The Trend in Family Income from 1979 to 2010

. The Trend in Family Income from 1979 to 2010 Gerald Mayer Analyst in Labor Policy February 8, 2012 Congressional Research Service 7-5700 www.crs.gov R42348 CRS Report for Congress Prepared for Members and Committees of Congress c11173008 . The Trend in Family Income from 1979 to 2010 Summary Policymakers often debate how government programs and legislation benefit or harm families at different income levels. Families can be defined in different ways. The analysis here uses two categories of families. The first category combines married couples, single parents, and individuals into one group. The second category consists only of families headed by married couples. Family income can also be defined differently. The analysis here is based on a consistent data series from the Annual and Social Economic (ASEC) supplement to the Current Population Survey (CPS). The analysis uses two definitions of income: income from all sources, and earnings from work only. Income does not include employer contributions for health insurance or other benefits, in-kind transfers (e.g., for food, housing, healthcare, or energy assistance), or capital gains. In recent decades, family income inequality has increased. In recent years, the level of family income has changed because of slow or declining real hourly earnings, a decline in annual hours worked, and the 2007-2009 recession. It may be difficult to separate the effects of longer-term changes in income inequality and hours worked from the effects of the recent recession. From 1979 to 2010, average real income (i.e., actual income adjusted for inflation to constant 2010 dollars) increased for all families and for all married couples. However, the average income of families at the lowest quintile (i.e., the lowest fifth of families by income) fell by $3,350 (22.3%), while the average income of the top 5% of families increased by at least $55,280 (25.5%). The average income of married couples at the lowest quintile fell by $1,550 (5.9%), while the average income of the top 5% of married couples increased by at least $85,590 (34.4%). Because of differences in the growth of real family income by quintile, income inequality increased from 1979 to 2010. Most of the increase took place from 1979 to 1989. Inequality also increased during the recession years of 2007-2009. Real hourly earnings were higher in 2010 than in 1979. But total hours worked by all families have been falling since about 1989 and for married couples since about 2000. For families at all quintiles, average total earnings fell during the economic expansion between 2000 to 2007 and again during and after the 2007-2009 recession. In 2010, the average earnings of families at the three lowest quintiles were lower than in 1979. The loss of earnings from 2000 to 2010 was greater than the growth in earnings in previous years. The lower earnings were due to slow or declining wage growth and a reduction in hours worked. The average total earnings of married couples at the lowest quintile were lower in 2010 than in 1979. The average earnings of couples at the next-to-lowest and middle quintiles were higher in 2010 than in 1979, but the gains in earnings from the expansions of the 1980s and 1990s were partially offset by a loss of earnings from 2000 to 2010. The rise in earnings from 1979 to 2010 for couples at the next-to-lowest and middle quintiles was due, in part, to higher real hourly earnings. But, most of the growth in earnings was due to an increase in hours worked. The average earnings of married couples at the top two quintiles were also higher in 2010 than in 1979. But, unlike couples at the next-to-lowest and middle quintiles, who benefited from an increase in hours worked, most of the increase in earnings for couples at the top two quintiles was due to higher earnings per hour worked. The total number of hours worked by these couples (over 4,000 annually) suggests that they are two-earner couples. Congressional Research Service . The Trend in Family Income from 1979 to 2010 Contents Family Income: Why it Matters and Policies to Improve It ............................................................ 1 Definitions Used in the Report ........................................................................................................ 2 The Trend in Family Income ........................................................................................................... 4 Longer-Term Trends .................................................................................................................. 5 The Distribution of Family Income..................................................................................... 5 Annual Hours Worked......................................................................................................... 6 The Level of Family Income ..................................................................................................... 7 All Families......................................................................................................................... 7 Married Couples.................................................................................................................. 9 Family Earnings Per Hour and Total Hours Worked ............................................................... 11 Earnings Per Hour Worked................................................................................................ 11 Hours Worked.................................................................................................................... 13 The Separate Effects of Hourly Earnings and Hours Worked on Total Family Earnings................................................................................................................................ 16 All Families....................................................................................................................... 16 Married Couples................................................................................................................ 18 Other Reasons For Changes in Family Income ............................................................................. 18 Conclusion ..................................................................................................................................... 20 Figures Figure 1. Annual Unemployment Rates, 1979 to 2010.................................................................... 4 Figure 2. The Distribution of Family Income, as Measured by the Gini Coefficient, 1979 to 2010 .......................................................................................................................................... 5 Figure 3. Average Annual Hours Worked, All Families and Married Couples, 1979 to 2010 ................................................................................................................................. 7 Figure 4. Average Real Income, All Families and Married Couples, 1979 to 2010 ........................ 8 Figure 5. Changes in Average Real Family Income, by Quintile, All Families, 1979 to 2010 ........................................................................................................... 8 Figure 6. Changes in Average Real Family Income, by Quintile, Married Couples, 1979 to 2010 .................................................................................................. 10 Figure 7. Changes in Average Real Hourly Earnings, by Quintile, All Families, 1979 to 2010 ............................................................................................................................................ 11 Figure 8. Changes in Average Real Hourly Earnings, by Quintile, Married Couples, 1979 to 2010 .................................................................................................. 13 Figure 9. Changes in Total Hours Worked Annually, by Quintile, All Families, 1979 to 2010 ......................................................................................................... 14 Figure 10. Changes in Total Hours Worked Annually, by Quintile, Married Couples, 1979 to 2010 .................................................................................................. 15 Figure A-1. Average Real Family Income and Average Real Income of the Top 5% of Families, 1979 to 2010.............................................................................................. 23 Congressional Research Service . The Trend in Family Income from 1979 to 2010 Figure A-2. Family Earnings as a Share of Family Income, by Quintile, 1979, 2007, and 2010 ............................................................................................. 24 Tables Table 1. Family Type and Average Family Size, 1979 and 2010..................................................... 6 Table 2. Estimates of the Relative Effects of Changes in Real Hourly Earnings and Total Hours Worked on Family Earnings, by Quintile, 1979 to 2010 ................................................. 17 Table 3. Changes in the Age and Education of the Labor Force, 1979 to 2010............................. 19 Table A-1. Average Hours Worked by All Family Members, by Quintile, 1979 to 2010 .............. 25 Table A-2. Illustration of the Decomposition of Average Annual Earnings Due to Changes in Real Hourly Earnings and Annual Hours Worked.................................................................. 26 Table A-3. Illustration of Equivalence Scale ................................................................................. 27 Table A-4. Average Real Income: All Families and Married Couples, 1979 to 2010.................... 28 Table A-5. Gini Coefficient: All Families and Married Couples, 1979 to 2010 ............................ 31 Table A-6. Average Real Family Earnings Per Family Hour Worked: All Families and Married Couples, 1979-2010 ...................................................................................................... 32 Appendixes Appendix. Data and Methodology................................................................................................. 21 Contacts Author Contact Information........................................................................................................... 