Emily Beller and Michael Hout examine trends in U.S. social mobility, especially as it relates to the degree to which a person's income or occupation depends on his or her parents' background and to the independent contribution of economic growth. They also compare U.S. social mobility with that in other countries. They conclude that slower economic growth since 1975 and the concentration of that growth among the wealthy have slowed the pace of U.S. social mobility.
In measuring mobility, economists tend to look at income and sociologists, occupation. The consensus among those measuring occupational mobility is that the average correlation between the occupations of fathers and sons today ranges from 0.30 to 0.40, meaning that most variation in the ranking of occupations is independent of social origins. Those measuring income mobility tend to agree that the elasticity between fathers' and sons' earnings in the United States today is about 0.4, meaning that 40 percent of the difference in incomes between families in the parents' generation also shows up in differences in incomes in the sons' generation.
Beller and Hout show that occupational mobility increased during the 1970s, compared with the 1940s-1960s, but there is some evidence to suggest that by the 1980s and 1990s it had declined to past levels. Existing data on income mobility show no clear trends over time, but increases in economic inequality during the 1980s made mobility more consequential by making economic differences between families persist for a longer time.
In international comparisons, the United States occupies a middle ground in occupational mobility but ranks lower in income mobility. Researchers have used the variation in mobility to study whether aspects of a country's policy regime, such as the educational or social welfare systems, might be driving these results. There is as yet, however, no scholarly consensus about the sources of cross-national differences in mobility.
Most Americans think it unfair when things they cannot control limit their chances to succeed in life. Particularly un-American is the notion that circumstances of birth set life on a course that may be hard to alter through one's own efforts. So, rags-to-riches stories are popular, and crowds cheer for the underdog. Academic research on social mobility goes beyond the stories and the drama to quantify the link between circumstances of birth and economic success, both for the population as a whole and for important and interesting groups within it. Sociologists and economists put numbers to patterns by comparing the social and economic success of Americans with an absolute standard that is completely free of traces of birth and with a relative standard that is based on recent experience or the current experience of other countries. On the absolute standard, Americans' occupations and incomes are tied much more closely to their parents' occupations and incomes than they would be in a world where circumstances of birth were irrelevant for adult success. (1) On the relative scale, ties between people's current occupations and incomes and those of their parents are about what they have been over the past twenty-five years, but substantially weaker than they were in the early 1960s. (2)
Social mobility from one generation to the next is the difference between a person's current income, wealth, or occupation and that of the family that raised her. (3) An opportunity structure promotes social mobility if it allows people to escape poverty while limiting the degree to which those who grow up in privileged homes get advantages throughout their lives. Growth promotes mobility, too, by raising everyone, regardless of background, above the level of that background. In this article we will focus most on the opportunity structure because scholars have written more about it. But it is important to keep in mind how important growth can be. Nearly everyone who grew up in the Great Depression experienced substantial upward mobility in adulthood. It was not that America was more equitable when the children of the Great Depression grew up than it was before or has been since; it was that the nation recovered from its economic collapse and therefore most people were much better off. Social mobility should not be confused with inequality, which refers to differences among people in wealth, income, and occupational status at any point in time. Social mobility would not matter in a society in which there was no inequality. Parents would have no advantages to bequeath to their children, and no one would care where they ended up. But when inequality is great, social mobility matters a lot. The advantages of rising to the top are large, and the consequences of remaining stuck at the bottom are much more serious.
Social mobility is high if the opportunity structure is open--that is, if the barriers and advantages associated with a person's background are few. But openness of that sort is not the only way to spur intergenerational mobility. Mobility is also high if growth is strong and widespread enough to make everyone better off. The opportunity structure, in the form of barriers and advantages, is symmetrical in the sense that in the absence of growth, removing a barrier that might block a person who starts low also implies removing an advantage from a person who starts high. Growth, on the other hand, can--in President John F. Kennedy's famous phrase--lift all boats. If growth is widespread, it creates new opportunities that can lift a person who starts low without knocking down a person who starts high. But slow growth reduces social mobility, as does a closed opportunity structure. (4)
Growing inequality does not necessarily increase or decrease the prevalence of social mobility, but it does increase the difference between the upwardly mobile and the downwardly mobile. When inequality increases, extreme incomes, occupations, and amounts of wealth (high and low) become more prevalent, and fewer people occupy the middle of the distribution. So an upwardly mobile person has farther to rise and a downwardly mobile person has farther to fall in a more unequal society. Also (and this is a little less intuitive) an increase in inequality over a person's lifetime increases the probability that someone who starts life in extreme privilege will stay there and (simultaneously) increases the probability that someone whose parents were poor will also be poor. Those increases in immobility are offset, though, by a decrease in the probability that someone whose parents were about average will end up near the average (because rising inequality eliminates positions near the average). The increased immobility at the extremes and mobility in the center do not imply a stronger or weaker correlation between circumstances of birth and adulthood; they follow from the definition of inequality--more extreme outcomes, fewer average ones.
It is possible to talk about social mobility in general terms, but most researchers focus on one of five specific forms of mobility: educational mobility, occupational mobility, wage mobility, family income mobility, and wealth mobility. Each has its own interesting properties. We focus on two types: family income mobility and occupational mobility. The first--typically the domain of economists--is the extent to which an adult's (or family's) relative income or rank in the income distribution is similar to his or her father's (or father's family's) relative income or rank. The second--most often the province of sociologists--is the extent to which the status or type of job a person winds up with resembles that of his or her father or mother.
