Issues in labor supply.
THE DETERMINANTS oF LABOR SUPPLY
In an accounting sense, the supply of labor available for the production of goods and services depends on the size and the composition of the population, the proportion of the population working or looking for work (the rate of labor force participation), the number of hours worked each week, and the number of weeks worked each year. I Over the past four decades, these determinants have affected the supply of labor in various ways.
The size and composition of the population is the first major building block of the labor force. Over the past forty years, the working-age population of the United States (that is, noninstitutionalized civilians sixteen years of age and older) has grown significantly: At the end of 1990, it stood at 189 million (chart 1), 84 million more than its level in 1950. Three sources of change determine the pattern of growth in the population: births, deaths, and net immigration.
During the 1960s and 1970s, the growth of the working-age population was influenced mainly by the sharp rise in birth rates from the end of World War II to the early 1960s-the period of the so-called baby boom. As the large baby-boom cohort reached working age between 1961 and 1979, the overall working-age population grew 1.9 percent per year, and the number of inexperienced workers seeking jobs bulged (chart 2). For example, sixteen to twenty-four year olds, who had accounted for 18 percent of the working-age population in the 1950s, made up nearly 22 percent of the working-age population by 1970. In contrast, the baby bust, which began in the early 1960s, slowed the pace of growth of the working-age population to about I percent in the 1980s; and the proportion of sixteen to twenty-four year olds in the working-age population shrank back to 17 percent by 1990.
The average lifespan, the second determinant of population trends, has lengthened because of advances in health care. Average life expectancy has increased from seventy years in 1960 to seventy-five years in 1989. As a result, persons sixty-five years of age and older are a growing proportion of the population.
The final factor that has influenced population growth in recent decades is the net flow of immigrants. Between 1980 and 1989, legal immigration to the United States averaged 580,000 persons per year-about one-quarter of I percent of the U.S. population. This pace was well above that of the 1970s, which was 450,000 per year. Although estimates are imprecise, illegal immigration, according to the Immigration and Naturalization Service, probably has added another 100,000 to 300,000 per year to the total of legal immigration.
Recently, new laws have resulted in large fluctuations in the flow of immigrants. Efforts to control illegal immigration led to the Immigration Reform and Control Act of 1986, which attempted to restrict the employment opportunities of illegal aliens by imposing penalties on employers who hired them. However, the recent Immigration Act of 1990 works in the opposite direction: It allows for an increase in total immigration, for an increase in the immigration of individuals with skills that are in short supply, and for the admission of immigrants from underrepresented countries. The Immigration and Naturalization Service estimates that this act could increase legal immigration from roughly 600,000 in 1989 to as much as 800,000 annually over the next several years.
Labor Force Participation
Besides sizable increases in the working-age population, proportion of the working-age population that is either working or looking for work has been on an upward trend throughout the past four decades (chart 3). In the early 1950s, the rate of labor force participation averaged roughly 59 percent; by 1990, the rate had risen to more than 66 percent.
Perhaps the most noteworthy change that has taken place in the labor market over the past forty years is the vast increase in the proportion of women in the civilian labor force. As chart 3 shows, increasing participation among women has more than accounted for the overall rise in the labor force participation rate. As late as 1950, only 34 percent of women were in the labor force. By 1980, women's participation rate had risen to 52 percent; and by 1990, it had moved up to 58 percent.
Rising participation among women has occurred in most age categories (chart 4), with the greatest increases evident for women of childbearing age. The participation rate for women twenty to twenty-four years of age increased from 46 percent in 1960 to 69 percent in 1980 and now stands at nearly 72 percent. For women twenty-five to thirty-four years of age, the participation rate has more than doubled, from 36 percent in 1960 to 74 percent in 1990. Indeed, the jump in the labor force participation rate of married women with children under the age of six has been dramatic, tripling from 19 percent in 1960 to 59 percent in 1990.
