the survey data are not perfectly comparable to data from the 1990 census because the bureau asked slightly different questions of the two different groups of people surveyed. However, Professor Beveridge said the Census Bureau numbers released yesterday, and another set of annual bureau surveys, suggest similar trends.
Bruce Smith, principal economist for the California Department of Finance, expressed skepticism similar to Mr. Kagann's. The population studied by the Census Bureau changed between 1990 and 2000, he said. While the incomes of middle- class people who were in California in 1990 went up, he insisted, a lot of new arrivals probably make less that those who were there before.
Others found the findings less surprising.
Edward N. Wolff, a professor of economics at New York University who recently concluded from his own research that most families in the United States saw their net worth stagnate over the past decade, said that while median family income had risen in many places since the mid- 1990's, "a lot of this was simply covering losses that occurred in the early 1990's."
The numbers Professor Beveridge analyzed came from the Census 2000 Supplementary Survey, a survey of 700,000 households nationwide that is being used to test the feasibility of collecting census data in a new way. He compared those data with income figures from the 1990 census, adjusted for inflation using the index used by the Census Bureau.
Comparing those two sets of numbers, he found that median family income in New York State declined to an estimated $52,313, down $2,876; in California, to $53,037, off $3,288; in Connecticut, to $64,502, down $3,821; and in Washington, D.C., to $45,943, down $4,406. Median family income measures the income of blood relatives living together.
Median household income declined in many states, Professor Beveridge found. But some economists suggested that median family income is a more telling measure of the position of people in the middle, in part because the nationwide rise in one- person households may account for much of the drop in household income, the income of anyone, related or not, in a home.
A similar pattern is visible in the median household income numbers from another Census Bureau survey, the annual Current Population Survey, a trusted source of income and poverty data at the national level. Median household income for the tristate area around New York rose throughout the 1980's until 1990, then dropped sharply before turning around in 1996.
Looking more closely at family income, Professor Beveridge found that the biggest gains occurred among the most affluent. In New York, the average family income in the richest fifth of the population rose to $173,418, or 13.6 times the average income of the poorest fifth. Nationally, the average for the top group rose to 10.9 times that of the bottom group, up from 10.4 in 1990.
Professor Beveridge found that average income rose nationally, in part because the average was pulled upward by large increases in income among the rich. His analysis looked only at income, not at wealth or overall net worth.
Census Bureau officials said they could not say how much of any apparent decline in median family incomes might be attributable to differences in design of the the 1990 census and the 2000 supplementary survey. Ed Welniak, chief of the income surveys branch, said a test done by the bureau found differences in the way people answer different sets of income questions.
When asked their income in the previous 12-month period, people tended to give a lower figure than when asked their income in the previous calendar year, he said. In the 1990 and 2000 censuses, people were asked about the previous calendar year; in the supplementary survey and the Current Population Survey, they were asked their income in the previous 12 months.
"What will help us in understanding what might be going on is when we produce the 2000 census data," Mr. Welniak said. "We will have two sets of data, both using the previous calendar year. We will have a much better comparison to compare the 1990 data with. It will negate that methodologic difference."
Jared Bernstein, an economist with the Economic Policy Institute, a liberal research organization in Washington, who had reservations about comparing the survey data with the 1990 census, said that even so, there were real trends in certain sectors of the economy in some states that could shed light on any apparent income decline in a place like New York.
"Particularly New York City was very hard hit by the recession in 1990-91," he said. "For New York, it was a much deeper recession than for other areas of the country. It didn't really come on line with the economic boom until a year or two after the rest of the country."
In addition, he said: "The financial markets and high-rolling bond traders at the very top of the economic scale did very well over the latter 90's when incomes were growing quickly. But at the same time, New York generated reams of low-end service jobs - jobs in retail, cashiers, the tourism industry, hotels and the low end of health care. If you add a bunch of low-wage workers at the beginning of the income scale, that lowers the median."