The changing shape of the nation's income distribution

Abstract

Are the rich getting richer and the poor getting poorer?Historical Census Bureau income statistics can shed some light onthis debate. Although the Census Bureau has been measuring incomefor a half-century, and a large number of factors have been identi-fied as contributing to changes in inequality, the causes are stillnot entirely understood.The Current Population Survey (CPS) is a rich source of dataon income inequality.During the past 50 years, the annual demographic supplement to theMarch CPS has provided researchers with a wealth of data on the incomedistribution. Since 1947, the Census Bureau has employed a commonlyused measure, the Gini coefficient (also known as the index of incomeconcentration), to measure family income inequality. With two exceptions,the Gini coefficient decreased between 1947 and 1968. During thisperiod, the Gini for families indicated a decrease in income inequalityof 7.5 (±2.1) percent. Since 1968, however, this trend has reversed.Income inequality for families, measured by the Gini coefficient,increased between 1968 and 1998 (see Figure 1). The net effect overthe entire 1947-1998 period is an increase in family income inequality.A New Mode of Data CollectionIn 1993, the Census Bureau began using a new method of collectingincome data, allowing respondents to report greater income valuesin the Current Population Survey. A change that may affect only asmall number of cases (particularly those at the upper end of theincome distribution) can have a considerable effect on inequalitymeasures, like the Gini coefficient and shares of aggregate income,while making little or no change to median income. This had a profoundeffect on the upper end of the income distribution by recording incomelevels that had been previously underreported. The impact of thischange on measured income inequality was quite large, and we areunable to determine precisely the proportion of the increase in incomeinequality between 1992 and 1993 that is attributable to this change.