We review research on income and occupational mobility, examining changes in the opportunity structure and growth, as well as the effects of inequality. We first try to quantify the extent of intergenerational occupational and income mobility in the United States. We then compare estimates of mobility in the United States today with evidence both from the American past and from cross-national comparisons. Where possible, we discuss the intergenerational persistence of wealth and property as well. Intergenerational educational mobility is another fascinating topic, but it is beyond our scope in this review.
Measuring Intergenerational Social Mobility
Important differences in the concepts of occupational and income mobility can help to explain how it is possible that mobility in one domain might be greater than mobility in another. People's incomes vary significantly even if their jobs share the same occupational category. Analyses of occupational mobility and analyses of income mobility provide different pictures of people's prospects, because they ask different questions. Intergenerational persistence in occupational status is not a good proxy for persistence in income, and vice versa; a person who is upwardly mobile occupationally does not necessarily enjoy a higher relative income than his or her parents (and vice versa). (5)
In addition, analysts investigating occupational and income mobility face different limitations and use different methodologies. On the one hand, occupation is easier to measure than income because people remember their parents' occupations reliably and with a high degree of accuracy, whereas dollar amounts are much harder to recall, and most people plainly do not know their parents' incomes. Inflation erodes the value of the dollar over time, too, further complicating the task of evaluating parents' incomes, even if they are known. On the other hand, occupations can be hard to rank, whereas income is straightforwardly scored in dollars (or the relevant local currency). In addition, researchers interested in occupational mobility often want to measure the component of mobility that is independent of growth, whereas income mobility researchers do not typically distinguish between the two.
Researchers interested in occupational mobility must first come to grips with the problem of how to rank occupations, getting beyond the qualitative detail of specific job descriptions to arrive at useful categories or scores. Some solve the problem by grouping occupations into relatively large classes. Others rank them on a scale from 0 to 100. (6) In the first approach, researchers gather occupations into several broad classes, such as professionals (for example, doctors and lawyers), skilled trade workers (for example, electricians and carpenters), or the self-employed, and then create a matrix that allows them to compare each person's occupation with his or her father's occupation. While this approach reveals details of which occupations are linked across the generations and which are not, its results are hard to summarize unless the categories are clearly ranked.
Ranking allows the straightforward estimation of an overall intergenerational correlation between the ranking of a person's occupation and that of his or her father. A correlation of 0 implies that a person's occupational rank is completely independent of that of his or her parents, and therefore that there is perfect mobility between ranks across generations. A correlation of 1 implies that ranks do not change from generation to generation. The correlation that a researcher calculates for a real society places that society on the continuum from perfect mobility to complete rigidity.
In principle, one could use an intergenerational income correlation to measure income mobility as well as occupational mobility, but in practice researchers (usually economists) typically measure income mobility slightly differently. They look at the strength (persistence) of the relationship between parents' and children's income in percentage terms; that is, they ask how much (what percentage) of the income difference between families in one generation persists into the next generation. This estimate is called the intergenerational elasticity. If the elasticity is 0.4, for example, they would conclude that a 10 percent difference in parents' income would lead to a 4 percent difference in offspring's incomes. The advantage of using the intergenerational elasticity, from the researcher's point of view, is that it can capture the amplifying effects on mobility of rising income inequality, or the dampening effects of falling income inequality (the formula for the intergenerational correlation discards this useful information). On the low extreme, an elasticity of 0 describes a society in which family economic background is not at all related to adult income, whereas an elasticity of 1 describes a society in which each person ends up in exactly the same economic position as her or his parents (just like the correlation). But unlike the correlation, the elasticity is unbounded, so one could, in principle, discover that two people who started life in families 10 percent apart ended up 15 percent apart (if the elasticity was 1.5). Mobility is the complement of the elasticity--a low intergenerational elasticity translates to a high mobility rate, and a high elasticity translates to a low mobility rate.
Social Mobility in the Contemporary United States
Having defined our terms and introduced some of the analytical distinctions that researchers use, we turn now to the heart of the matter: how much mobility Americans have experienced from their youth till now. We discuss occupational mobility first, and then turn to income and wealth mobility.
Intergenerational Occupational Mobility
One way to assess occupational mobility in the United States is to categorize occupations into a few classes and then measure the extent of class immobility, downward mobility, and upward mobility between generations. Using this technique, we analyzed nationally representative data on men and women born after 1950. (7) We distinguished six general occupational categories in descending order: upper professional or manager, lower professional or clerical, self-employed, technical or skilled trade, farm, and unskilled and service workers. (8) Among men, 32 percent were immobile (their occupation was in the same category as their father's), 37 percent were upwardly mobile, and 32 percent were downwardly mobile. Fifteen percent of the mobility was driven by structural change in the economy, or economic growth--more professional jobs and fewer farm jobs were available to sons than to their fathers; that also accounts for why upward mobility was more common than downward mobility. Women's mobility patterns reflect the gender segregation of the labor force, as well as opportunity and growth. Among women, 27 percent were immobile, 46 percent were upwardly mobile, and 28 percent were downwardly mobile. Most Americans regard sales and clerical jobs as better than most blue-collar jobs, so the millions of blue-collar men's daughters who work in stores and offices are upwardly mobile (just not very much). That particular type of short-range upward mobility accounts for the fact that more American women than men are upwardly mobile.