35 Congressional Research Service . The Trend in Family Income from 1979 to 2010 P olicymakers often debate how government programs and legislation benefit or harm families at different income levels. For example, some argue that the middle class has been squeezed by declining or stagnant incomes and rising costs for food, energy, healthcare, and other goods and services. Others maintain that, because of improvements in productivity and advances in technology, real incomes and the standard of living have improved for most, if not all, families. Also, because of economic mobility, a middle income family may become an upper income or lower income family. This report analyzes the trends in the level and distribution of family income from 1979 to 2010. The beginning year of 1979 was chosen because the U.S. economy was at the peak of an economic expansion. Other peaks occurred in 1989, 2000, and 2007. The most recent data for family income are for 2010. A recession officially began in the United States in December 2007 and ended in June 2009.1 In addition to a loss of jobs, a recession typically affects both hourly earnings and hours worked. Therefore, this report examines the total number of hours families work outside the home and how much they earn per hour worked. The report separates the relative effects on family earnings of changes in hourly earnings and hours worked. Finally, the report examines changes in family income, hourly earnings, and hours worked over three economic expansions before the recent recession: 1979-1989, 1989-2000, and 2000-2007. The report also examines the period during and after the 2007-2009 recession. Earnings inequality has increased in recent decades and, in recent years, the total number of hours worked by families outside of the home has fallen.2 It may be difficult to separate the effects of these longer-term trends from the shorter-term effects of the business cycle. Family Income: Why it Matters and Policies to Improve It Family income can be affected by economic, social, and demographic changes. Economic expansions and recessions can affect employment and hours worked. Changes in the labor force participation of men and women can affect family earnings. A younger or older population can affect the age of the workforce and the sources of family income. For most families, earnings from work are the largest part of family income. Real earnings may grow with increased productivity and advances in technology. Policies to increase productivity may include programs to raise both private and public saving, increase capital investment per worker, expand investment in human capital (e.g., education, training, and healthcare), and encourage the development of new technology.3 1 The National Bureau of Economic Research (NBER) dates the beginning and end of recessions. National Bureau of Economic Research, U.S. Business Cycle Expansions and Contractions, http://www.nber.org/cycles/. (Hereinafter cited as NBER, U.S. Business Cycle Expansions and Contractions.) 2 See CRS Report RL33835, Real Earnings, Health Insurance and Pension Coverage, and the Distribution of Earnings, 1979-2009, by Gerald Mayer. 3 Technological changes may include improved equipment; the introduction of new goods and services; or improved methods of production, transportation, or communication. Congressional Research Service 1 . The Trend in Family Income from 1979 to 2010 Real family income may also rise with improved economic efficiency, which consists of a more efficient allocation of resources.4 Economic efficiency can be improved through policies that provide consumers with greater access to goods and services; for example, policies that expand the exchange of goods or reduce the costs of exchanging goods (e.g., improved infrastructure). Economic efficiency can also be improved through a better allocation of labor and capital; for example, policies that remove barriers to capital and labor mobility or reduce the effect of the tax system on decisions to work, save, and invest. Inequality may be reduced by using either direct or indirect policies. Direct policies include income transfer programs. Indirect policies consist of programs that improve the incomeproducing human capital of lower-skilled workers (e.g., education, training, or healthcare) or that reduce the relative supply of less-skilled labor, increase the relative supply of skilled labor, or do both. The supply of less-skilled workers can be reduced with policies that improve investment in preschool, grade school, and high school education; better adult education; and improved access to healthcare for lower income workers and their families. The supply of skilled workers can be increased with policies that lower the cost of higher education or increase educational assistance to lower income students. Immigration policies that allow more skilled workers, fewer unskilled workers, or both, into the country can also reduce inequality. Some policies may be more cost effective than others, however. Direct policies to reduce income inequality include programs such as progressive taxation— including refundable tax credits like the Earned Income Tax Credit (EITC). Direct policies also include in-kind transfers of food, housing, healthcare, and energy assistance. Programs to reduce inequality may involve tradeoffs, however, with policies to improve economic efficiency. For example, progressive taxation may reduce inequality in the distribution of aftertax income, but high marginal income tax rates may affect decisions to work, save, and invest. Transfer payments may affect the supply of labor (i.e., decisions to work and the number of hours worked). Finally, macroeconomic policies may affect the level and distribution of family income. Fiscal and monetary policies can help lower unemployment or maintain low unemployment. Low unemployment can increase the average number of hours that individuals work and increase the hourly earnings and total income of lower-wage workers and their families. Definitions Used in the Report This section describes the concepts and data used in the report. More detail is provided in the Appendix. Families can be defined in different ways. This report analyzes the incomes of two categories of families. The first category, called “all families,” includes households of all types. It is comprised of families headed by married couples, single parents, and individuals living alone or sharing a household with another, or other, unrelated individuals. The second category consists only of families headed by married couples. The analysis is restricted to families headed by persons 4 Resources consist of individuals with different skills, capital goods (e.g., computers, machinery, and buildings), and natural resources. Congressional Research Service 2 . The Trend in Family Income from 1979 to 2010 between the ages of 22 and 62, the age group that is most likely to be affected by changes in labor market conditions. The report uses data from the Annual Social and Economic (ASEC) supplement to the monthly Current Population Survey (CPS). The CPS is a household survey conducted by the U.S. Census Bureau for the Bureau of Labor Statistics (BLS). The monthly CPS is the source of the national unemployment rate and other labor market information. In the ASEC supplement, family income is defined as the total money income of all family members. Money income includes earnings from work and income from nonlabor sources (e.g., interest, dividends, unemployment compensation, or public assistance). Earnings include wages and salaries, self-employment income, tips, commissions, and cash bonuses. Earnings do not include employer contributions for health insurance or to a retirement plan. Money income is before taxes or other deductions. Money income does not include in-kind transfers (e.g., for food, housing, healthcare, or energy assistance) or investment income such as interest, dividends, or capital gains.5 In this report, family income is reported in constant 2010 dollars. To analyze the trend in family income, the report uses a consistent method to calculate the average income of higher income individuals. In the CPS, to protect the confidentiality of survey participants, the actual incomes of higher income persons are not reported. Instead, the CPS assigns these individuals an income amount; their incomes are topcoded. Year-to-year changes in these amounts can affect the observed trend in average income for families with the highest incomes. This report uses data from independent research to develop a consistent data series for family income for the years 1979 to 2010. Nevertheless, for individuals with the highest incomes, the data series does not include actual income. Instead, the incomes of these individuals are averages based on their gender, race, ethnicity, and whether or not they work full-time, yearround. These averages can affect the analysis of the distribution of family income. (See the discussion of topcoding in the Appendix.) Family income may change because of longer-term trends and because of shorter-term changes over the course of the business cycle. From 1979 to 2010, there were five official recessions in the United States.6 The recent 18-month recession was the longest since the Great Depression of the 1930s. The unemployment rate reached 10.1% in October 2009.7 In 2010, the unemployment rate was 9.6%. From 2007 to 2010, employment fell by an estimated 7.8 million jobs (from 137.6 million to 129.8 million).8 The deepest recession since 1979 was the 16-month recession of July 1981 to November 1982. The unemployment rate reached 10.8% in both November and 5 For an analysis of changes in the distribution of income from 1996 to 2006 among tax filers, see CRS Report R42131, Changes in the Distribution of Income Among Tax Filers Between 1996 and 2006: The Role of Labor Income, Capital Income, and Tax Policy, by Thomas L. Hungerford. 6 The five recessions occurred from January 1980 to July 1980, July 1981 to November 1982, July 1990 to March 1991, March 2001 to November 2001, and December 2007 to June 2009. NBER, U.S. Business Cycle Expansions and Contractions. 7 The unemployment rate for October 2009 is seasonally adjusted. U.S. Department of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, available at http://stats.bls.gov/cps/. (Hereinafter cited as BLS, Labor Force Statistics from the Current Population Survey.) 8 Because the most recent data for family income are for 2010, this report analyzes the trend in family income from 1979 to 2010. Nevertheless, employment data for 2011 show that from 2010 to 2011, employment increased by 1.3 million (from 129.8 million to almost 131.2 million). Also, unemployment data for 2011 show that from 2010 to 2011, the unemployment rate fell from 9.6% to 8.9%. U.S. Department of Labor, Bureau of Labor Statistics, Employment, Hours, and Earnings from the Current Employment Statistics Survey, available at http://www.bls.gov/ces; BLS, Labor Force Statistics from the Current Population Survey. Congressional Research Service 3 . The Trend in Family Income from 1979 to 2010 December 1982.9 The 1981-1982 recession was preceded by a six-month recession in 1980, which was the shortest recession since the Great Depression. As context, Figure 1 shows the national unemployment rate for the years 1979 to 2010. This report analyzes changes in the level and distribution of family income over three periods that, for the most part, extend from the peak of one economic expansion to the next. The three periods are 1979 to 1989, 1989 to 2000, and 2000 to 2007.10 The six-month recession of 1980 is included in the period from 1979 to 1989. Figure 1 also shows the peak years between the three economic expansions of the 1980s, 1990s, and 2000s. Finally, the report examines changes in family income during and after the 2007-2009 recession. Figure 1. Annual Unemployment Rates, 1979 to 2010 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% Peak of economic expansion 20 10 20 05 20 00 19 95 19 90 19 85 19 79 0.0% Unemployment rate Source: U.S. Department of Labor, Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey. Notes: Because the most recent data for family income are for 2010, this report analyzes family income from 1979 to 2010. However, from 2010 to 2011, the unemployment rate fell from 9.6% to 8.9%. The Trend in Family Income This section summarizes the findings of the report. The income and earnings amounts discussed are rounded. The Appendix shows unrounded amounts. 