In contrast, participation rates among men generally have declined over the past three decades. The most substantial decreases have been among men from fifty-five to sixty-four years of age, whose rate fell from 87 percent in 1970 to 68 percent in 1990; the participation rate for men over sixty-five years of age also has dropped significantly from 33 percent in 1970 to 16 percent by 1990. These declines reflect a trend toward earlier retirement. In contrast, participation rates for men from twenty-five to fifty-four years of age have declined only slightly.
The dramatic demographic shifts associated with the baby boom and the baby bust and the secular changes in various social and economic conditions have influenced individuals' decisions about how much labor to supply to the market. The following section lays out a conceptual framework for analyzing individual labor supply. Subsequent sections use the framework to analyze shifts in labor supply in the short and the long terms.
LABOR SUPPLY DECISIONS: A CONCEPTUAL FRAMEWORK
The basic framework for analyzing an individual's decision to participate in the labor market is a model of the allocation of time between work and leisure.2 The model describes the factors that will influence an individual's response to an increase in the wage rate and, in particular, identifies two opposing influences the substitution effect and the income effect. As the wage rate goes up, the opportunity cost of leisure rises-that is, the income forgone by not working rises as the wage rate rises. The substitution effect is the tendency for individuals to allocate their time away from leisure and toward more work when the wage rate rises-that is, substitute work time for leisure time. In contrast, the income effect assumes that leisure is like most goods: As income rises, individuals want to purchase more leisure. Thus, an individual can be expected to allocate a portion of the increase in potential income associated with a hike in wage rates to acquiring more leisure; this allocation can offset part or all of the substitution effect associated with the higher wage or can more than offset it. In summary, with an increase in wage rates, the substitution effect reduces desired leisure and increases the amount of labor services an individual wants to offer to the labor market, whereas the income effect increases desired leisure and decreases labor services.
The presence of both effects creates ambiguity in predicting individuals' overall labor supply response to changes in labor market conditions. Although economic theory cannot predict whether the income or the substitution effect will dominate an individual's decision to supply labor, empirical evidence can be brought to bear on this question. The relative importance of these income and substitution effects can vary, depending upon the time frame and the demographic groups considered, and can change over time.
LABOR SUPPLY AND THE BUSINESS CYCLE
Typically, labor force participation rises during an expansionary period and falls during a recession. This relation occurs because of the effects of fluctuations in expected real wages on labor supply decisions. Although the level of real wages does not change much over the business cycle, the probability of finding a job changes substantially. Consequently, during a recession, the expected real wage falls for those without jobs. For some workers, the expected payoff from looking for work is so low that they decide that spending time at home is more productive than spending time in searching for a job-that is, in the terminology of the model of the allocation of time presented in the preceding section, the substitution effect dominates. This result is referred to as the discouraged worker effect. However, if labor supply decisions of family members are interdependent, the need to maintain living standards during a recession can draw other workers into the labor force despite the low expected real wage. For example, if one family member loses his or her job during a recession, the family may decide that another family member should enter the labor market. This response is referred to as the addedworker effect.
Of course, the added-worker and the discouraged-worker effects can coexist because the added workers and discouraged workers will be different groups of people. Which group dominates is an empirical issue. Research has indicated that in the aggregate the discouraged-worker effect dominates the added-worker effect: More households are affected by the decline in expected real wages during a recession than are induced to enter the labor force by the layoff of a family member.3 In addition, the increased attachment of women to the labor force has diminished the potential importance of the added-worker effect as the pool of nonparticipating women has shrunk. Thus, the number of discouraged workers clearly increases during economic downturns and declines during expansions; or stated another way, the rate of labor force par - LABOR SUPPLY OVER THE LONGER RUN
These income and substitution effects are important over longer periods as well. Broadly speaking, hourly wage rates tend to be relatively low when young people first enter the work force and are learning new skills. Wage rates rise until they peak, typically when workers are around forty-five to fifty-five years of age, and then they decline. The so-called lifecycle model of labor supply predicts that the labor supply of individual households will change as the members of the household age.4 In particular, workers are assumed to participate most heavily during the period of their life cycle when their expected wages are highest and to substitute leisure for work when expected wages are low.