Table 1 shows the data for men from which the above estimates were generated. It shows the outflow of sons from each class background category to current occupational categories (in percentages). The bold diagonal entries show the percentage of men from each class background who stay where they began; this "stickiness" is greatest for the most and least advantaged class background categories. If we consider the column percentages instead (that is, the share in each class from each background category [data not shown]), it is striking that the proportion of immobile incumbents is almost always higher than the proportion drawn from any other class category. The most extreme example is that 66 percent of men in the farm class came from a farm background.
Another way to assess occupational persistence is to examine intergenerational occupational correlations. As noted, these correlations differ depending on which characteristic of occupations is the focus of research. For example, the intergenerational correlation of the prestige of fathers' and sons' occupations is lower than the correlation of the education level associated with their occupations. (9) One of the most commonly used scales for measuring occupations is the socioeconomic status index (SEI), which provides a rank for each occupation. Average intergenerational father-son correlations in the SEI and similar indexes are in the neighborhood of 0.35 to 0.45, implying that some 12 to 21 percent of the variation in sons' occupations can be accounted for by fathers' occupations. (10) The larger estimates are mostly from the early 1960s; the smaller ones are from the 1980s and 1990s. (11) For the men in table 1 we calculate the correlation to be 0.32. (12)
Assessing whether a given intergenerational correlation or mobility rate reflects a low or high degree of occupational mobility requires determining an appropriate reference for comparison. Complete mobility is neither plausible nor, arguably, desirable, given that some of the factors leading to the intergenerational persistence of social position, such as cognitive ability or work effort, seem acceptable--that is, fair. (13) Complete immobility is also implausible. In the absence of accepted definitions of what constitutes low or high mobility, one strategy is to contrast the U.S. estimates with those from a range of comparable countries. Comparisons with other industrialized countries (to which we turn later) support the prevailing idea that occupational mobility in the United States is reasonably high, as does the finding that U.S. occupational persistence does not extend past two generations. (14)
But one complication in analyzing occupational mobility using either SEI correlations or class mobility tables such as table 1 is that there is no straightforward way to incorporate two parents' occupations into the intergenerational correlations or class background categories. Thus occupational mobility research is limited, for the most part, to studies of father-child (or household head-child) occupational persistence.
The case of income mobility, to which we turn next, is instructive: intergenerational associations appear to be weaker when calculations do not include both parents' earnings and other sources of family income. Of course, occupational statuses do not add together the way incomes do, so we use multivariate regression to calculate the total association between family background and occupational status. For the men in table 1 we find the multiple correlation is 0.38.
Intergenerational Income Mobility
The current consensus among researchers is that intergenerational persistence, or elasticity, between fathers' and sons' earnings in the United States lies at about 0.4 on the 0-1 scale described above. (15) The persistence between total childhood family income and adult sons' family income or personal earnings is even greater, in the range of 0.54 to 0.6. (16) An elasticity of 0.54 means that, for example, a 10 percent difference between two families' incomes is associated with a 5.4 percent difference in their sons' earnings. The corresponding elasticity between family income and daughters' earnings is lower, at 0.43. When analysts focus on married women, the elasticity between total childhood family income and adult daughters' total family income is 0.39. The same elasticity for married sons is 0.58. These gender-specific patterns occur because men contribute about 70 percent of family income, on average, and because there is an association between childhood family income and spouses' income. (17)
The conclusion that the intergenerational elasticity between father's and son's earnings in the United States is as high as 0.4 was reached only recently, and these estimates may understate the true income persistence, as more recent research has tended to raise estimates of the elasticity. Early estimates placed the father-son earnings elasticity at 0.2 or lower--indicating substantially more economic mobility than an estimate of 0.4 would imply. (18)
The upward trend in estimates reflects methodological improvements, probably not real-life trends. In the 1970s researchers had to estimate the size of intergenerational elasticities from one year of data about fathers and one year of data about sons. The newer, higher estimates accumulate income over five or more years for both fathers and sons. (19) Another improvement has been the recognition that a person's age affects his or her earnings. Calculations based on young people's earnings understate the persistence that is seen when we observe people during their top-earning years. (20) New, logically similar corrections are resulting in a further increase in the estimated elasticity to 0.6. (21) We have doubts about this higher estimate for father-son earnings persistence. The theory behind accumulating data is that each family or person has a "true" income level but minor ups and downs (and measurement errors) produce variations around the true value that lower the elasticity. In the short run, this theory is credible. Over longer and longer spans, it becomes harder to believe that there is just one true value.
Elasticities are good indicators of a society's average level of intergenerational economic persistence, but they do not provide much information about mobility patterns. Mobility matrices that give the probability of children's economic position conditional on fathers' or family position provide a more detailed picture of intergenerational mobility. Similar to the pattern of occupational mobility shown in table 1, the income mobility matrix in table 2 shows that economic immobility is highest among children whose family incomes fall in the top or the bottom quartiles of the earnings distribution. (22) This pattern is consistent with other U.S. economic mobility matrices, which show the greatest rigidity at the extremes of the distribution. (23)
That overall mobility rates are higher in the middle of the income distribution does not necessarily mean that the impact of family income is weaker in the middle than it is at the top and bottom of the distribution--by definition, people at the bottom of the distribution can experience only upward mobility, and the reverse is true at the top of the distribution. People in the middle have the prospect of moving either up or down.