9 The unemployment rates for November and December 1982 are seasonally adjusted. BLS, Labor Force Statistics from the Current Population Survey. 10 The unemployment rate was 5.8% in 1979, 5.3% in 1989, 4.0% in 2000, and 4.6% in 2007; BLS, Labor Force Statistics from the Current Population Survey. Congressional Research Service 4 . The Trend in Family Income from 1979 to 2010 Longer-Term Trends Family income may change because of longer-term demographic, social, and economic trends. This report takes into account two longer-term trends: changes in the distribution of income and changes in the total number of hours worked by families outside the home. The Distribution of Family Income Figure 2 shows changes in the distribution of family income as measured by the Gini coefficient. The Gini coefficient is a measure of equality that ranges from 0 to 1. If the incomes of all families are the same, the Gini coefficient is equal to 0, representing complete equality. If one family receives all income and all other families receive no income, the Gini coefficient is equal to 1. Thus, a larger coefficient indicates a greater degree of inequality. Income inequality increased from 1979 to 2010, as measured by changes in the Gini coefficient. Most of the increase occurred from 1979 to 1989. Inequality is greater among all family types analyzed together than among married couples analyzed separately. From 1979 to 2010, inequality increased more among all families than among families headed by married couples. Inequality also increased during the recession years of 2007-2009. Figure 2. The Distribution of Family Income, as Measured by the Gini Coefficient, 1979 to 2010 0.45 0.40 0.35 Peak of economic expansion All families 20 10 20 05 20 00 19 95 19 90 19 85 19 79 0.30 Married couples Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: Estimates are for families headed by persons ages 22 to 62. Changes in the Gini coefficient between 1992 to 1993 are not shown because a change in the CPS affected the observed incomes of high-income families. See the discussion of topcoding in the Appendix. Congressional Research Service 5 . The Trend in Family Income from 1979 to 2010 Changes in family size and composition can affect the trend in the distribution of family income. From 1979 to 2010, the percentage of families that were married couples fell from 62.2% to 46.1%, a decline of 16.1 percentage points. From 1979 to 2010, average family size fell from 3.6 to 3.3 persons for families headed by married couples and from 1.9 to 1.7 for other types of families. Other types of families include single parents and individuals who may be widowed, divorced, separated, or never married. (See Table 1.) In this report, family income is adjusted for changes in family size and the number of adults and children in the family. (See Table A-5 and the discussion of the Gini coefficient in the Appendix.) Table 1. Family Type and Average Family Size, 1979 and 2010 1979 2010 Change from 1979 to 2010 Family Type Married couple families 62.2% 46.1% -16.1 Other family types 37.8% 53.9% +16.1 Average Family Size Married couple families 3.6 3.3 -0.3 Other family types 1.9 1.7 -0.2 Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. Annual Hours Worked The average number of family hours worked generally falls during a recession. The average number of hours worked by all family types increased after the 1981-1982 recession and peaked in 1989. But the number of hours worked has fallen since 1989. For married couples, the average number of hours worked increased until 2000, but it fell during the years preceding the 20072009 recession. (See Figure 3.) Average hours worked can change for a number of reasons, including changes in household size and composition, the age of the workforce, the effects of nonlabor income, the demand for labor, changes in the labor force participation of men and women, and other reasons. Congressional Research Service 6 . The Trend in Family Income from 1979 to 2010 Figure 3. Average Annual Hours Worked, All Families and Married Couples, 1979 to 2010 4,000 3,500 3,000 2,500 Peak of economic expansion All families 20 10 20 05 20 00 19 95 19 90 19 85 19 79 2,000 Married couples Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. The Level of Family Income Between 1979 and 2010, the average real income of all families peaked in 2000 and was falling before the 2007-2009 recession. For married couples, average income peaked in 2006, but it had changed little from 2000 to 2006. (See Figure 4.) All Families Figure 5 shows the changes in average real family income over each of the three business cycles from 1979 to 2007 and from the beginning of the 2007-2009 recession to 2010. (Recall that the period from 1979 to 1989 includes both the six-month recession in 1980 and the 16-month recession in 1981-1982.) In 2010, average real family income was $6,200 (9.8%) higher than in 1979. Average income increased during the two economic expansions of the 1980s and 1990s, but it fell during the 2000-2007 expansion and fell again during and after the 2007-2009 recession. Congressional Research Service 7 . The Trend in Family Income from 1979 to 2010 Figure 4. Average Real Income, All Families and Married Couples, 1979 to 2010 (in 2010 dollars) $120,000 $100,000 $80,000 $60,000 Peak of economic expansion All families 20 10 20 05 20 00 19 95 19 90 19 85 19 79 $40,000 Married couples Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. Figure 5. Changes in Average Real Family Income, by Quintile, All Families, 1979 to 2010 (in 2010 dollars) $60,000 $50,000 $40,000 $30,000 $20,000 $10,000 $0 -$10,000 -$20,000 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 Average Lowest quintile Next-to-lowest quintile Middle quintile Next-to-highest quintile 81st to 95th percentiles Top 5% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: Estimates are for families headed by persons ages 22 to 62. Because of a change in the CPS, the changes in the average income of all families and the average income of the top 5% of families do not include the change in income from 1992 to 1993. See the discussion of topcoding in the Appendix. Congressional Research Service 8 . The Trend in Family Income from 1979 to 2010 From 1979 to 2010, average real family income increased more for upper income than lower income families. In 2010, the average income of families at the three lowest quintiles was lower than in 1979. For families at the two lowest quintiles, average income fell during the expansion of the 1980s. During the expansion of the 1990s, average income increased at all quintiles. By contrast, average income fell for families at all quintiles during the expansion of 2000-2007. During and after the 2007-2009 recession, average income fell again for families at all quintiles. By 2010, the average incomes of families at the lowest and next-to-lowest quintiles were $3,350 (22.3%) and $4,510 (13.1%) lower, respectively, than in 1979. At the middle quintile, average real family income in 2010 was $1,660 (3.2%) lower in 2010 than in 1979. Although the average income of families at the middle quintile increased during both expansions in the 1980s and 1990s, it fell during the 2000-2007 expansion and again during and after the 2007-2010 recession. For families at the top two quintiles, average real income was higher in 2010 than in 1979. For families at the next-to-highest quintile, average income increased by $7,720 (10.6%). At the 81st to 95th percentiles, average income increased by $26,140 (25.0%). For the top 5% of families, average income increased by $55,280 (25.5%). As was the case with families at the middle quintile, the average income of families at the top two quintiles increased during both the 1980s and 1990s, but it fell from 2000 to 2007 and again from 2007 to 2010. Although the average real income of the top 5% of families, as reported here, increased by $55,280 from 1979 to 2010, a change in the CPS affected the observed change in income from 1992 to 1993. Because of a change in the coding of the incomes of persons with the highest incomes, the increase in the average income of the top 5% of families from 1979 to 2010 may have been more or less than $55,280. However, during the expansion of the 1990s, the average income of the top 5% of families increased annually from 1993 to 2000. Thus, the average income of the top 5% of families likely increased from 1992 to 1993 as well. Therefore, the increase in average income from 1979 to 2010 for the top 5% of families was probably at least $55,280. (See the discussion of topcoding in the Appendix.) Married Couples The average real income of married couple families is higher than the average income of all family types combined. Moreover, from 1979 to 2010, the average income of married couples increased more, or fell less, than it did for all families. In 2010, the average real income of married couples was $18,610 (23.4%) higher than in 1979. Average income increased by $9,800 (13.4%) and $15,080 (17.7%), respectively, during the economic expansions of the 1980s and 1990s. Although average income rose during the expansion of 2000-2007, the increase was only $110 (0.1%). Average income fell by $6,370 (6.2%) during and after the 2007-2010 recession. (See Figure 6.) As was the case with all families, average real income increased more for upper income than lower income married couples. In 2010, the average income of couples at the lowest quintile was $1,550 (5.9%) lower than in 1979. The average income of couples at the lowest quintile fell during the 1980s, but increased during the 1990s. Their average income fell during the expansion of the 2000s and again from 2007 to 2010. Congressional Research Service 9 . The Trend in Family Income from 1979 to 2010 Figure 6. Changes in Average Real Family Income, by Quintile, Married Couples, 1979 to 2010 (in 2010 dollars) $100,000 $80,000 $60,000 $40,000 $20,000 $0 -$20,000 -$40,000 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 Average Lowest quintile Next-to-lowest quintile Middle quintile Next-to-highest quintile 81st to 95th percentiles Top 5% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: Estimates are for families headed by persons ages 22 to 62. Because of a change in the CPS, the changes in the average income of all married couples and the average income of the top 5% of married couples do not include the change in income from 1992 to 1993. See the discussion of topcoding in the Appendix. For married couples at the other four quintiles, average real income was higher in 2010 than in 1979. For couples at the next-to-lowest quintile, average income rose by $3,880 (8.0%) from 1979 to 2010. For couples at the middle quintile, income increased by $12,830 (19.7%), and for couples at the next-to-highest quintile, income increased by $25,250 (30.0%). For couples at the 81st to 95th percentiles, average income rose by $47,800 (40.5%). For couples