The rates of men's participation in the labor force over the past three decades appear to be consistent with this view, as they rise from men's teenage years into middle age and then fall off in their later years (chart 4). The rates of labor force participation for women in 1960 and 1970 were broadly consistent with the life-cycle model, although participation among women of childbearing age tended to drop off sharply. By 1990, however, the profile of labor force participation rates for women had essentially the same shape as that for men.
As previously discussed, over the past few decades movements in labor supply for men and women have diverged: The participation of women in the labor force has risen sharply, while the participation of men has trended down. The life-cycle model alone does not fully explain these long-term trends in labor force participation. Other economic forces that underlie these developments include demographic changes in the labor force as well as shifts in social customs and tastes that have influenced changes in relative income, fertility patterns, attachment to the labor force, and retirement decisions.
Relative Income Effects
One way that demographic changes can affect participation decisions is through their influence on the relative incomes of various age groups.1 Over the past three decades, relative income effects have been most pronounced for younger workers. The income of year-round, full-time workers aged fifteen to twenty-four years (the best available proxy for the wage rate), relative to that of men aged forty-five to fifty-four years. began to decline sharply in the mid-1960s and continued falling through the 1970s and into the 1980s chart 6). The decline coincided with the influx of young workers into the labor market (chart 7). Relative income flattened out for young men and women in the second half of the 1980s, as the small current generation of youth the baby-bust generation) replaced the baby-boom generation, although reduced demand for lower-skilled workers apparently prevented a reversal of the earlier decline.
The changes in the relative wage levels for young men and women affected the labor supply decisions of individual young people in two ways: first, through changes in their own relative wage and, second, through changes in the relative wages of their spouses or other family members. For young women, econometric evidence suggests that the effect of the declines in their own relative wages, which ordinarily would have reduced labor supply, was more than offset by the positive influence on participation of the potential drop in family living standards associated with the decline in the relative wage of young men.6 In contrast, the econometric evidence suggests that these effects are about offsetting for the participation decisions of young men.
Fertility and Home Responsibilities
Women's decisions to participate in the labor force and to have children are not independent. For example, a woman's decision to increase the size of her family may be affected by the level of her family's income and by the wage she expects to receive by entering the labor market. At the same time, a woman s decision to participate in the labor force may be affected by whether she has young children. In the 1960s and the first half of the 1970s, the birth rate fell sharply (chart 8). For the next ten years, birth rates flattened out and then in the late 1980s began to increase.
The lower birth rates probably were closely related to the sharp rise during the 1970s in the participation rate of women twenty-five to thirty-four years of age. Of course, the importance of the birth rate to levels of labor force participation is influenced by changes in the availability, cost, and quality of child care and the propensity for mothers of young children to work outside the home. Currently, the proportion of mothers of young children who are in the labor market is at a historic high. This finding suggests that childrearing and young women's participation in the labor force are more compatible than in the past.
The trend toward increased labor force participation by women is mirrored in the data on individuals who are out of the labor force. Data on the trends in nonparticipation have been tabulated since the mid-1970s (tables I and 2). Between the late 1970s and the 1980s, the downward trend for younger women in the category of keeping house apparently slowed. These data are consistent with the small increase in the birth rate shown in chart 8. For adult women, increasing labor force participation continued to shrink this category of non-participation well into the 1980s.
Retirement An important element of the slow decline in labor force participation rates for men is a trend, beginning as early as the 1930s, toward earlier retirement.7 Between 1960 and 1990, labor force participation rates for men fifty-five and over fell substantially; and despite an increase in participation among women twenty to fifty-four years of age, participation rates for women fifty-five and over dropped to rates similar to those for men (chart 4). Data on nonparticipation also show the trend toward early retirement for men between the late 1970s and 1980s table 2).