Besides looking at descriptive income mobility matrixes, another way that researchers can learn more about mobility patterns than the average intergenerational elasticity can provide is to calculate separate estimates for people who start life at low, middle, and high points on the income distribution for their parents' generation. Some evidence suggests that the effect of childhood family income on adult income is stronger at the high end of the father's earnings distribution than at the low end. (24)
A different question is how the effect of family background differs along the son's earnings distribution rather than that of the father. Such analyses suggest that father's income is more persistent among sons with low earnings than among sons with high earnings. (25) This implies that opportunity for upward mobility is more equal than the opportunity for downward mobility--presumably, advantaged parents are able to protect their children from downward mobility, but children from more disadvantaged backgrounds do have a greater chance of upward mobility than the intergenerational elasticity (which, as noted, describes the average level of mobility) would suggest.
Intergenerational Wealth Mobility
Finally, how does wealth mobility compare to occupation and income mobility? First, there is substantial intergenerational persistence in family wealth; the correlation is in the neighborhood of 0.50. (26) Wealth is important because its distribution is far more unequal than the distribution of family income and because it seems to have greater effects on other aspects of family well-being, especially homeownership and investment in children's education. (27)
The disparity in wealth not only persists between the generations, it mushrooms. Without a cushion of inherited wealth, emergencies hit harder; and people who have no nest egg have to let opportunities pass by. Because of a wealth deficit, African Americans are more vulnerable to shocks and less able to capitalize on breaks than whites with the same income, so the next generation will inherit less too. (28) The wealth gap will not close any time soon; wealthy people's assets grow at a rate that approximates that of the New York Stock Exchange. (29) Furthermore, inherited wealth can put families in better neighborhoods and school districts than they could afford if they had to rely exclusively on their incomes.
At the very top of the wealth distribution, innovations in computer and telecommunications technologies created new fortunes in the 1980s and 1990s and pushed new people to the top of lists like the Fortune 400. As interesting as the extremely high tail of the wealth distribution is, however, those 400 wealthy people are not, by definition, representative of their 300 million fellow citizens. Thus most analyses of wealth mobility focus on the wealth differentials in representative samples of American families and households. (30) Wealth mobility in the United States resembles occupational and income mobility in a few key respects.
A number of familiar features show up in the wealth mobility matrix in table 3. First, in each row, the main diagonal entries are the largest, indicating the relative strength of persistence and mobility. They are somewhat higher than comparable figures for family income. Second, the richest and poorest quintiles are less mobile than the middle groups, as was true for income mobility.
Contemporary and Past U.S. Mobility Rates
The best way to evaluate contemporary mobility is to compare it with the past. Strong evidence from several approaches shows that barriers to mobility weakened substantially for American men from the 1960s through the mid-1980s, thus increasing opportunity. Trends since that time are not yet clear, since the children born from the late 1980s onward have not yet entered the labor force. However, assuming there has been no change in mobility patterns, the rise in income inequality over the last few decades means that economic differences between families will persist for a longer time.
Barriers to mobility weakened substantially for American men during the 1960s; the intergenerational correlation in occupations fell from roughly 0.4 to 0.3 on the scale from 0 to 1. (31) These trends continued for men until the mid-1980s; the correlation declined another 30 percent in ten years. (32) Data sets for women are available only from the 1970s; they show that women's and men's intergenerational occupational mobility differed because the occupational distributions differed, but that the occupational mobility rate, adjusted for differences in distribution, did not differ for men and women. Trends for women resembled those for men as well. (33)
One major reason for the declining correlation between fathers' and sons' occupations was the climbing share of men with college degrees--occupational opportunity for college graduates is quite high, though of course the likelihood of college graduation is itself highly dependent on class background. (34) State interventions on behalf of disadvantaged groups may also have contributed to the increased mobility. (35)
By itself, the declining effect of fathers' occupation should have increased overall occupational mobility in the United States between 1972 and 1985, but for these later years the increase in opportunity was counteracted by a slowdown in economic growth. (Conversely, the growth slowdown would have resulted in less social mobility in the United States were it not for the opposite trend in opportunity.) The combined effect of increased opportunity and slower growth kept overall mobility relatively high through 1985. Some evidence suggests that the trend toward greater opportunity slowed or reversed for men born after 1970, but data limits prevent firm conclusions at this time. (36)
Historically, over the course of the twentieth century economic growth produced more upward than downward movement. Mobility that results from a more open opportunity structure--a decreased advantage to upper-status background--leads, however, to both downward and upward movement. (37) So the changes in the American mobility pattern since the early 1970s have resulted in more downward mobility, especially for the offspring of the most privileged classes, and somewhat less upward mobility. Table 4 shows our calculations of the amount and direction of men's occupational mobility; that is, the share of men upwardly mobile, downwardly mobile, or immobile by year of birth. The earliest cohort (born in the 1930s) first entered the labor force in the 1950s and reached its top earning potential around 1980; the latest cohort (born in the 1970s) first entered the labor force in the 1990s and will reach its top earning potential around 2020. Almost half of the cohort born in the 1930s was upwardly mobile; only one-fourth of that cohort was downwardly mobile. Since then, fewer men have been upwardly mobile and more have been downwardly mobile. Among men born in the 1960s and 1970s, downward mobility is almost as prevalent as upward mobility. Immobility rose across cohorts from one-fourth to one-third.
