(Archived document, may contain errors) 10/18/89 253 POVERTY AND PLENTY IN AMERICA: UNDERSTANDING CENSUS BUREAU DATA The Census Bureau today released its official statistics on poverty and family income for. 1988. The figures s how continued strong improvement. For the sixth straight year, real per capita income is up and the number of Americans below the "poverty threshold" is down. Inflation-adjusted per capita income increased to $13,120 last year, an historic high and a 1.7 p ercent improvement over 1987. Real per capita income is up dramatically since 1982 - 19.2 percent. The real median income for black married couple families rose over the past year by a startling 6.8 percent, to $30,390. The poverty rate was down to 13.1 p e rcent, from 13.5 percent last year. Yet even this decline actually overstates the number of poor Americans. Census Bureau methodology continues to exaggerate the number of persons in poverty. A more accurate measure would find millions fewer poor American s than the 32.5 million reported today. The economic well-being of American families is understated because. the. Census Bureau's official data fail to count as income major government assistance programs for the poor and to measure the rate of inflation a c curately. The statistics ignore both assets (about 40 percent of "poor" households, for instance, own their own home) and the impact of such non-cash "in-kind" government programs as Medicare, Medicaid, housing subsidies, and food assistance. T11is; non-c a sh aid totals $174 billion. The Census Bureau ignores it as though it has no effect on poverty. Admittedly, the Bureau has published "experimental" income definitions that include some non-cash benefits. Such technical appendices, however, rarely are noti c ed by the press. It is thus the unadjusted figures that become the "official" poverty rate and are used as the basis for public policy debate and legislation. Erroneous Conclusion. Even the Census Bureau's incomplete inclusion of some programs, however, r e duces the official poverty rate significantly. In 1988, it brings the rate down from-13' A percent to 12.4 percent or 10.5 percent (depending on the method used to place a value on non-cash benefits). Policy makers should recognize that until the Bureau i n cludes the total benefits of non-cash programs as income in its official income and poverty data, poverty will be exaggerated in the U.S. and many non-cash programs will be deemed, erroneously, to have no impact on poverty. The Census Bureau's statistics o n both median family income and the poverty rate are distorted further by a mistake in the measurement of inflation. The original poverty in come thresholds have been adjusted upward since 1963 to keep pace with inflation. However,, it has been long recog n ized that the government's normal measurement of inflation has overstated the rate for the past 25 years. Experts agree that this inaccuracy, due to such things as overstating the importance of housing in certain family budgets, has led to a serious overe stimate in the official count of the number of poor persons. This year the Bureau has experimented with an improved and widely-accepted measure of inflation, which reduces the current poverty income threshold from
$12,092 in 1988 to $10,997. This experim ental measure is not used to adjust the "official" data. If it were, it would indicate that in 1988 there were 3.6 million fewer poor persons in the U.S. than suggested in the Bureau's own official report. Despite the deficiencies in the overall official n umbers in the report, the Census Bureau's yearly data continue to indicate that family instability and a lack of work effort account for most of the persistent poverty continuing in the midst of the economic boom of the - 1980s. In 1988, the poverty rate a mong intact families was 5.6 percent; by contrast, the poverty rate: for female-headed families was 33.5 percent. Single motherhood is a recipe for poverty, and one which represents an increasing share of America's poverty population. In 1959, just 28 per c ent of poor families with children were headed by women. But in 1988, some 61.3 percent of poo'r families with children were female-headed. The growth in the number of female-headed families and their persistently high poverty rate should demonstrate to p o licy makers that family breakup - not alleged government inaction, "low" wages, or uneven economic growth - is the most important cause of poverty. TackIing the Root Cause. It also appears that government "anti-poverty" programs actually have undermined f a mily self-sufficiency, giving rise to a new social class of government dependents. In the 1950s, before the "War on Poverty," nearly one-third of poor families were headed by adults who worked full-time throughout the year. In 1988, only 16.4 percent of p o or families had full-time working heads of households. Over half of poor families are headed by persons who do not work at all. The proper reaction to these figures by lawmakers should be to tackle family breakup and disincentives to work. Adding new bene f it programs likely would only exacerbate the root cause of poverty and welfare dependency. The 1988 Census data, though still deficient, thus show steady and significant improvement for most Americans. If the methodology used by the Bureau were improved, the figures would show the condition of American families to be even better.
Kate Walsh O'Beirne Visiting Fellow
Robert Rector Policy AnalystF or further information: Kate Walsh O'Beirne, "U.S. Income Data: Good Numbers Hiding Excellent News," Heritage Fo undation Backgrounder No. 667, August 19, 1988. Robert Rector and Kate Walsh O.'Beirne, "Dispelling the Myth of Income Inequality," Heritage Foundation Backgrounder No. 710, June 6,1989.