Participation rates for older men reached historically low levels in the mid-1980s and since then have changed little. Factors contributing to the decline in participation include wider coverage of the labor force by social security, the extension of old age assistance to persons sixty-two to sixty-four years old, the greater availability f social security and other types of disability payments, the greater prevalence of pension plans, the provision of early retirement benefits in many pension plans, and efforts by employers to cut back payrolls by inducing early retirement. These factors dramatically lowered the cost, in terms of forgone earnings, of retirement. In addition, as lifetime incomes rose, individuals substituted away from work to leisure in their later years. In recent years, the enactment of laws prohibiting mandatory retirement may have been a factor working to stem the decline in participation among older men.
The cause of increased retirement among women is somewhat different table 2). From 1960 to 1990, among women in their preretirement years (forty-five to fifty-four years of age) labor force participation increased markedly (chart 4). Consequently, in 1990 a larger proportion of women fifty-five years of age and older were eligible for retirement than had been eligible in earlier years. Nevertheless, the finding that the rate of labor force participation for these women has remained fairly steady since 1960 suggests that factors favoring retirement have influenced women as well as men.
Labor Force Attachment
The rise in participation rates among women is also due to a greater attachment to the labor force of women who are currently working. If all labor force participants stayed in the labor force for the entire year, the annual average labor force participation rate would be a direct measure of annual labor force experience. However, some individuals will be out of the labor force for part of the year. Consequently, to study the effect of changing patterns of work experience on labor force participation, annual participation rates are divided into two components: the proportion of the population that worked or looked for work at some time during the year and the mean number of weeks those individuals spent in the labor force.
The labor force participation rate provides a snapshot of labor force attachment, and data on the cumulative experience of individuals in the labor market over an entire year provide a more comprehensive view of changing trends in labor force activity. The Work Experience Surveys conducted with the March Current Population Survey summarize the cumulative employment and unemployment experience of respondents for the preceding year.8 These data can shed light on the extent to which changes in the labor force participation rate reflect changes in the number of participants in the labor market each year as opposed to changes in the number of weeks during the year workers participate in labor market activity.
The two panels of chart 9 show the trends in the two components of labor force participation between 1958 and 1989, the last year for which the work experience data have been tabulated. The average number of weeks worked annually by men who were in the labor force at any time during the year has drifted up slightly over the thirty years to just under fifty weeks. This finding reflects the fact that most men with labor market experience work at full-time, year-round jobs or, if unemployed at some point in the year, that they remain active in searching for work. However, the proportion of men with labor force experience during the year has shrunk slowly over time because of the trend toward early retirement.
For women, the average number of weeks worked has steadily increased from thirty-nine weeks in 1958 to more than forty-seven weeks in 1989. This finding highlights the extent to which, over time, women have developed firmer attachments to the labor market. This trend has occurred as women's exits from jobs, or from a search for a job, to nonparticipation each year have become less frequent. Indeed, by the late 1980s, the average amount of time women spent in the labor force had moved to within three weeks of the average for men. At the same time, the proportion of women with some labor market activity during the year has trended higher, this trend further boosting the participation rate of females in the labor force.
The relative contributions of the two components to the overall rise in the female labor force participation rate have shifted over time. Between 1958 and 1964, all of the increase in the participation rate reflected growth in the proportion of women with labor market experience. Over the subsequent decade or so, however, the rise in the number of weeks spent working or looking for work each year and the rise in the proportion of women with labor market experience contributed in equal proportions to the sharp acceleration in the participation rate for women. The growing attachment of women to the labor force, as measured by the average weeks of experience, continued into the mids, 1980even though the rise in incidence of women with labor market experience slowed.
Although many jobs place constraints on the number of hours per week worked by employees, employees can, in effect, exercise some choice over their hours of work by choosing their type of employment. Those who choose part-time schedules because they do not want full-time work or are unavailable for full-time work are considered voluntary part-time workers. Those individuals who work short hours because of slack work, the inability to find a full-time job, or for other reasons related to the demand for their labor are considered involuntary part-time workers.