Less is known about change over time in intergenerational income mobility. Some research has suggested a decline in the intergenerational elasticity among recent cohorts--indicating higher mobility--but the decline is not statistically significant (that is, it might appear by chance because of insufficient data), and using alternate data generates the conflicting finding that the elasticity may have increased. (38) The most convincing finding is that there has been no change over the past century in intergenerational income mobility. (39)
Because the data for the analysis of trends over time in income mobility are limited, it is possible that trends did change but that data could not detect them. One way around that difficulty is to study change over time in how a set of family background indicators broader than father's earnings or family income may affect adult income. One such study used parent income, parent education, parent occupation, family race and ethnicity, family structure, number of siblings, and region to investigate how background affected economic outcomes over time. It found that the effect of family background on men's economic outcomes declined during the 1960s, then remained constant during the 1970s, 1980s, and 1990s. Although the effect of family background remained constant over those three decades, the economic gap between advantaged and disadvantaged men increased because economic inequality increased during this period. On the other hand, the effect of family background on women's outcomes declined between the 1970s and 1990s. Coupled with the increased economic inequality, this meant that the gaps in women's outcomes remained constant over the period. (40)
A yet broader approach investigates the joint impact of state residence, ancestry, and family income on men's economic outcomes and finds a down-up cycle. The intergenerational correlation between social origins and adult incomes was fairly constant from 1940 until 1960, fell substantially in 1970s (indicating increased mobility), and then returned to previous levels in the 1980s and 1990s. The intergenerational elasticity, on the other hand, declined between 1940 and 1980 but increased during the 1980s and 1990s. The measures are different because the elasticity is sensitive to income inequality, which followed the same trend over time--declining from 1940 to 1980 and then rising during the next two decades. The trends in the intergenerational correlation considered together with the elasticity suggest that income mobility was not abnormally low in the 1980s. But because economic inequality increased, the consequences of a historically normal degree of mobility were greater, and a greater share of the economic differences between families could persist for a longer time. (41)
U.S. Social Mobility Rates in International Comparison
Direct comparisons of intergenerational social mobility in different countries are difficult to make, because both data availability and research methodologies differ from country to country. Until recently it has been hard to compare occupational mobility in the United States with that in other countries because of differences in occupational coding, but new research using comparable coding shows that the United States is at the median in terms of opportunity: lower than the most open nations, such as Sweden, Canada, and Norway; but higher than the more rigid nations, such as West Germany, Ireland, or Portugal. (42) Other research suggests that Italy, France, and Great Britain are among the other societies that now display the lowest comparative mobility rates. (43)
Recent research attributes these international differences in occupational mobility to two dimensions of educational inequality--the share of adults who attend college and equality of educational opportunity (the strength of the effect of family background on educational attainment). (44) Opportunity is much greater among college-educated adults of different class backgrounds than it is among adults with less education. The United States has one of the highest levels of college attendance, but also a relatively low level of equality in overall educational opportunity.
Although the United States occupies a middle ground in international comparisons of occupational mobility, its ranking in terms of income mobility is lower. Both the United States and Great Britain have significantly less economic mobility than Canada, Finland, Sweden, Norway, and possibly Germany; and the United States may be a less economically mobile society than Great Britain. (45) Much of the higher intergenerational elasticity in the United States is due to greater income immobility at the top and bottom of the earnings distribution; the mobility of middle earners looks more similar to that in the other countries. (46)
Two explanations for these international differences in income mobility appear particularly compelling. First, it seems plausible that high income inequality at a given time could cause a high intergenerational persistence of economic status. The United States and Great Britain have high income inequality coupled with low income mobility, whereas Scandinavian countries display the opposite pattern. Canada, however, casts doubt on this explanation, because it has relatively high income inequality coupled with high income mobility.
Second, given the limited ability of low-income parents to invest in their children's education, it is possible that progressive public policies toward education financing could explain why some countries have higher rates of economic mobility. Research shows that differences in education financing alone do not explain mobility differences between countries, but education financing is an important part of the explanation, together with other factors that differ between countries, such as the earnings return to education (how much another year of education increases one's earnings) and the heritability (either genetic or environmental) of income-predictive traits. (47)
Higher economic returns to education and lower levels of public financing of education decrease intergenerational mobility because when income depends on education, children from low-income families need to go to college to be upwardly mobile. But with less public financing of education, fewer low-income children can go to college. Both factors also increase income inequality at a given time, because lower public financing of education lowers equality of educational opportunity, while higher returns to education increase the earnings gap between more and less educated people. These patterns may explain why most countries either have low income inequality and high income mobility or high income inequality and low income mobility.
The economic returns to education are higher in the United States than they are in other countries, which may explain the stronger intergenerational income persistence. The role of heritability also implies that differing degrees of assortative marriage in a country--differing rates of couples from similar economic backgrounds marrying--will affect intergenerational mobility. Marital sorting increases intergenerational inequality. (48)
Consequences and Policy Discussion
The research literature--and by necessity our review of it--focuses on the way the economy affects mobility. To us, though, that leaves unexplored the most profound changes affecting families and their potential to promote or hinder their children's prospects: the way family structure itself affects both income and occupational mobility. Sophisticated mobility studies came of age in the era when most people grew up in a relatively stable family structure, anchored by the earnings of a paternal breadwinner. Tying the circumstances of birth to the income of the family breadwinner greatly simplifies the task of quantifying social mobility. And as long as that was an appropriate simplification, researchers made significant progress.