at these income levels, average income increased during each of the three economic expansions from 1979 to 2007, but it fell during and after the 2007-2009 recession. From 1979 to 2010, the largest increase in average real income was for the top 5% of married couples. Their average income increased by $85,590 (34.4%). The average income of the top 5% of couples increased during both the 1980s and 1990s, but it fell during the expansion of the 2000s and again from 2007 to 2010. Again, because a change in the CPS affected the observed change in income from 1992 to 1993, the increase in average income from 1979 to 2010 for the top 5% of married couples was probably at least $85,590. (See the discussion of topcoding in the Appendix.) Congressional Research Service 10 . The Trend in Family Income from 1979 to 2010 Family Earnings Per Hour and Total Hours Worked Income inequality can change for a number of reasons. For most families, earnings (i.e., income from work) are the largest source of income. (See Figure A-2 in the Appendix.) Total family earnings may rise or fall because of a change in earnings per hour, a change in the total number of hours worked, or both. Earnings Per Hour Worked All Families In 2010, the average real hourly earnings of all families was $3.60 (17.7%) higher than in 1979. The largest increases occurred during the economic expansions of the 1980s and 1990s. Average earnings fell during the expansion of the 2000s and again from 2007 to 2010. (Figure 7.) Figure 7. Changes in Average Real Hourly Earnings, by Quintile, All Families, 1979 to 2010 (in 2010 dollars) $16 $14 $12 $10 $8 $6 $4 $2 $0 -$2 -$4 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 Average Lowest quintile Next-to-lowest quintile Middle quintile Next-to-highest quintile 81st to 95th percentiles Top 5% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: Estimates are for families headed by persons ages 22 to 62. Because of a change in the CPS, the changes in the average hourly earnings of all families and the average hourly earnings of the top 5% of families do not include the change in earnings from 1992 to 1993. See the discussion of topcoding in the Appendix. As was the case with average family income, average real hourly earnings increased more for upper income than lower income families. In 2010, the average hourly earnings of families at the two lowest quintiles were lower than in 1979. At the lowest and next-to-lowest quintiles, average Congressional Research Service 11 . The Trend in Family Income from 1979 to 2010 hourly earnings fell by $0.30 (3.9%) and $0.60 (4.3%) an hour. Average earnings for families at these two quintiles fell during the economic expansion of the 1980s, increased during the expansion of the 1990s, changed little from 2000 to 2007, and fell during and after the 2007-2009 recession. Unlike the change in average income, which increased during the 1980s, the average real hourly earnings of families at the middle quintile fell during the expansion of the 1980s. Their average earnings increased during the expansion of the 1990s, fell during the expansion of the 2000s, and fell again from 2007 to 2010. However, unlike families at the two lowest quintiles, the average earnings of families at the middle quintile were higher, by $0.30 (1.6%) an hour, in 2010 than in 1979. From 1979 to 2010, the largest increases in average real hourly earnings were for families at the two top quintiles. For families at the next-to-highest quintile, earnings increased by an average of $1.60 (8.3%) an hour. For families at the 81st to 95th percentiles, average hourly earnings increased by $6.60 (27.7%). For families at the next-to-highest quintile and at the 81st to 95th percentiles, earnings increased during each of the three economic expansions from 1979 to 2007, but they fell during and after the 2007-2009 recession. On the other hand, the average hourly earnings of the top 5% of families increased during both the 1980s and 1990s, but they fell from both 2000 to 2007 and 2007 to 2010. From 1979 to 2010, the top 5% of families experienced the largest increase, at least $13.70 (30.0%), in average hourly earnings. Married Couples Average real hourly earnings are greater for married couple families than for all family types combined. Moreover, from 1979 to 2010, average hourly earnings increased more for married couples than for all families. In 2010, the average real hourly earnings of married couples were $4.70 (21.8%) higher than in 1979. Earnings were higher at all quintiles, but increased more for upper income than lower income couples. Average earnings increased during each of the three economic expansions from 1979 to 2007. (See Figure 8.) During the expansion of the 1980s, average real income fell only for married couples at the bottom quintile. (See Figure 6.) But, during this same time period, average real hourly earnings fell for couples at the three lowest quintiles. Their earnings increased during the expansion of the 1990s and again during the expansion between 2000 and 2007. But, their average earnings fell during and after the 2007-2009 recession. From 1979 to 2010, married couples at the top two quintiles experienced the largest increases in average real hourly earnings. Unlike married couples at the three lowest quintiles, average hourly earnings for couples at the top two quintiles increased during the expansion of the 1980s. Their earnings also increased during the expansion of the 1990s. During the expansion of the 2000s, hourly earnings increased for couples at the next-to-highest quintile and for couples at the 81st to 95th percentiles, but they fell for the top 5% of couples. In 2010, the average earnings of couples at the next-to-highest quintile were $4.50 (21.4%) higher than in 1979. For couples at the 81st to 95th percentiles, average earnings were $10.80 (43.1%) higher in 2010 than 1979. The top 5% of couples experienced the largest increase, at least $17.10 (33.8%), in hourly earnings. Congressional Research Service 12 . The Trend in Family Income from 1979 to 2010 Figure 8. Changes in Average Real Hourly Earnings, by Quintile, Married Couples, 1979 to 2010 (in 2010 dollars) $18 $14 $10 $6 $2 -$2 -$6 -$10 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 Average Lowest quintile Next-to-lowest quintile Middle quintile Next-to-highest quintile 81st to 95th percentiles Top 5% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: Estimates are for families headed by persons ages 22 to 62. Because of a change in the CPS, the changes in the average hourly earnings of all married couples and the average hourly earnings of the top 5% of married couples do not include the change in earnings from 1992 to 1993. See the discussion of topcoding in the Appendix. Hours Worked In addition to earnings per hour, family earnings are affected by the total number of hours worked. For all family types combined, from 1979 to 2010, annual hours worked peaked in 1989. For married couple families, annual hours worked peaked in 2000. All Families In 2010, families worked an average of three weeks (120 hours) less than they worked in 1979. Average hours worked fell the most, by an average of 6½ weeks (260 hours), for families at the lowest and next-to-lowest quintiles. At the middle quintile, average hours worked fell by 4½ weeks (180 hours). For families at the 81st to 95th percentiles, average hours worked fell by an average of three weeks (120 hours). For the top 5% of families, average hours worked fell by over four weeks (170 hours). On the other hand, families at the next-to-highest quintile worked an average of one-half week (20 hours) more in 2010 than in 1979. (See Figure 9.) During the economic expansion of the 1980s, average hours worked by all families increased by six weeks (240 hours), peaking at 3,240 hours in 1989. Total hours worked increased for families at all quintiles, but the increases were greater for middle and upper income families. At the next- Congressional Research Service 13 . The Trend in Family Income from 1979 to 2010 to-highest quintile, average hours worked increased by almost nine weeks (350 hours). For families at the 81st to 95th percentiles, average hours worked increased by 7½ weeks (400 hours). At the middle quintile, average hours worked increased by over six weeks (250 hours). For the top 5% of families, average hours worked increased by over five weeks (210 hours). After the 1989 peak, average hours worked fell during the expansion of the 1990s and the expansion of the 2000s. During the expansion of the 1990s, except for families at the lowest quintile, average hours worked fell at all quintiles. During the expansion of the 2000s, average hours worked fell at all quintiles except for the top 5% of families. Average hours worked fell at all quintiles during and after the 2007-2009 recession. Figure 9. Changes in Total Hours Worked Annually, by Quintile, All Families, 1979 to 2010 400 300 200 100 0 -100 -200 -300 -400 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 Average Lowest quintile Next-to-lowest quintile Middle quintile Next-to-highest quintile 81st to 95th percentiles Top 5% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. Married Couples Married couple families work more hours annually than all family types combined. In addition, unlike all families, who worked fewer hours on average in 2010 than in 1979, married couples worked an average of 2½ weeks (100 hours) more in 2010 than in 1979. Nevertheless, average hours worked fell by over five weeks (210 hours) for couples at the lowest quintile. Average hours worked increased for couples at the three middle quintiles. The largest increase in hours worked, nine weeks (360 hours), was for couples at the middle quintile. At the next-to-lowest quintile, average hours worked increased by three weeks (120 hours). At the next-to-highest quintile, average hours rose by almost six weeks (230 hours). But, couples at the 81st to 95th Congressional Research Service 14 . The Trend in Family Income from 1979 to 2010 percentiles worked three weeks (120 hours) less in 2010 than in 1979. The top 5% of couples worked almost a week (30 hours) more in 2010 than in 1979. (See Figure 10.) Figure 10. Changes in Total Hours Worked Annually, by Quintile, Married Couples, 1979 to 2010 400 300 200 100 0 -100 -200 -300 -400 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 Average Lowest quintile Next-to-lowest quintile Middle quintile Next-to-highest quintile 81st to 95th percentiles Top 5% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. The average number of hours worked by married couples peaked at 3,780 hours in 2000. During the expansion of the 1980s, average hours worked increased by six weeks (240 hours). Average hours worked increased at all quintiles. The largest increase, almost 10 weeks (390 hours), occurred among married couples at the middle quintile. Unlike the decline in average hours worked by all family types, during the expansion of the 1990s, average hours worked by married couples increased by three weeks (120 hours). Except at the top quintile, average hours increased for couples at all quintiles. At the middle quintile, average hours worked increased by over five weeks (210 hours). The