The numbers seemed outdated almost as soon as they appeared. Last September, the Census Bureau released its poverty and income reports for 2000, showing that the poverty rate declined to 11.3 percent, matching the all-time low, and median household income held steady at the all-time high of about $42,000.
Normally, these findings would be hailed as signs that the economy was strong and that more and more citizens were benefiting from the expansion of the 1990s. But last year, coming on the heels of the Sept. 11 tragedy, even economists at the bureau expressed doubts. â€œThe day we put out the numbers, some people were already saying, â€˜Poverty levels are probably a half percentage point higher.â€™ And that was on the day it was released,â€? recalls Chuck Nelson, assistant division chief for income, poverty and health statistics at the bureau.
As if to confirm this impression, just weeks later, the Commerce Department reported that for the first time in eight years, the nation's economy shrank in the third quarter, marking the end of the longest expansion in American history. A few days after that, the Labor Department announced that more than 400,000 Americans were laid off in October, the most for a single month in 21 years.
Data from the Current Population Survey (CPS), which is collected in March for the previous year, is, by definition, always out-of-date by the time it is released the following September. But rarely has the economic climate changed so radically from one year to the next, as it did in 2001. So the question naturally arises: What do you do with the Census Bureau's findings, contained in two reports, â€œMoney Income in the United States: 2000â€? and â€œPoverty in the United States: 2000?â€? What relevance, if any, do they have today and how can economists, demographers and business executives best use these numbers?
Interviews with a half dozen economists and demographers who track such data make it clear that the CPS numbers are best used to plot longer-term trends and track information over time. â€œBy their nature, reports are always a look back,â€? says Nelson. â€œThrough this data, you build up a time series. You get a picture of the effect of the economy on income and poverty. It's the basic information of what's happening to the American economy.â€?
And some of that information does not change that quickly. Certain underlying strengths built up during the years of economic growth will not disappear overnight even as we enter a recession â€” a recession that many economists believe will be short-lived and shallow. By the end of 2002, analysts say, the economy may be back at or near 2000 levels. It would therefore be foolhardy to disregard the CPS figures entirely, economists say.
At the same time, however, for those interested in strategizing for the next few months, â€œyou've almost got to turn to other indicators,â€? says Steven Cochrane, a senior economist at Economy.com, a West Chester, Pa.-based consulting firm. â€œI wouldn't use [the CPS] numbers right now.â€? Analysts may need to look to more current monthly or quarterly reports to get a clearer picture of today's economy.
Keeping things in perspective seems to be the order of the day. When asked how to use the CPS numbers, Ken Hodges, chief demographer at Claritas, a San Diego-based market research firm, replies, â€œCarefully and with judgment.
â€œThere are widespread notions that everything has changed since Sept. 11, so we make the assumption that anything reported before is not relevant,â€? says Hodges. â€œThat is overdone. It's more proclamation than [the result of] any real data. Yes, unemployment is up, but it's still low historically.â€?
Besides, according to the Census Bureau's Nelson, â€œshort-term changes in income are not very accurate as a measure of well-being.â€? That's why the CPS figures, which take the measure of a whole year, give a more comprehensive understanding of how people are functioning in the economy.
An increase in unemployment, for example, doesn't automatically translate into an increase among those living in poverty. More people in a family are working today, and the makeup of families is so different these days that the loss of one person's job may not have as dramatic an effect on household life as it did in the past, he adds.
Regarding the use of the CPS figures, Cochrane believes, â€œwhat you really have to do is look beyond this year. This year is so extraordinary. The near-term outlook is being driven by exogenous events that we don't have answers for, like terrorism and war. You have to look beyond, to late 2002, early 2003, when I expect the economy to begin expanding again.â€?
At that point, certain trends reflected in the CPS data will bear up once more. â€œThe essentials are all still there for recovery,â€? asserts Martin Holdrich, senior economist at Woods & Poole Economics, a Washington, D.C.-based market research firm. The â€œunderlying reasonâ€? for household income going up and poverty rates going down in the 1990s was expanding global growth combined with a tight labor market, which in turn drove wages up, says Holdrich. And â€œthe underlying causes [for tight labor markets] haven't changed. It's because of demographics,â€? he explains â€” the smaller Gen X cohort being unable to replace the aging Baby Boomers in the work force. â€œOnce the economy starts growing again, this [demographic reality] will return,â€? says Holdrich, as will demand from global markets.
Even beyond these fundamentals, some of the numbers in the CPS reports â€œare representative of longer-term trends,â€? that do not fluctuate much from year to year, says Daniel Weinberg, chief of the Census Bureau's housing and household economic statistics division. Income inequality, for example, which was on a downward trend after World War II to the 1960s, and began climbing through the '80s, has remained pretty flat over the past decade, as can be seen by the 2000 Income report. This more equitable sharing of the wealth in the '90s helped lead to record lows for poverty levels among blacks and Hispanics, as demonstrated in the 2000 Poverty report. â€œThat's not going to change,â€? says Weinberg. Also, certain subgroups of the population, like the elderly, will not be as hurt by a rise in unemployment.
To get a snapshot of what's going on today, especially given the shock of Sept. 11, researchers and marketers can turn to reports such as the monthly unemployment numbers, the Bureau of Economic Analysis' monthly series on personal income, welfare rolls, bankruptcy filings and quarterly updates on economic output. But for comprehensive data on income and poverty rates on a national level, economists and demographers agree there's just no real substitute for the CPS figures, which have always had to be evaluated on a time-delayed basis.
Within the next few months, Hodges will prepare his projections for 2002. â€œFor those of us who wrestle with the lag on a regular basis, having the 2000 numbers is still better than [only having] '99,â€? he says. â€œWe'll be very glad to have these numbers to use as a point of comparison for next year and the year after. Hang onto [the 2000 CPS] because it'll be useful to compare it with future releases.â€?
Forecast is American Demographics ' monthly newsletter devoted to census coverage. For a subscription, call (212) 204-3877 or send an e-mail to [email protected].