But changes in family structure since the 1970s have contributed to growing economic inequality. Two-earner families have significantly higher standards of living than single-parent families. (49) At least part of the connection between parents' incomes and the success of their adult children is presumably due to the disrupting effects of family breakup. To be sure, researchers have considered family structure in important papers and books over the past forty years. (50) But so far they have not been able to take fully into account what they know about family structure when they measure social origins. This gap in the research reduces our confidence in current estimates of social mobility. In particular, it appears that a father's absence from his family can reduce the correlation between his occupation and the success of his children. He is more able to pass on the advantages of his accomplishments if he lives with his family. (51) Until this issue gets sorted out, it will be hard to say what family policy is most appropriate for promoting social mobility.
What we can say is that greater opportunity and increased growth promoted social mobility during the 1960s and 1970s. The importance of socioeconomic background for adult success declined during those decades, while economic growth further boosted all job seekers and earners. Research has tied the declining importance of socioeconomic background to better educational opportunities and equal opportunity legislation and its enforcement. (52) Each seems to be a potential tool for leveling differences in the American opportunity structure. Institutions need to compensate for the ways that family differences lead to differences in achievement--a point made by James Coleman twenty years ago. (53) Educational opportunity promotes social mobility not only by distributing human capital in many ways that are independent of social origins, but also by loosening the ties between occupational and income origins and destinations among college graduates. (54) Establishing norms of fairness and enforcing them seems like a particularly sixties-style idealist solution. But establishing and enforcing those norms during the 1960s improved the life chances of disadvantaged people during the 1960s and 1970s and could, in principle, do so again.
(1.) We will cite many specific studies in the course of this review. For overviews, consider the reviews by Gary Solon, "Cross Country Differences in Intergenerational Earnings Mobility," Journal of Economic Perspectives 16, no. 3 (2002): 59-66; and Michael Hout, "How Inequality May Affect Intergenerational Mobility," in Social Inequality, edited by Kathryn M. Neckerman (New York: Russell Sage Foundation, 2004), pp. 969-87.
(2.) David L. Featherman and Robert M. Hauser, Opportunity and Change (New York: Academic Press, 1978); David J. Harding and others, "The Changing Effects of Family Background on the Incomes of American Adults," in Unequal Chances: Family Background and Economic Success, edited by Samuel Bowles, Herbert Gintis, and Melissa Osborne Groves (New York: Russell Sage Foundation and Princeton University Press, 2005), pp. 100-44; Michael Hout, "More Universalism, Less Structural Mobility," American Journal of Sociology 93, no. 6 (1988): 1358-400.
(3.) The term social mobility can also refer to intragenerational mobility--the changes in a person's income level or occupational status during his or her adult life course--but our focus is intergenerational.
(4.) Hout, "How Inequality May Affect Intergenerational Mobility" (see note 1).
(5.) Robert M. Hauser, "Intergenerational Economic Mobility in the United States: Measures, Differentials, and Trends," Working Paper 98-12 (University of Wisconsin-Madison Center for Demography and Ecology, 1998); Daniel P. McMurrer and Isabel Sawhill, Getting Ahead: Economic and Social Mobility in America (Washington: Urban Institute Press, 1998), p. 49.
(6.) Otis D. Duncan. "A Socioeconomic Index for All Occupations," in Occupations and Social Status, edited by Albert J. Reiss (New York: Free Press, 1961), pp. 109-38.
(7.) General Social Surveys, 1972-2004 (cumulative file).
(8.) We substituted mothers' occupations if the father was not part of the household.
(9.) Robert M. Hauser and John Robert Warren, "Socioeconomic Indexes for Occupations: A Review, Update, and Critique," Sociological Methodology 27 (1997): 177-298.
(11.) Peter M. Blau and Otis Dudley Duncan, The American Occupational Structure (New York: Wiley, 1967); Thomas A. DiPrete and David B. Grusky, "Structure and Trend in the Process of Stratification for American Men and Women," American Journal of Sociology 1 (1990): 107-43.
(12.) The regression coefficient is also 0.32.
(13.) For example, since cognitive abilities are inherited to some degree (through genetics and environment), perfect mobility would imply no link between ability and outcome. See Harding and others, "The Changing Effects of Family Background" (see note 2); John E. Roemer, "Equal Opportunity and Intergenerational Mobility: Going Beyond Intergenerational Income Transition Matrices," in Generational Income Mobility in North America and Europe, edited by Miles Corak (Cambridge University Press, 2004), pp. 48-57.
(14.) John Robert Warren and Robert M. Hauser, "Social Stratification across Three Generations: Evidence from the Wisconsin Longitudinal Study," American Sociological Review 62, no. 4 (1997): 561-72.