largest increase, almost 5½ weeks (220 hours), took place among couples at the next-to-highest quintile. However, among the top 5% of couples, average hours worked fell by almost seven weeks (270 hours). From 2000 to 2007, average hours worked fell for couples at all income levels except for the top 5% of couples. The largest decline, over four weeks (170 hours), was for couples at the lowest quintile. Average hours worked fell by two weeks (80 hours) at the next-to-lowest quintile and by over a week (50 hours) for couples at the middle quintile. At the next-to-highest quintile, average hours worked fell by almost two weeks (70 hours) and by almost three weeks (110 hours) for couples at the 81st to 95th percentiles. For the top 5% of couples, on the other hand, total hours worked increased by almost five weeks (190 hours), partially offsetting the decline in hours worked during the expansion of the 1990s. Congressional Research Service 15 . The Trend in Family Income from 1979 to 2010 During and after the 2007-2009 recession, average hours worked fell for married couples at all quintiles. Average hours worked fell more for lower income than upper income couples. The Separate Effects of Hourly Earnings and Hours Worked on Total Family Earnings The previous two sections analyzed the changes in real hourly earnings and total hours worked. The change in average family earnings can be separated into the effect of higher or lower real hourly earnings and the effect of working more or less hours. The analysis in this section is based on estimates of average total hours worked shown in Table A-1 and the estimates of average hourly earnings from Table A-6. Both tables are in the Appendix. All Families In 2010, average total family earnings were $8,090 higher than in 1979. However, for families at the three lowest quintiles, average family earnings were lower in 2010 than 1979 (by $2,530, $4,510, and $2,260 for families at the lowest, next-to-lowest, and middle quintiles, respectively). Most of the decline in earnings was the result of families working fewer hours in 2010 than in 1979. On the other hand, for families at the top two quintiles, average earnings were higher in 2010 than in 1979 (by $6,050, $23,500, and $51,460 for families at the next-to-highest quintile, 81st to 95th percentiles, and the top 5% of families, respectively). Unlike the change in earnings for the three lowest quintiles, the increase in average earnings for families at the top two quintiles was due mostly to an increase in earnings per hour worked. (See Table 2.) During the expansion of the 1980s, the average total earnings of families at the lowest quintile fell because the increase in total hours worked was more than offset by the decline in real hourly earnings. Conversely, at the next-to-lowest and middle quintiles, the decline in real hourly earnings was more than offset by an increase in hours worked. For families at the 81st to 95th percentiles and for the top 5% of families, most of the increase in average earnings was due to greater earnings per hour worked. During the expansion of the 1990s, average total earnings increased for families at all quintiles. Except for families at the lowest quintile, families worked fewer hours. But, the decrease in hours worked was more than offset by an increase in real hourly earnings. During the expansion of the 2000s, average total earnings fell for families at all quintiles. Except for families at the 81st to 95th percentiles and the top 5% of families, average earnings fell mainly because families worked fewer hours. For families at the 81st to 95th percentiles, an increase in earnings per hour was more than offset by a reduction in hours worked. For the top 5% of families, all of the decline in average earnings was due to a drop in earnings per hour. During and after the recession of 2007-2009, average total family earnings fell at all quintiles. The decline was due to both a decline in real hourly earnings and a reduction in hours worked. Congressional Research Service 16 . Table 2. Estimates of the Relative Effects of Changes in Real Hourly Earnings and Total Hours Worked on Family Earnings, by Quintile, 1979 to 2010 (in 2010 dollars) 1979 to 1989 Total change in real earnings Change due to earnings per hour 1989 to 2000 Change due to hours worked Total change in real earnings Change due to earnings per hour 2000-2007 Change due to hours worked Real Earnings Average Total change in real earnings Change due to earnings per hour 2007 to 2010 Change Total Change due to change due to hours in real earnings worked earnings per hour 1979 to 2010 Change Change Total due to Change due to change earnings due to hours in real per hours worked earnings hour worked All Families $7,600 $2,960 $4,640 $5,320 $7,760 -$2,440 -$1,330 $520 -$1,850 -$3,670 $150 -$3,820 -680 -980 300 1,450 1,280 170 -850 50 -900 -2,450 -760 -1,690 -2,530 -480 -2,050 600 -1,880 2,470 1,010 2,610 -1,600 -1,550 50 -1,600 -4,570 -1,940 -2,630 -4,510 -1,270 -3,240 Middle quintile 2,830 -1,170 4,000 1,740 3,460 -1,730 -2,050 -400 -1,660 -4,770 -920 -3,850 -2,260 770 -3,030 Next-to-highest quintile 7,750 720 7,030 4,860 6,000 -1,150 -1,110 520 -1,630 -5,440 -1,060 -4,370 6,050 5,640 410 16,890 9,480 7,410 11,520 17,440 -5,910 -360 2,660 -3,020 -4,550 -1,420 -3,130 23,500 26,630 -3,120 50,770 41,640 9,130 25,590 40,910 -15,320 -14,570 -14,900 330 -11,360 -5,420 -5,940 51,460 59,070 -7,610 Lowest quintile Next-to-lowest quintile 81st – 95th percentiles Top 5% $8,090 $10,620 -$2,530 Married Couples Average 8,880 3,930 4,950 14,020 11,310 2,710 -60 1,910 -1,970 -4,520 180 -4,700 18,570 16,350 2,230 -240 -2,240 2,000 3,480 2,870 610 -420 1,330 -1,750 -4,250 -920 -3,330 -1,430 720 -2,150 Next-to-lowest quintile 1,620 -2,280 3,890 6,150 3,530 2,610 -140 1,010 -1,150 -5,550 -1,690 -3,850 2,070 310 1,760 Middle quintile 5,090 -1,950 7,040 10,230 6,400 3,820 130 1,170 -1,040 -4,720 -1,060 -3,650 10,740 4,010 6,720 Next-to-highest quintile 10,080 5,240 4,830 14,990 9,880 5,110 1,870 3,700 -1,830 -4,090 -570 -3,510 22,850 17,510 5,350 81st 17,610 14,960 2,650 25,780 26,350 -560 2,320 6,310 -3,990 -2,490 520 -3,010 43,230 46,790 -3,560 55,860 48,630 7,230 56,690 73,470 -17,000 -18,000 -33,920 15,920 -17,600 -14,180 -3,420 74,440 73,090 1,350 Lowest quintile – 95th percentiles Top 5% Source: Estimates of the change in annual earnings due to changes in average hours worked and in average hourly earnings are from the unrounded estimates of annual hours worked in Table A-1 in the Appendix and of earnings per hour worked in Table A-6, also in the Appendix. An explanation and illustration of how the estimates in Table 2 were calculated are provided in the Appendix. Note: Estimates are for families headed by persons ages 22 to 62. CRS-17 . The Trend in Family Income from 1979 to 2010 Married Couples In 2010, married couples earned an average of $18,570 more than they earned in 1979. However, couples at the lowest quintile earned an average of $1,430 less in 2010 than in 1979. Their earnings fell mainly because of a reduction in hours worked. Average earnings increased at other quintiles. Most of the increase in average earnings of couples at the next-to-lowest and middle quintiles was due to an increase in hours worked. By contrast, at the top quintile, most of the increase in average earnings was due to higher earnings per hour. During the economic expansion of the 1980s, real hourly earnings fell for married couples at the lowest quintile because the decline in real hourly earnings was greater than the increase in hours worked. For couples at the next-to-lowest and middle quintiles, an increase in hours worked more than offset the decline in real hourly earnings. For couples at the 81st to 95th percentiles and for the top 5% of couples, most of the increase in average earnings was due to greater earnings per hour worked. Average total earnings increased for married couples at all quintiles during the expansion of the 1990s. For couples at the first four quintiles, most of the increase in average earnings was due to higher hourly earnings. But, for couples at the top quintile, all of the increase in average earnings was due to higher earnings per hour. During the economic expansion of the 2000s, average total earnings fell for married couples at the two lowest quintiles. Despite an increase in hourly earnings, average earnings fell because of a reduction in hours worked. At the middle quintile, the small increase in average earnings, an average of $130, was due to higher hourly earnings, which offset a reduction in hours worked. The average earnings of couples at the next-to-highest quintile and at the 81st to 95th percentiles increased despite a reduction in hours worked. On the other hand, for the top 5% of couples, despite an increase in hours worked, average earnings fell because of lower hourly earnings. Other Reasons For Changes in Family Income Family income may change for reasons not considered in this report. For example, family income may change if the age distribution of the labor force changes. Earnings generally rise as workers age and gain work experience. Also, better-educated workers generally earn more than workers with less education. In 2010, the labor force was both more experienced and better educated than in 1979. In 2010, the percentage of the labor force between the ages of 45 and 62 was 10.8 percentage points higher than in 1979 (42.7% versus 31.9%). The percentage of the labor force that had completed at least four years of college increased by 13.0 percentage points (from 21.5% to 34.5%). (See Table 3.) These changes may or may not affect the distribution of income. For example, inequality could rise as the workforce ages and workers gain experience if the increase in earnings is greater for workers who already earned more than other workers. Conversely, inequality could fall as the workforce becomes more educated if the demand for skilled workers falls relative to the supply of educated workers. Congressional Research Service 18 . The Trend in Family Income from 1979 to 2010 Earnings may also be affected by changes in the distribution of workers by occupation or industry.11 For example, from 1979 to 2010, manufacturing employment fell by 7.9 million workers (from 19.4 million to 11.5 million). The manufacturing industry is an important source of employment for less-educated workers, especially men.12 In contrast to the decline in employment in manufacturing, from 1979 to 2010, employment increased by 12.8 million in educational and health services (from 6.8 million to 19.6 million) and by 9.4 million in professional and business services (from 7.3 million to 16.7 million).13 Finally, earnings may change because of institutional changes, such as changes in government regulations that affect labor, product, or financial markets. Increased globalization may affect wages. Changes in wage standards, such as federal, state, and local minimum wages, can affect family earnings. The degree of unionization may affect the earnings of workers and their families. Table 3. Changes in the Age and Education of the Labor Force, 1979 to 2010 Year Age 1979 2010 Percentage Point Change, 1979 to 2010 22-34 35-44 44.6% 32.4% -12.2 23.5% 24.9% 1.5 45-54 19.7% 26.9% 7.2 55-62 12.2% 15.8% 3.5 Total 100.0% 100.0% Education Less than a high school education 19.5% 8.5% -11.0 High school graduatea 37.6% 28.2% -9.4 Some college 21.4% 28.9% 7.5 College graduatea 13.2% 22.9% 9.7 8.3% 11.6% 3.2 100.0% 100.0% Post graduate education Total Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). a. For 2010, high school and college graduates are persons with degrees. For 1979, high school and college graduates are persons who completed four years of high school or college. 