Part-time employment has made up a growing share of all jobs over the past four decades. The fraction of employed workers who voluntarily work part time rose from 13 percent in the late 1960s to 14 percent in the 1970s and accounted for the trend in total part-time work over this period (table 3). The increase occurred partly because employers sought to accommodate the preferences for short hours of students and housewives, the fastest growing groups in the labor force in the 1970s. However, voluntary part-time work has remained a fairly constant proportion of total employment since the 1970s. Instead, involuntary part-time employment has propelled the upward trend in total part-time work since the early 1970s, largely because employers view part-time work as a means to cut labor costs, and not because workers want shorter schedules.9 With a sizable proportion of the workforce holding part-time jobs, it would seem that the aggregate amount of labor supplied could be increased by lengthening the workweek of those working part time. However, many workers hold more than one part-time job and thus essentially work full-time schedules.
Indeed, a final way that employees can exercise choice over their hours of employment is to work at more than one job. Between 1970 and the middle of the 1980s, the percentage of workers holding two or more jobs showed no apparent trend. with the proportion fluctuating around 5 percent (table 4). Historically, most holders of multiple jobs were men working a second job to supplement income from a primary, full-time job. However, while the aggregate multiple-job-holding rate held fairly steady over this period, the rate for men moved down from 7 percent in 1970 to around 6 percent by the mid-1970s, where it stayed through the middle of the 1980s. At the same time, the rate for women moved up from 2 1/4 percent in 1970 to 4 3/4 percent in 1985.
Between mid-1985 and mid-1989, the number of persons holding more than one job increased nearly 1 1/2 million, and the proportion of multiple-job holders climbed substantially from 5 1/2 percent in 1985 to 6 1/4 percent in 1989. Women accounted for nearly two-thirds of the increase in multiple-job holding over this period. Multiple job holding for men continues to be a way of supplementing income from a full-time job with work after hours. In contrast, nearly one-third of 4. Percentage of workers holding more than one job. 1970-89(1) the female multiple-job holders work at more than one part-time job; they work, on average, nearly fifty hours per week.
All told, more women are participating in the labor force than at any time in the past four decades, and those women who are participating work more hours per week and more weeks per year. In many ways, women's labor supply decisions are more and more resembling those of men.
THE SLOWDOWN IN 1990: PERMANENT OR TRANSITORY?
The overall rise in labor force participation over the past three decades has not been smooth. In numerous episodes, participation stopped increasing for a time or even declined slightly. The steady upward movement in labor force participation rates after 1982 stopped suddenly in 1990. In some respects, a flattening in the participation rate seems a natural reaction to the recent slowing in the growth of employment opportunities. Over the four quarters of 1990, employment, as measured by the current population survey, was down about 270,000-the first decline in eight years. Also during 1990, the labor force participation rate declined, and the civilian labor force increased only 1/2 percent-the smallest annual gain in almost thirty years. Apparently, the halt in labor force growth in 1990 reflected both cyclical and secular developments. In response to slowing economic growth, the number of discouraged workers rose, although the number remains well below that seen early in the 1980s (chart 5 and table 2). The pickup in nonparticipation in 1990 also reflected increased school attendance for the younger age groups table 1). If jobs are scarce, schooling is a natural alternative for the young. But rising school enrollment Fates may reflect the increasing returns to higher education over the 1980s as well.(10)
The larger number of young nonparticipants in 1990 who reported school as their major activity and as their reason for nonparticipation is mirrored in the school enrollment rate for sixteen to twenty-four year olds (enrollees as a proportion of the population of sixteen to twenty-four year olds), which has increased steadily in recent years (chart 10). Through 1989, the distribution of school enrollment by employment status had remained fairly constant. Between 1989 and 1990, the proportion of school enrollees who reported themselves as out of the labor force (the bottom slice of chart 10) jumped from 23 percent to 25 percent. In other words, full-time students now make up a larger share of all school enrollees.