(15.) Most of the recent estimates pinpoint the father-son earnings elasticity between 0.35 and 0.5; see Gary Solon, "Intergenerational Mobility in the Labor Market," in Handbook of Labor Economics, vol. 3A, edited by Orley Ashenfelter and David Card (Amsterdam: North Holland, 1999), pp. 1761-800; Gary Solon, "Intergenerational Income Mobility in the United States," American Economic Review 82, no. 3 (1992): 393-408; David J. Zimmerman, "Regression toward Mediocrity in Economic Stature," American Economic Review 82, no. 3 (1992): 409-29. These estimates exclude non-full-time workers. Estimates of persistence that include all sons with incomes are lower; see Joseph G. Altonji and Thomas A. Dunn, "Relationships between the Family Incomes and Labor Market Outcomes of Relatives," in Research in Labor Economics, edited by Ronald G. Ehrenberg (Greenwich, Conn.: JAI Press, 1991): 269-310; Nathan D. Grawe, "Intergenerational Mobility for Whom? The Experience of High and Low Earning Sons in International Perspective," in Generational Income Mobility in North America and Europe, edited by Corak, pp. 58-89 (see note 13); H. Elizabeth Peters, "Patterns of Intergenerational Mobility in Income and Earnings," Review of Economics and Statistics 74, no. 3 (1992): 456-66.
(16.) Laura Chadwick and Gary Solon, "Intergenerational Income Mobility among Daughters," American Economic Review 92, no. 1 (2002): 335-44; Bhaskhar Mazumder, "The Apple Falls Even Closer to the Tree than We Thought: New and Revised Estimates of the Intergenerational Inheritance of Earnings," in Unequal Chances, edited by Bowles, Gintis, and Groves, pp. 80-99 (see note 2); Peters, "Patterns of Intergenerational Mobility in Income and Earnings" (see note 15).
(17.) Chadwick and Solon, "Intergenerational Income Mobility among Daughters" (see note 16).
(18.) Gary Becker and Nigel Tomes, "Human Capital and the Rise and Fall of Families," Journal of Labor Economics 4, no 3, pt. 2: "The Family and the Distribution of Economic Rewards" (1986): S1-39.
(19.) Gary Solon, "Biases in the Estimation of Intergenerational Earnings Correlations," Review of Economics and Statistics 71, no. 1 (1989): 172-74; Solon, "Intergenerational Income Mobility in the United States," (see note 15); Zimmerman, "Regression toward Mediocrity in Economic Stature" (see note 15).
(20.) Steven Haider and Gary Solon, "Life Cycle Variation in the Association between Current and Lifetime Earnings," Working Paper 11943 (Cambridge, Mass.: National Bureau of Economic Research, 2006).
(21.) Mazumder, "The Apple Falls Even Closer to the Tree" (see note 16).
(22.) For example, lower overall mobility on the low end of the income distribution may be due to factors such as the lower ability of poor parents to invest in children's education, the debilitating effects of living in poor neighborhoods, or the employment effects of high rates of incarceration.
(23.) Grawe, "Intergenerational Mobility for Whom?" (see note 15); Markus Jantti and others, "American Exceptionalism in a New Light: A Comparison of Intergenerational Earnings Mobility in the Nordic Countries, the United Kingdom and the United States," Discussion Paper 1938 (Bonn, Germany: IZA [Institute for the Study of Labor], 2006); Mazumder, "The Apple Falls Even Closer to the Tree" (see note 16).
(24.) Kenneth A. Couch and Dean R. Lillard, "Non-Linear Patterns of Intergenerational Mobility in Germany and the United States," in Generational Income Mobility in North America and Europe, edited by Corak, pp. 190-206 (see note 13).
(25.) Eric R. Eide and Mark H. Showalter, "Factors Affecting the Transmission of Earnings across Generations: A Quantile Regression Approach," Journal of Human Resources 34, no. 2 (1999): 253-67; Grawe, "Inter generational Mobility for Whom?" (see note 15). There is some conflicting evidence on this issue when a different data source is used, but this may be due to a problem with some of the data collection.
(26.) Casey B. Mulligan, "Galton versus the Human Capital Approach to Inheritance," Journal of Political Economy 107, no. 6, pt. 2: "Symposium on the Economic Analysis of Social Behavior in Honor of Gary S. Becker" (1999): S184-224.
(28.) Thomas M. Shapiro, The Hidden Cost of Being African American (Oxford University Press, 2004), pp. 60-85.
(29.) Lester C. Thurow, Generating Inequality: The Distributional Mechanisms of the Economy (Springfield, Va.: National Technical Information Service, 1975).
(30.) Dalton Conley, Being Black, Living in the Red: Race, Wealth, and Social Policy in America (University of California Press, 1999); Lisa A. Keister, Getting Rich: America's New Rich and How They Got That Way (Cambridge University Press, 2005); Melvin Oliver and Thomas M. Shapiro, Black Wealth/White Wealth: A New Perspective on Racial Inequality (New York: Routledge, 1995).
(31.) Featherman and Hauser, Opportunity and Change (see note 2).
(32.) DiPrete and Grusky, "Structure and Trend in the Process of Stratification" (see note 11); Hout, "More Universalism, Less Structural Mobility" (see note 2).
(33.) There is limited information on trends before this time, and firm conclusions on occupational mobility after the mid-1980s are hard to come by because of data limitations. Timothy J. Biblarz, Vern L. Bengston, and Alexander Bucur, "Social Mobility across Three Generations," Journal of Marriage and the Family 58, no. 1 (1996): 188-200; DiPrete and Grusky, "Structure and Trend in the Process of Stratification" (see note 11); Michal Hout, "Status, Autonomy and Training in Occupational Mobility," American Journal of Sociology 89, no. 6 (1984): 1379-409; Hout, "More Universalism, Less Structural Mobility" (see note 2).