11 For an analysis of employment in the manufacturing industry, see CRS Report R41898, Job Creation in the Manufacturing Revival, by Marc Levinson. 12 According to a study of less-skilled workers ages 18 to 54, in 2002, 39% of men and 27% of women were employed in manufacturing. Less-skilled workers were defined as persons with a high school education or less. Rebecca M. Blank and Heidi Shierholz, Exploring Gender Differences In Employment and Wage Trends Among Less-Skilled Workers, National Bureau of Economic Research, Working Paper 12494, August 2006, pp. 4, 15, available at http://www.nber.org/papers/w12494. 13 U.S. Department of Labor, Bureau of Labor Statistics, Current Employment Statistics, available at http://stats.bls.gov/ces. Congressional Research Service 19 . The Trend in Family Income from 1979 to 2010 Conclusion Whether families are defined to include only those headed by married couples or defined to include married couples, single parents, and individuals combined, the average real earnings of families at the middle three quintiles increased during the economic expansions of the 1980s and 1990s. But, for all family types grouped together, the average earnings of families at the three middle quintiles fell during the economic expansion of the 2000s and again during the 2007-2009 recession. As a result, the average real earnings of all families at the next-to-lowest and middle quintiles were lower in 2010 than in 1979. On the other hand, for families headed by married couples at the middle three quintiles, average real earnings were higher in 2010 than in 1979. At the next-to-lowest and middle quintiles, there was little change in average earnings during the economic expansion of the 2000s. The loss of earnings during the recent recession did not offset the gains from the expansions of the 1980s and 1990s. From 1979 to 2010, married couples at the three middle quintiles benefited from an increase in both real hourly earnings and total hours worked. At the next-to-lowest and middle quintiles, most of the growth in the earnings of married couples was due to an increase in the number of hours worked. At the lowest quintile, for all family types and for married couples only, average real earnings were lower in 2010 than 1979. Under both family definitions, the decline in average earnings was due mainly to a reduction in hours worked. From 1979 to 2010, average real earnings increased the most for families at the top quintile. Regardless of how families are defined, the average earnings of these families grew mainly because of an increase in earnings per hour. The average hours worked (over 4,000 annually) by the top quintile of families suggests that they are two-earner families. Congressional Research Service 20 . The Trend in Family Income from 1979 to 2010 Appendix. Data and Methodology This appendix provides a more detailed description of the data and methodology used in this report. Table A-4, Table A-5, and Table A-6 show the data discussed in the body of the report. Data The analysis in this report is based on data from the Annual Social and Economic (ASEC) supplement to the monthly Current Population Survey (CPS). The CPS is a household survey conducted by the U.S. Bureau of the Census for the Bureau of Labor Statistics (BLS) of the U.S. Department of Labor. The monthly CPS is the source of the national monthly unemployment rate and other labor market information. The sample for the ASEC supplement is representative of the civilian noninstitutional population of the United States. The sample for the supplement includes members of the Armed Forces living in civilian housing units on a military base or in a family not on a military base. The sample does not include persons living in institutions (such as psychiatric hospitals, nursing homes, or correctional facilities). The 2011 supplement collected information from families living in about 75,900 households.14 The ASEC supplement asks questions about individual income and earnings for the previous year. Income consists of money income, and includes wages and salaries, income from selfemployment, interest and dividends, social security and pension benefits, public assistance, unemployment and worker’s compensation, alimony and child support, and other types of money income. Wages and salaries include earnings from self-employment, tips, commissions, and cash bonuses. Wages and salaries tend to be reported more accurately than income from other sources. Money income does not include in-kind transfers for food, housing, healthcare, or energy assistance. Nor does it include capital gains or noncash or deferred compensation (e.g., employer contributions for health insurance or to a retirement plan). Money income is income before taxes or other deductions.15 In this report, total family income consists of the sum of money income of all members of a family. In Table A-4 and Table A-6, comparisons of real income and earnings between consecutive years should be made with caution. When answering questions about annual income or earnings, some respondents may round off their answers. For example, some individuals may report that they earned $50,000 the previous year, when they may have actually earned either more or less than $50,000. From one year to the next, this rounding may affect the observed trend in real income or earnings. In the CPS, a family is defined as a group of two or more persons who are living together and who are related by birth, marriage, or adoption.16 The CPS defines single individuals as either “nonfamily householders” or “secondary individuals.” 14 U.S. Census Bureau, Current Population Survey, 2011 Annual Social and Economic (ASEC) Supplement, pp. 1-1, 93, G-2, available at http://www.census.gov/apsd/techdoc/cps/cpsmar11.pdf. 15 Ibid., p. 9-4. 16 Ibid., p. 9-2. Congressional Research Service 21 . The Trend in Family Income from 1979 to 2010 This report uses two categories of “family.” First, the report analyzes the incomes and earnings of “all families,” where married couples, single parents, and unrelated individuals are combined. Second, the incomes and earnings of married couples are analyzed separately. In this report, a primary family and a related or unrelated subfamily living in the same family are treated as separate families. An unmarried couple with children is treated as a single parent and a single individual. Unmarried couples without children are treated as two single individuals. The analysis in this report includes families headed by persons between the ages of 22 and 62. In both 1979 and 2007, the labor force participation rate first reached 75% for persons age 22. In 2010, the labor force participation rate first reached 75% for persons age 24. In both 2007 and 2010, for persons over age 62 (in 1979, for persons over 61), the labor force participation rate fell below 50%. Topcoding To protect the confidentiality of survey participants, the CPS assigns an income amount to higher income persons. Changes in these amounts, or topcodes, can affect the observed trend in real income and earnings. Since 1996, for persons with earnings above the topcoded amounts, the amount of earnings reported in the public CPS files is the average earnings of workers with similar characteristics. In the CPS, average earnings are calculated for persons based on gender, race, ethnicity, and whether or not a person works full-time, year-round. Since 1999, for persons with incomes above the topcoded amounts, the public CPS files report the average income of persons with topcoded income. This report uses average topcoded earnings and income as reported in the CPS for the years that these averages are in the public-use files.17 For consistency over the period from 1979 through 2010, the analysis in this report uses data published in a report by Jeff Larrimore et al.18 Data in their report replicate the calculations of average topcoded earnings and income for the years before these averages were available to the public in the CPS. The calculations of average income and earnings in the report by Larrimore et al. follow the approach used in the CPS; average earnings and income are calculated for persons based on gender, race, ethnicity, and whether or not a person worked full-time, year-round. Nevertheless, average topcoded earnings and income published by Larrimore et al. are subject to some limitations. The economists were given access to internal CPS data. But the internal data are also topcoded, albeit at higher levels than the data available to public users. In some years, changes in the highest earnings amounts in the internal CPS data files may affect the observed trend in inequality. Between the years 1993 and 1994, the highest amount of earnings available in the internal CPS data increased from $299,999 to $999,999.19 This increase affects the observed 17 Instead of average income, median income is an alternative measure of family income. If the incomes of all families are ranked from lowest to highest, the median is the value in the middle of the distribution. Because some families have higher incomes, average income is greater than median income. Evidence suggests that the growth in average wage and salary income has kept pace with the growth in productivity, while the growth in median income has not. See Ian DewBecker and Robert Gordon, Where Did the Productivity Growth Go? Inflation Dynamics and the Distribution of Income, National Bureau of Economic Research, Working Paper 11842, December 2005, available at http://www.nber.org. 18 Jeff Larrimore, Richard V. Burkhauser, Shuaizhang Feng, and Laura Zayatz, Consistent Cell Means for Topcoded Incomes in the Public Use March CPS (1976-2007), National Bureau of Economic Research, Working Paper 13941, April 2008, available at http://www.nber.org/papers/w13941. 19 Ibid., p. 49. Congressional Research Service 22 . The Trend in Family Income from 1979 to 2010 change in average earnings and income from the years 1992 to 1993. (Recall that the ASEC supplement asks questions about earnings and income for the previous year.) Because of the changes in the internal CPS data from 1993 to 1994, this report follows the approach of Richard Burkhauser et al. and does not include the changes in earnings and income from 1992 to 1993 in the calculation of changes in average earnings and income. 