For adult women, the increase in labor force participation continued at the expense of the number keeping house-at least through the end of the 1980s. However, some reversal may have occurred in 1990. That is, the number of women leaving housework to enter the labor force may not be dropping as fast as it had previously. This reversal may reflect, in part, the recent increases in the birth rate (chart 8). As fertility has increased, the availability of child care has become an important factor in decisions about labor force participation, especially for women. Also, as the population ages, the care of elderly parents and spouses may become a limiting factor on labor supply decisions of individuals. Indeed, a moderate decline among adult men in total participation in 1990 relative to the mid-1980s is attributed to more men reporting that they are keeping house. This decline may reflect that a growing number of men are caring for young children and ailing spouses.(11)
Taken together, these data do not unambiguously point to the increase in nonparticipation as being either a permanent or temporary phenomenon. If, for example, the increase in the number of young women who are staying out of the labor force because of home or family responsibilities persists, the 1990 slowdown in participation may last for a number of years. However, the rise in the number of discouraged workers, albeit moderate, combined with several other factors suggests that some of the increase in nonparticipation may be transitory or cyclical in nature. In particular, the substantial increase in school enrollment has occurred among nonparticipants who are devoting themselves full time to education, rather than among participants who are dividing their time between schooling and the labor force. These students may be choosing more intensive schooling as an alternative to what they perceive to be a poor labor market. In any case, this increase in school enrollment is not a negative development for the long-term growth of the economy. These young people are investing in their human capital and are likely to move into the labor force when their schooling is completed. Also, greater schooling at this time may result in stronger labor force attachment later in their careers. Consequently, their increase in nonparticipation is not necessarily indicative of a permanent, secular decline in labor force growth. It appears rather to be a transitory phenomenon that should be reversed over the medium term.
THE LONGER-TERM OUTLOOK
The gap in work experience patterns of men and women has significant implications for the outlook for further increases in labor force participation. The difference in the proportions of men and women with some labor market experience during the year has shrunk considerably over the past thirty years, but it remains sizable: In 1958, about 50 percent of all women aged sixteen years and over worked or looked for work at some time during the year, compared with more than 90 percent of men aged sixteen years and over. By the mid-1970s, the proportions were 57 percent and 84 percent respectively; by 1989, they were 64 percent and 81 percent. Thus, considerable room exists for more women to enter the work force, although home and family responsibilities may be a constraint on how high the proportion of women will go.
Similarly, the average number of weeks of labor market experience for men which has been relatively stable at approximately fifty weeks per year) may be viewed as an effective ceiling for women. Under this assumption, closing the remaining gap for women has the potential of adding only another 3 percentage points to their labor force participation rate. How fast that gap may disappear is unclear; between 1986 and 1988, when labor market conditions were generally favorable, the average weeks of experience for women were little changed. if no further increase in weeks of work experience were to occur, the upward trend in the labor force participation rate for adult women would be reduced by about 1112 percentage points per year. Alternatively. if that gap were to continue to close at the same rate as over the 1976-88 period and if the proportion of women participating in the labor force were to continue to rise at the same rate as during the 1980s, growth in the female labor force participation rate would remain at the 1980-90 average of 1112 percentage points per year for another decade. Thereafter, however, the contribution of growing labor force attachment to the rise in the labor force participation rate for women would be exhausted.
As indicated in the preceding section, trends in labor force participation show no clear signs of a sharp structural break, despite the drop in late 1990. Nevertheless, the increase in the rate of participation for women is likely to slow gradually as it nears the rate of participation now experienced by men. As formal child cafe has become more affordable and more accessible, the effect of fertility on labor force decisions has been muted. Expanded access to child care may weaken this relationship further.
Past experience suggests that the participation rates of young people are likely to be affected by movements in their relative incomes. The relative income of young people is likely to rise for the next several years as the baby-bust cohort continues to move into the labor force. This higher relative income may place less pressure on young families to have dual careers. Although a greater emphasis on educational attainment may depress labor force participation rates for the next several years, it could boost long-term labor force attachment by increasing the expected wages of school enrollees over their lifetimes.
The data in this article derive from the Current Population Survey (CPS), a monthly survey that the Bureau of the Census conducts for the Bureau of Labor Statistics. The CPS, collected from a probability sample of approximately 60,000 households, provides statistics on the employment status of the civilian population.
