Chart 4-3 Real Household Income Relative to 1967 Income for Selected Quintiles Real income of low-, middle-, and high-income households generally rose from 1967 to 1979, fell from 1979 to 1982, and rose after 1982. 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 Note: CPI-U-X1 used as deflator. Sources: Department of Commerce and Department of Labor. terns: wages for high-wage workers were roughly the same level in 1979 and in 1982, while wages for low-wage workers fell. Chart 4-3 understates the improvement in income for the lowest group because, among other reasons, money income omits noncash transfers. Real Federal and State spending on means-tested medical assistance, the vast majority of which is medicaid, grew by $67 billion (in 1990 dollars) from 1967 to 1990, while spending on other means-tested noncash transfers grew by $46 billion. Real payments for medicare, which is not means-tested, grew by $96 billion. In 1990, households in the lowest income quintile received about 10 percent of medicare payments, 17 percent of medicaid payments, and 59 percent of other means-tested noncash transfers. Maintaining these allocations over time and using the Census Bureau's best estimates of the value of these transfers to recipients provides estimates of noncash transfers per household. For households in the lowest quintile, money income plus the estimated value of noncash transfers, adjusted for inflation, increased by 48 percent between 1967 and 1990, nearly double the 25-percent growth rate for money income alone. Gini Ratios The Gini ratio is a measure of the dispersion of income that ranges between 0 and 1. A lower value indicates less dispersion in the income distribution; a Gini of 0 would occur if every family had the exact same amount of income. A higher value indicates more dispersion; a Gini of 1 would occur if all income accrued to only one family. Chart 4-4 shows that from 1947 to 1968, despite some fluctuations, the dispersion of money income for families fell gradually. Since then, dispersion has risen slowly but steadily, by about 14 percent. Almost one-third of the increase occurred between 1979 and 1982, when wage and income patterns diverged sharply at the high and low ends of the spectrum. Slightly more than a third of the increase occurred between 1968 and 1979, with the increase from 1982 to 1990 accounting for the remaining third. Chart 4-4 Gini Ratios for Family Income The dispersion of money income followed a downward trend from 1947 to 1968 and increased gradually after that. The Gini for households followed similar trends. One study found that shifting household composition accounted for almost half of the increase in dispersion between 1969 and 1989. The Gini ratio is a measure of relative income rather than of the absolute level of income. Thus, changes in the Gini do not provide any information about the level of income for various groups in the population. In the 1980s, increasing dispersion of income did not mean that the rich became richer while the poor became poorer. Incomes grew in all quintiles, but income in the top quintile grew fastest. Trends in the share of income received by families in each quintile mirror those of the Gini coefficient. The share received by the lowest quintile rose from 5.0 percent in 1947 to 5.7 percent in 1968, and then fell to 4.6 percent in 1990. The share for the highest quintile fell from 43.0 percent in 1947 to 40.5 percent in 1968, before rising to 44.3 percent by 1990. Similar trends apply to income shares received by households. Like the Gini, however, measures of income shares do not show how the level of income has evolved for each group, and thus give an incomplete picture of income patterns. Similar trends in income distribution appear in other countries as well. One study found that in the early 1980s, the distribution of earnings for prime-age males who headed households and worked full time became more unequal in all five countries studied: Canada, Sweden, Australia, West Germany, and the United States. The widening distribution in many countries indicates that the causes of the shift are more likely to be due to factors common to all of the countries rather than to any factor specific to only one of the countries. THE DISTRIBUTION OF LONG-TERM INCOME AND Families and households display a substantial amount of mobility across income classes in the United States. For this reason, analyses of income distribution that focus only on annual income tend to overstate the degree of income inequality. One reason annual income data are misleading is that earnings of individual workers tend to rise as they acquire training and experience and then to fall when they retire. A 20-year-old worker just starting out and a 45-year-old worker who is in his or her peak earning years could have equal incomes over their careers, but very different wages in the same calendar year. Data on annual income can also prove misleading because of transitory income, that is, income gains or losses that are thought to be temporary. A person who owns a small business, for example, may face greater year-to-year fluctuations in income than someone who works at a steady wage. There is substantial mobility across income classes from year to year. One study found that in the mid-1980s, about one-third