20 For the same reason, the calculations of changes in the average income and earnings of families with the top 5% of income and earnings do not include the changes from 1992 to 1993. Figure A-1 shows how the change in topcoding affects average real family income. Using the estimates in the report by Larrimore et al. to calculate average income, the income of the top 5% of families increased by $43,600 from 1992 to 1993 (from $223,230 to $266,830). This amount cannot be separated into the actual change in income and the amount that is due to the change in internal topcoding. If the change in income from 1992 to 1993 was known, the actual change in average income from 1979 to 2010 would probably be more than the $55,280 amount shown in this report. Average income increased during the economic expansion of the 1990s and likely increased from 1992 to 1993 as well.21 Figure A-1. Average Real Family Income and Average Real Income of the Top 5% of Families, 1979 to 2010 (in 2010 dollars) $350,000 $300,000 $250,000 $200,000 $150,000 $100,000 $50,000 Average Average, 1992-1993 20 10 20 05 20 00 19 95 19 90 19 85 19 79 $0 Top 5% Top 5%, 1992-1993 Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. 20 Richard V. Burkhauser, Jeff Larrimore, and Kosali I. Simon, A “Second Opinion” on the Economic Health of the American Middle Class, National Bureau of Economic Research, Working Paper 17164, p. 7, available at http://www.nber.org/papers/w17164. 21 For a discussion of the topcoding issue, see Paul Ryscavage, “A Surge in Growing Income Inequality?” Monthly Labor Review, vol. 118, August 1995, pp. 51-61. Congressional Research Service 23 . The Trend in Family Income from 1979 to 2010 Controlling for Family Hours Worked From 1979 to 2010, the labor force participation rate of women increased significantly, while the labor force participation rate of men declined. Thus, part of the observed trend in family income from 1979 to 2010 may be due to the increased hours worked by women. One way to control for changes in total hours worked would be to divide total family income by the total number of hours worked by all family members. But, the share of earnings in family income is not consistent across the income distribution. Earnings typically account for a smaller share of income for families in the bottom than at the top of the distribution. (See Figure A-2.) Thus, in this report, to account for changes in hours worked, total family earnings are divided by total family hours worked. Figure A-2. Family Earnings as a Share of Family Income, by Quintile, 1979, 2007, and 2010 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Lowest quintile Next-to-lowest Middle quintile quintile 1979 Next-tohighest quintile 2007 81st to 95th percentiles Top 5% 2010 Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families headed by persons ages 22 to 62. Estimating the Relative Effects of Changes in Earnings Per Hour and Changes in Total Hours Worked on Total Family Earnings Table 2 in the main body if this report decomposes changes in average total family earnings into the amount that is due to changes in family earnings per hour worked and the amount that is due to changes in total hours worked by all family members. The decomposition is based on the estimates of the changes in hours worked and earnings per hour from Table A-1 and Table A-6 shown in this Appendix. Congressional Research Service 24 . The Trend in Family Income from 1979 to 2010 Table A-1. Average Hours Worked by All Family Members, by Quintile, 1979 to 2010 Total Hours Worked 1979 1989 2000 2007 Annual Hours Change in Hours Worked 2010 19791989 19892000 20002007 20072010 19792010 All Families Average 3,002 3,237 3,123 3,044 2,882 235 -114 -79 -162 -120 Lowest quintile 1,675 1,714 1,736 1,627 1,415 39 22 -109 -212 -260 Next-to-lowest quintile 2,403 2,600 2,474 2,353 2,147 197 -126 -121 -206 -256 Middle quintile 2,954 3,200 3,096 2,999 2,771 246 -104 -97 -228 -183 Next-to-highest quintile 3,427 3,779 3,724 3,649 3,447 352 -55 -75 -202 20 81st to 95th percentiles Top 5% 4,123 4,422 4,210 4,110 4,007 299 -212 -100 -103 -116 4,386 4,598 4,305 4,310 4,218 212 -293 5 -92 -168 Married Couples Average 3,421 3,660 3,779 3,702 3,520 239 119 -77 -182 99 Lowest quintile 2,417 2,627 2,690 2,522 2,205 210 63 -168 -317 -212 Next-to-lowest quintile 3,072 3,340 3,518 3,443 3,190 268 178 -75 -253 118 Middle quintile 3,342 3,731 3,936 3,883 3,697 389 205 -53 -186 355 Next-to-highest quintile 3,801 4,025 4,243 4,170 4,032 224 218 -73 -138 231 81st to 95th Top 5% percentiles 4,391 4,490 4,472 4,358 4,274 99 -18 -114 -84 -117 4,264 4,412 4,146 4,333 4,290 148 -266 187 -43 26 Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: Estimates are for families headed by persons ages 22 to 62. The estimates are for families with reported earnings. Families are ranked by total income. Average total earnings over time change for three reasons: the change in earnings per hour worked, the change in hours worked, and the interaction of changes in earnings and hours worked. Table 2 shows that the change in average real family earnings from 1979 to 2010 was $8,090. Table A-1 shows that the average number of hours worked per year by all families fell by 120 hours, from 3,002 hours to 2,882 hours. Table A-6 shows that average real earnings per hour worked increased from $19.25 to $23.59. Taking into account the issue of topcoding discussed above, from 1979 to 2010, average real earnings per hour worked increased by $3.61. Table A-2 illustrates the decomposition of the change in average total earnings from 1979 to 2010. Congressional Research Service 25 . The Trend in Family Income from 1979 to 2010 Table A-2. Illustration of the Decomposition of Average Annual Earnings Due to Changes in Real Hourly Earnings and Annual Hours Worked Change in average annual earnings due to the change in annual hours worked, holding real hourly earnings constant at their 1979 level: (Average annual hours worked2010 - Average annual hours worked1979) x Average real hourly earnings1979 ( 2,882 - 3,002 ) x $19.25 = -120 x $19.25 = -$2,310.00 Change in average annual earnings due to the change in real hourly earnings, holding annual hours worked constant at their 1979 level (removing the change in hourly earnings from 1992 to 1993 because of the issue of topcoding): ((Average real hourly earnings2010 - Average real hourly earnings1979)) (Average real hourly earnings1993 - Average real hourly earnings1992)) x Average annual hours worked1979 ($23.59 - $19.25) – ($20.57- $19.84) x 3,002 = ($4.34 - $0.73) x 3,002 = $3.61 x 3,002 = $10,837.22 Change in average annual earnings due to the changes in both real hourly earnings and annual hours worked (i.e., the interaction effect): ((Average real hourly earnings2010 - Average real hourly earnings1979)) (Average real hourly earnings1993 - Average real hourly earnings1992)) x (Average annual hours worked2010 - Average annual hours worked1979) (($23.59 - $19.25) - ($20.57- $19.84)) x ( 2,882 - 3,002 ) = ($4.34 - $0.73) x -120 = $3.61 x -120 = - $433.20 Decomposition of the change in average annual earnings due to the changes in real hourly earnings and average annual hours worked. One-half of the interaction effect is assigned to the change in real hourly earnings and one-half is assigned to the change in annual hours worked: Change due to the change in average real hourly earnings: Change due to the change in average annual hours worked: Total change in average total earnings: $10,837.22 - $216.60 = $10,620.62 -$2,310.00 - $216.60 = -$2,526.60 $8,094.02 The change in average total earnings of $8,094.02 rounds to $8,090, which is shown in Table 2. Gini Coefficient This report uses the Gini coefficient to analyze changes in the distribution of family income. The Gini coefficient ranges from 0 to 1, with a higher coefficient indicating greater inequality. Because of topcoding, the Gini coefficient may understate the degree of family inequality. In addition, changes in family size and composition may affect family income. In this report, the calculation of the Gini coefficient adjusts real family income using an equivalence scale from the U.S. Census Bureau. The scale accounts for the different needs of adults and children and the economies of scale of living in a larger family. The equivalence scale assumes that, on average, children consume less than adults and that, as family size increases, expenses do not increase at the same rate. The scale also assumes that for a first child, the increase in expenses is greater for a single parent than for two adults.22 22 Carmen DeNavas-Walt, Bernadette D. Proctor, and Jessica Smith, Income, Poverty, and Health Insurance Coverage in the United States: 2007, U.S. Census Bureau, P60-235, August 2008, pp. 9-10, available at http://www.census.gov/ prod/2008pubs/p60-235.pdf. Congressional Research Service 26 . The Trend in Family Income from 1979 to 2010 The scale used by the Census Bureau divides family income by (A + PC)F, where A represents the number of adults in a family, C represents the number of children, P represents the adult equivalent of one child, and F is an economy of scale factor. For single parents, the scale is (A + 0.8 + 0.5*C)0.7, where 0.8 is the adult equivalent of the first child and 0.5*C represents each additional child. For other families, the scale is (A + 0.5*C)0.7.23 In calculating equivalenceadjusted income, family members who are 18 and over are counted as adults. Persons under 18 are counted as children, unless they are married, parents, or living on their own, in which case they are counted as adults. Table A-3 illustrates how the equivalence scale adjusts family income for differences in family size and composition. The examples include an individual with no children, a family with one adult (e.g., a single parent) and either one or two children, a family with two adults (e.g., a married couple) and either one or two children, and a family of three adults and either one or two children. First, the above scales are used to calculate the values in column 3. Next, all values are divided by 2.158 (row 5, column 3), which sets the scale for a two-adult, two-child family to 1.000. The Gini coefficient is calculated using family income divided by the values shown in column 4. Table A-3. Illustration of Equivalence Scale Number of Adults (1) Number of Children (2) Calculated Value (3) Calculated Value Standardized to a Family of Two Adults and Two Children (4) 1 1 1 2 2 3 3 0 1 2 1 2 1 2 1.000 1.509 1.791 1.899 2.158 2.404 2.639 0.463 0.699 0.830 0.880 1.000 1.114 1.223 23 Kathleen Short, Experimental Poverty Measures: 1999, U.S. Census Bureau, P60-216, October 2001, pp. A-1 to A-2, available at http://www.census.gov/prod/2001pubs/p60-216.pdf. Congressional Research Service 27 . Table A-4. Average Real Income: All Families and Married Couples, 1979 to 2010 1979 1980 1981 1982 1983 1984 1985 1986 Real Income Average 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 All Families $60,094 $57,896 $56,575 $56,361 $56,410 $57,970 $59,667 $62,021 $63,373 $64,064 $65,261 $63,375 $62,212 $62,336 $64,646 $66,313 $66,786 $68,331 Lowest quintile 15,038 14,058 13,208 12,410 11,979 12,200 12,686 12,789 12,944 13,118 13,488 12,939 12,615 12,119 11,915 12,359 12,726 12,972 Second quintile 34,435 32,832 31,420 30,644 30,201 30,870 31,630 32,510 32,959 33,264 33,649 32,601 31,824 31,102 30,632 31,247 31,973 32,464 Middle quintile 52,218 50,341 49,062 47,946 47,826 48,917 50,009 51,834 52,774 53,003 53,234 51,746 51,105 50,798 49,946 51,023 51,802 52,460 Next-to-highest quintile 72,503 70,471 69,360 68,734 69,230 71,233 72,518 75,078 76,723 77,135 77,795 75,977 75,326 75,615 75,641 77,143 77,895 79,050 81st – 95th percentiles Top 5% 104,666 102,378 101,086 101,979 102,754 105,943 108,317 112,439 114,964 116,404 118,402 115,707 114,624 114,961 117,811 120,388 121,539 123,509 191,046 179,918 176,001 182,321 182,968 188,670 200,965 214,216 220,941 225,872 237,307 227,209 216,820 223,225 266,829 277,957 273,461 288,194 Married Couples Average 73,163 70,820 69,634 69,115 69,616 72,287 74,507 77,826 79,908 80,893 82,960 81,134 80,112 81,053 84,945 87,467 88,174 90,242 Lowest quintile 26,487 25,379 24,058 22,296 22,075 22,985 23,375 24,586 25,254 25,583 25,923 25,536 25,056 24,664 24,189 25,055 25,823 26,345 Next-to-lowest quintile 48,778 47,052 45,666 43,843 44,058 45,713 46,388 48,497 49,837 50,092 50,746 49,899 49,324 49,702 48,735 50,332 51,362 52,383 Middle quintile 65,111 63,311 62,420 60,940 61,437 63,904 65,068 67,483 69,372 69,836 70,841 69,504 69,375 70,217 70,066 71,711 72,688 74,076 Next-to-highest quintile 84,283 82,809 81,972 81,624 82,381 85,702 87,647 90,690 93,098 94,147 95,679 94,101 93,762 94,272 96,083 97,921 99,209 100,373 81st – 95th percentiles 117,986 115,633 114,059 115,560 117,097 121,068 124,257 129,532 132,413 135,037 138,323 135,917 134,805 136,256 142,096 147,661 145,818 148,972 Top 5% 210,615 195,237 193,996 200,739 201,075 209,153 227,394 242,788 250,592 253,995 271,420 258,688 247,519 256,718 316,153 326,158 329,341 345,069 CRS-28 . 