The term civilian labor force refers to all persons sixteen years of age and older who are either working for pay or profit or looking for work. Unemployed refers to persons who are not working but are available and looking for work. Thus the labor force consists of die employed and the unemployed. The labor force participation rate represents members of the civilian labor force as a percent of the civilian noninstitutional population of persons sixteen years of age and older, or the working-age population. The unemployment rate represents the number of unemployed as a perpcentage of the civilian labor force.
All civilians sixteen years of age and older who are not classified as employed or unemployed are defined as not in the labor force. These persons are classified further, the classification depending upon two sets of questions asked of all respondents to the CPS. The first set (see table 1) asks respondents about their major activity during the survey period. Possible answers include "in school," keeping house," unable to work because of long-term illness," or other.' The second set of questions (see table 2) asks persons not in the labor force the major reason for their lack of participation. Respondents are split into those who would like a job now and those who would not. These questions, tabulated separately for men and women, provide a richer classification of respondents.
1. See the box at the end of the text for definitions of many terms used throughout this article.
The intensity of work effort and the education and training of the work force will help determine how productively the available supply of hours will be translated into the output of goods and services. These issues are beyond the scope of this article.
2. The terms work and leisure are used to distinguish between work for pay or profit or looking for such work) and all other activities, only some of which would be what is usually considered leisure.
3. See. for example, Jacob Mincer, "Labor Force Participation of Married Women: A Study of Labor Supply," Aspects of Labor Economics Princeton University Press for the National Bureau of Economic Research. 1962), pp. 63-97; and Shelly Lundberg, "The Added Worker Effect," Journal of Labor Economics, vol. 3 (January 1985), pp. 11-37.
4. For an illustration of the importance of intertemporal substitution in labor supply see Gilbert Ghez and Gary Becker. Allocation of Time and Goods over the Life Cycle Columbia University Press for the National Bureau of Economic Research, 1975); and James Heckman and Thomas MaCurdy. "A Life Cycle Model of Female Labour Supply." Review of Economic Studies. vol. 47 (January 1980). pp. 47-74.
5. For it more complete discussion of demographic effects on relative income. seel William Wascher. Susan Burch, and John Goodman, Jr., "Economic Implications of Changing Population Trends. " Federal Reserve Bulletin (December 1986), pp. 815-26; and Michael L. Wachter, "Intermediate Swings in Labor Force Participation," Brookings Papers on Economic Activity, vol. 1 (1977), pp. 545-76.
6. See Wachter, "Intermediate Swings in Labor Force Participation" and Robert S. Gay and William Wascher, "Persistence Effects in Labor Force Participation," Eastern Economic Journal, vol. 15 October-December 1989), pp. 177-87.
7. See Donald O. Parsons. The Decline in Male Labor Force Participation," Journal of Political Economy vol. 88 (1980), pp. 117-34: Laurence J. Kotlikoff and David Wise, The Wage Carrot and the Pension Stick W. E. Upjohn Institute for Employment Research, 1989); and Gary S. Fields and Olivia Mitchell, Retirement, Pensions, and Social Security (MIT Press, 1985).
8. The Current Population Survey is also the source of data on labor force participation rates and the unemployment rate.
9. See Chris Tilly. "Reasons for the Continuing Growth of Part-time Employment," Monthly, Labor Revievs, (March 1991), pp. 10-18.
10. See Lawrence Katz and Kevin Murphy, "Changes in Relative Wages, 1963-87: Supply and Demand Factors," unpublished manuscript. April 1990.
11. The distinction between discouraged workers and other nonparticipants is fuzzy given that the classification of respondents is based on self-reporting. For example, an individual who is truly discouraged may choose to report housekeeping as the reason for nonparticipation if he or she views discouragement as an admission of failure.
|Printer friendly Cite/link Email Feedback|
|Publication:||Federal Reserve Bulletin|
|Date:||Jun 1, 1991|
|Previous Article:||Record of policy actions of the Federal Open Market Committee.|
|Next Article:||Bank holding company investments for community development.|