of all families were in a different income quintile than they had been in the previous year. In each of the lowest three quintiles, about 18 percent of the families moved to a higher quintile the following year. In each of the highest three quintiles, more than 20 percent of the families moved to lower quintiles the following year. Another study found that more than half of families in the highest quintile in 1971 had fallen into lower quintiles by 1978. Similarly, almost half of those in the lowest quintile had risen to a higher quintile. Over longer periods, the extent of mobility increases. One study, using data from the 1970s and 1980s, found that more than 75 percent of households are in a different decile when ranked by lifetime income than when ranked by current income. A decile includes onetenth of the households. About 44 percent had current income two or more deciles away from their lifetime income. More than half of households in each of the lowest three deciles for annual income had lifetime income in a higher decile. More than half of households in the top three deciles for annual income had lifetime income in a lower decile. A recent study, using tax return data from the 1960s, 1970s, and 1980s, estimates that the Gini coefficients for income over 4-year or 7-year periods are between 5.0 percent and 7.7 percent less than the average of the Gini coefficients for the individual years. Another study, using data from 1969 to 1981, found that the Gini for lifetime income in the United States was 19 percent lower than the Gini for annual income, indicating less dispersion in lifetime income. These findings underscore the importance of income mobility for a large number of families. Nevertheless, even after removing temporary income changes and the effects of the life-cycle on income, part of the population still faces very low long-term income prospects. Because the distribution of long-term income. is less dispersed than are annual incomes, trends in the distribution of annual income may not accurately reflect trends in the distribution of long-term income. For example, an increase in income mobility or in the importance of transitory income can increase inequality of annual income but have no effect on the distribution of long-term income. Nevertheless, one study found that, like annual incomes, incomes averaged over 4- and 7-year periods became more dispersed between 1967-73 and 1979-85. A related issue is the distribution of wealth. A family's wealth holdings consist of financial assets, such as saving accounts; property, such as a house or family business; pensions and future Social Security benefits; and human capital, the value of future labor earnings. For most households, housing, public and private pensions, and human capital constitute the vast bulk of wealth. One study found that between 1983 and 1989 the median value of households' real financial net worth and property rose 11 percent and that holdings of these assets became more concentrated. 305-592 0-92--5 SUMMARY • Median levels of family and household money income have shown sustained long-term growth since the mid-1960s. Median income is influenced by cyclical and long-term economic activity and demographic patterns. • Since the mid-1960s and in particular since the early 1980s, income growth has occurred in all quintiles and the distribution of annual money income has become more dispersed in the United States. Earnings distributions have also become more dispersed in several other countries in recent years. • Because money income omits in-kind transfers, data on money income understate both the level of and improvement in income for the lowest income groups. • Families and households display significant mobility across income classes. The distribution of long-term income is more equal than the distribution of annual income. TRENDS IN TAXES AND TRANSFERS Tax and transfer policies in the United States have undergone major changes in level and composition in the last 30 years. These changes are among the principal ways that government influences the distribution of resources and the level and structure of economic activity. TRANSFERS The two main categories of transfers are means-tested programs and social insurance. Means-tested programs provide benefits or services to people and families whose financial resources have fallen below a certain level. Distributed by Federal, State, or local governments, means-tested transfers can be cash grants, such as aid to families with dependent children, or goods and services, such as food and health care. Transfers of goods and services ensure that assistance is used for the purposes intended. Means-tested service programs also provide education and job training. (Brief descriptions are in Boxes 4-1 and 4-2; fiscal 1990 spending totals are in Table 4-1.) Social insurance programs compensate people for income loss due to retirement, disability, and unemployment, and provide health insurance for the elderly. The three major Federal social insurance programs are Social Security, medicare, and unemployment insurance (Box 4-3). These programs are financed primarily by payroll taxes. Because they are predominantly not means-tested, social insurance programs can sometimes make large direct payments to the well-off (Box 4-4). |