1997 1998 1999 2000 2001 2002 2004 2005 2006 2007 2008 2009 2010 All Families Real Income Average 2003 $70,564 $72,669 $73,361 $75,558 $75,539 $74,000 $73,528 $72,979 $73,615 $74,884 $73,426 $71,429 $70,374 $68,599 Lowest quintile 13,306 13,950 14,385 14,779 14,467 14,163 13,639 13,614 13,527 13,945 13,703 13,055 12,271 11,688 Next-to-lowest quintile 33,156 34,672 35,425 35,488 35,079 34,473 33,848 33,453 33,495 33,960 33,739 32,199 30,854 29,924 Middle quintile 53,845 55,980 57,126 57,172 56,539 55,757 55,237 54,578 54,754 55,099 55,048 53,111 51,687 50,558 Next-to-highest quintile 81,396 84,239 86,878 86,727 86,436 85,468 85,527 84,214 84,391 85,621 85,376 82,710 81,345 80,220 81st – 95th percentiles 128,293 132,114 137,160 137,001 137,097 135,273 136,291 134,764 136,134 138,102 136,788 133,755 132,667 130,803 Top 5% 299,537 301,628 280,384 323,428 329,333 314,633 308,652 311,807 319,119 328,778 306,606 302,963 304,778 289,934 Married Couples Average 93,503 96,693 97,394 101,927 101,603 100,478 100,394 100,233 101,022 103,754 102,035 99,172 97,611 95,666 Lowest quintile 27,365 27,831 29,114 29,765 29,277 28,727 28,363 28,045 28,679 29,346 29,325 27,372 25,705 24,934 Next-to-lowest quintile 54,100 55,816 57,506 58,145 57,775 57,377 56,805 56,829 56,963 57,221 58,285 55,559 53,379 52,656 Middle quintile 75,944 78,577 81,493 82,166 81,870 81,921 81,821 81,321 81,015 82,382 83,227 80,029 78,204 77,938 Next-to-highest quintile 103,934 108,085 111,682 112,429 112,432 111,867 113,144 112,334 112,478 114,996 114,713 111,439 110,168 109,528 81st – 95th percentiles 154,672 160,058 166,970 168,571 168,354 167,612 168,584 168,439 170,707 173,720 171,502 168,932 167,045 165,785 Top 5% 360,288 372,374 327,721 402,698 401,418 387,085 381,362 385,027 391,660 418,018 383,843 378,782 381,198 355,643 CRS-29 . Dollar Change 1979-1989 1989-2000 2000-2007 Percent Change 2007-2010 Real Earnings 1979-2010 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 All Families Averagea $5,167 $7,987 -$2,132 -$4,827 $6,195 8.6% 12.4% -2.8% -6.6% 9.8% Lowest quintile -1,550 1,291 -1,076 -2,015 -3,350 -10.3% 9.6% -7.3% -14.7% -22.3% Next-to-lowest quintile -786 1,839 -1,749 -3,815 -4,511 -2.3% 5.5% -4.9% -11.3% -13.1% Middle quintile 1,016 3,938 -2,124 -4,490 -1,660 1.9% 7.4% -3.7% -8.2% -3.2% Next-to-highest quintile 5,292 8,932 -1,351 -5,156 7,717 7.3% 11.5% -1.6% -6.0% 10.6% 13,736 18,599 -213 -5,985 26,137 13.1% 15.7% -0.2% -4.4% 25.0% 46,261 42,517 -16,822 -16,672 55,284 24.2% 15.3% -5.2% -5.4% 25.5% 81st 95th – Top 5%a percentiles Married Couples Averagea 9,797 15,075 108 -6,369 18,611 13.4% 17.7% 0.1% -6.2% 23.4% Lowest quintile -564 3,842 -440 -4,391 -1,553 -2.1% 14.8% -1.5% -15.0% -5.9% Next-to-lowest quintile 1,968 7,399 140 -5,629 3,878 4.0% 14.6% 0.2% -9.7% 8.0% Middle quintile 5,730 11,325 1,061 -5,289 12,827 8.8% 16.0% 1.3% -6.4% 19.7% Next-to-highest quintile 11,396 16,750 2,284 -5,185 25,245 13.5% 17.5% 2.0% -4.5% 30.0% 81st – 95th percentiles 20,337 30,248 2,931 -5,717 47,799 17.2% 21.9% 1.7% -3.3% 40.5% Top 5%a 60,805 71,843 -18,855 -28,200 85,593 28.9% 22.0% -4.7% -7.3% 34.4% Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: All income amounts are in 2010 dollars. Estimates are for families that reported income and were headed by persons 22 to 62 years of age. a. CRS-30 Because of a change in topcoding, the change in average income for all families and the change in average income for the top 5% of families do not include the change in income from 1992 to 1993. . The Trend in Family Income from 1979 to 2010 Table A-5. Gini Coefficient: All Families and Married Couples, 1979 to 2010 Year All Families Married Couples 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987 1986 1985 1984 1983 1982 1981 1980 1979 0.44465 0.44488 0.43629 0.43178 0.43797 0.43746 0.43478 0.43450 0.43335 0.43685 0.42915 0.41948 0.42812 0.43297 0.43144 0.42555 0.43084 0.43028 0.40448 0.39942 0.40136 0.40187 0.39834 0.39340 0.39302 0.38999 0.38699 0.38651 0.38133 0.37030 0.36190 0.35764 0.40431 0.40866 0.40366 0.39763 0.40823 0.40158 0.40241 0.40125 0.40076 0.40176 0.40227 0.38306 0.39880 0.39481 0.39267 0.38817 0.39442 0.39386 0.36479 0.36152 0.36369 0.36655 0.35701 0.35375 0.35584 0.35585 0.34653 0.34641 0.34407 0.33010 0.32290 0.32267 Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Note: Estimates are for families that reported income and were headed by persons 22 to 62 years of age. Congressional Research Service 31 . Table A-6. Average Real Family Earnings Per Family Hour Worked: All Families and Married Couples, 1979-2010 1979 1980 1981 1982 1983 1984 1985 1986 Real Earnings Average Lowest quintile 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 All Families $19.25 $18.88 $18.68 $18.81 $18.78 $18.75 $19.23 $19.78 $19.90 $20.10 $20.20 $19.75 $19.66 $19.84 $20.57 $20.94 $20.85 $21.28 8.05 7.74 7.39 7.27 7.01 7.00 7.21 7.31 7.34 7.42 7.47 7.38 7.30 7.06 6.93 7.12 7.28 7.59 Next-to-lowest quintile 12.92 12.53 12.15 11.86 11.74 11.63 11.85 12.07 12.10 12.09 12.17 11.88 11.87 11.65 11.40 11.56 11.80 12.11 Middle quintile 16.44 16.21 16.06 15.73 15.66 15.54 15.79 16.23 16.17 16.35 16.06 15.64 15.58 15.68 15.31 15.57 15.51 15.73 Next-to-highest quintile 19.86 19.41 19.15 19.26 19.09 19.39 19.56 19.87 20.17 20.03 20.06 19.63 19.47 19.52 19.41 19.61 19.76 19.78 81st – 95th percentiles 23.66 23.22 23.34 23.55 23.76 24.04 24.49 25.18 25.47 25.64 25.88 25.43 25.57 25.64 25.83 26.49 26.68 26.94 Top 5% 38.41 37.56 36.30 37.75 37.00 36.73 40.47 42.36 42.35 45.22 47.68 45.54 43.98 45.39 57.22 59.52 56.57 59.59 Married Couples Average 20.14 19.70 19.49 19.59 19.50 19.59 20.16 20.75 20.90 21.08 21.25 20.74 20.66 20.95 21.93 22.34 22.29 22.64 9.99 9.66 9.22 8.75 8.68 8.64 8.77 9.16 9.27 9.20 9.10 9.04 9.14 8.90 8.76 8.97 9.20 9.55 Next-to-lowest quintile 14.88 14.62 14.31 13.83 13.78 13.83 13.83 14.23 14.27 14.33 14.17 13.77 13.67 13.74 13.31 13.74 13.87 13.98 Middle quintile 18.37 17.80 17.62 17.38 17.29 17.39 17.60 17.85 18.12 18.06 17.82 17.56 17.48 17.62 17.45 17.63 17.77 17.96 Next-to-highest quintile 20.91 20.70 20.55 20.54 20.37 20.87 21.28 21.55 21.96 21.75 22.25 21.62 21.54 21.73 22.01 22.37 22.29 22.22 81st – 95th percentiles 25.05 24.28 24.75 25.29 25.28 25.88 26.60 27.48 27.56 28.30 28.42 28.06 27.85 28.22 28.88 29.43 30.10 30.50 Top 5% 43.28 42.86 39.65 41.15 39.99 39.68 44.24 47.24 47.23 49.57 54.49 51.99 51.00 54.21 71.67 75.10 68.49 70.62 Lowest quintile CRS-32 . 1997 1998 1999 2000 2001 2002 Real Earnings Average 2003 2004 2005 2006 2007 2008 2009 2010 All Families $21.82 $22.36 $22.27 $23.37 $23.86 $23.87 $23.85 $23.52 $23.55 $23.79 $23.54 $23.42 $23.88 $23.59 7.63 7.84 7.93 8.21 8.20 8.20 8.05 8.01 7.99 8.17 8.24 8.02 7.89 7.74 Next-to-lowest quintile 12.47 12.81 13.04 13.20 13.28 13.34 13.09 12.99 12.88 13.09 13.22 12.80 12.59 12.36 Middle quintile 16.10 16.74 16.97 17.16 17.13 17.27 17.22 16.91 16.96 16.92 17.03 16.60 16.88 16.71 Next-to-highest quintile 20.41 21.05 21.49 21.66 21.81 22.03 22.09 21.69 21.53 21.72 21.80 21.45 21.64 21.50 81st – 95th percentiles 27.64 28.57 29.16 29.92 30.54 30.40 30.74 30.22 30.26 30.84 30.56 30.25 30.65 30.21 Top 5% 61.59 61.68 57.99 68.70 73.23 69.74 68.02 68.30 69.21 71.03 65.24 67.52 66.75 63.97 Lowest quintile Married Couples Average 23.24 23.99 23.79 25.27 25.65 25.87 25.87 25.70 25.62 26.10 25.78 25.66 26.17 25.83 9.70 9.88 9.98 10.18 10.36 10.41 10.33 10.22 10.26 10.48 10.69 10.32 10.40 10.30 Next-to-lowest quintile 14.39 14.74 15.04 15.20 15.29 15.44 15.34 15.37 15.27 15.16 15.49 14.97 15.10 14.98 Middle quintile 18.22 18.70 19.19 19.49 19.45 19.81 19.87 19.68 19.55 19.71 19.79 19.42 19.44 19.51 Next-to-highest quintile 23.10 23.84 24.31 24.64 25.20 25.46 25.51 25.30 25.19 25.53 25.52 25.02 25.57 25.38 81st 31.32 32.53 33.28 34.30 34.66 35.11 35.85 35.35 35.52 36.09 35.73 35.73 35.80 35.85 74.01 76.30 68.21 89.12 90.71 86.96 82.31 83.18 83.23 90.53 81.12 83.55 83.98 77.83 Lowest quintile – 95th percentiles Top 5% CRS-33 . Dollar Change 1979-1989 1989-2000 2000-2007 Percent Change 2007-2010 Real Earnings 1979-2010 1979-1989 1989-2000 2000-2007 2007-2010 1979-2010 All Families Averagea $0.95 $2.44 $0.17 $0.05 $3.61 4.9% 11.8% 0.7% 0.2% 17.7% Lowest quintile -0.58 0.74 0.03 -0.50 -0.31 -7.2% 9.9% 0.4% -6.1% -3.9% Next-to-lowest quintile -0.75 1.03 0.02 -0.86 -0.56 -5.8% 8.5% 0.2% -6.5% -4.3% Middle quintile -0.38 1.10 -0.13 -0.32 0.27 -2.3% 6.8% -0.8% -1.9% 1.6% Next-to-highest quintile 0.20 1.60 0.14 -0.30 1.64 1.0% 8.0% 0.6% -1.4% 8.3% 81st – 95th percentiles 2.22 4.04 0.64 -0.35 6.55 9.4% 15.6% 2.1% -1.1% 27.7% Top 5%a 9.27 9.19 -3.46 -1.27 13.73 24.1% 15.3% -5.0% -1.9% 30.0% Married Couples Averagea 1.11 3.04 0.51 0.05 4.71 5.5% 13.8% 2.0% 0.2% 21.8% Lowest quintile -0.89 1.08 0.51 -0.39 0.31 -8.9% 11.9% 5.0% -3.6% 3.1% Next-to-lowest quintile -0.71 1.03 0.29 -0.51 0.10 -4.8% 7.3% 1.9% -3.3% 0.7% Middle quintile -0.55 1.67 0.30 -0.28 1.14 -3.0% 9.4% 1.5% -1.4% 6.2% Next-to-highest quintile 1.34 2.39 0.88 -0.14 4.47 6.4% 10.7% 3.6% -0.5% 21.4% 81st – 95th percentiles 3.37 5.88 1.43 0.12 10.80 13.5% 20.7% 4.2% 0.3% 43.1% 11.21 17.17 -8.00 -3.29 17.09 25.9% 23.8% -9.0% -4.1% 33.8% Top 5%a Source: CRS analysis of data from the Annual Social and Economic (ASEC) supplement to the Current Population Survey (CPS). Notes: All income amounts are in 2010 dollars. Estimates are for families that reported income and were headed by persons 22 to 62 years of age. a. CRS-34 Because of a change in topcoding, the change in average hourly earnings for all families and the change in average hourly earnings for the top 5% of families do not include the change in earnings from 1992 to 1993. . The Trend in Family Income from 1979 to 2010 Author Contact Information Gerald Mayer Analyst in Labor Policy gmayer@crs.loc.gov, 7-7815 Congressional Research Service 35