(34.) Claude S. Fischer and others, Inequality by Design: Cracking the Bell Curve Myth (Princeton University Press, 1996), pp. 152-55; Hout, "Status, Autonomy and Training in Occupational Mobility" (see note 33).
(35.) DiPrete and Grusky, "Structure and Trend in the Process of Stratification" (see note 11).
(36.) Emily Beller and Michael Hout, "Income Inequality and Intergenerational Mobility: Change across Cohorts," unpublished Working Paper (University of California, Berkeley, 2005). The extension of the measurement of occupational class background to include mothers' as well as fathers' occupation will also change the estimated trends in mobility rates over time, because the impact of including mothers' occupation varies depending on the period (since, for example, mothers' labor force participation and parents' marital sorting by class also varies by period).
(37.) Michael E. Sobel, Michael Hout, and Otis D. Duncan, "Exchange, Structure and Symmetry in Occupational Mobility," American Journal of Sociology 91, no. 2 (1985): 359-72.
(38.) David I. Levine and Bhashkar Mazumder, "Choosing the Right Parents: Changes in the Intergenerational Transmission of Inequality between 1980 and the Early 1990s," Working Paper WP-02-08 (Federal Reserve Bank of Chicago, 2002); Susan E. Mayer and Leonard M. Lopoo, "What Do Trends in the Intergenerational Economic Mobility of Sons and Daughters in the United States Mean?" in Generational Income Mobility in North America and Europe, edited by Corak, pp. 90-121 (see note 13).
(39.) Jo Blandon, Paul Gregg, and Stephen Machin, "Intergenerational Mobility in Europe and North America," A Report Supported by the Sutton Trust (London: LSE Centre for Economic Performance, 2005); Chul-In Lee and Gary Solon, "Trends in Intergenerational Income Mobility," Working Paper 12007 (Cambridge, Mass.: National Bureau of Economic Research, 2006).
(40.) Harding and others, "The Changing Effects of Family Background" (see note 2).
(41.) Daniel Aaronson and Bhashkar Mazumder, "Intergenerational Economic Mobility in the United States: 1940-2000," Working Paper 05-12 (Federal Reserve Bank of Chicago, 2005).
(42.) Emily Beller and Michael Hout, "Welfare States and Social Mobility," Research in Social Stratification and Mobility (forthcoming).
(43.) Richard Breen and Jan O. Jonsson, "Inequality of Opportunity in Comparative Perspective: Recent Research on Educational Attainment and Social Mobility," Annual Review of Sociology 31 (2005): 223-43; Richard Breen, ed., Social Mobility in Europe (Oxford University Press, 2004), pp. 37-76.
(44.) Beller and Hout, "Welfare States and Social Mobility" (see note 42); Richard Breen and Ruud Luijkx, "Conclusions," in Social Mobility in Europe, edited by Breen, pp. 383-410; Michael Hout, "Maximally Maintained Inequality Revisited: Irish Educational Mobility in Comparative Perspective," in Changing Ireland, 1989-2003, edited by Maire Nic Ghiolla Phadraig and Betty Hilliard (University College Dublin Press, forthcoming).
(45.) Anders Bjorklund and Markus Jantti, "Intergenerational Income Mobility in Sweden Compared with the United States," American Economic Review 87, no. 5 (1997): 1009-18; Blandon, Gregg, and Machine, "Intergenerational Mobility in Europe and North America"(see note 39); Miles Corak, "Generational Income Mobility in North America and Europe: An Introduction," in Generational Income Mobility in North America and Europe, edited by Corak, pp. 1-37 (see note 13); Jantti and others, "American Exceptionalism in a New Light" (see note 23); Solon, "Cross Country Differences in Intergenerational Earnings Mobility" (see note 1).
(46.) Jantti and others, "American Exceptionalism in a New Light" (see note 23).
(47.) Casey B. Mulligan and Song Han, "Human Capital, Heterogeneity, and Estimated Degrees of Intergenerational Mobility," Working Paper W7678 (Cambridge, Mass.: National Bureau of Economic Research, 2000); Solon, "Cross Country Differences in Intergenerational Earnings Mobility" (see note 1).
(48.) Raquel Fernandez and Richard Rogerson, "Sorting and Long-Run Inequality," Working Paper W7508 (Cambridge, Mass.: National Bureau of Economic Research, 2000).
(49.) Sara McLanahan and Gary Sandefur, Growing Up with a Single Parent: What Hurts, What Helps (Harvard University Press, 1994).
(50.) Timothy J. Biblarz and Adrian E. Raftery, "Family Structure, Educational Attainment, and Socioeconomic Success: Rethinking the Pathology of Matriarchy," American Journal of Sociology 105, no. 3 (1999): 321-65; Beverly Duncan and Otis Dudley Duncan, "Family Stability and Occupational Success," Social Problems 16, no. 3 (1969): 273-85; Sara McLanahan and Larry Bumpass, "Intergenerational Consequences of Family Disruption," American Journal of Sociology 94, no. 1 (1988): 130-52; Kelly Musick and Robert D. Mare, "Family Structure, Intergenerational Mobility, and the Reproduction of Poverty: Evidence for Increasing Polarization?" Demography 41, no. 4 (2004): 629-48.
(51.) Susan E. Mayer, What Money Can't Buy: Family Income and Children's Life Chances (Harvard University Press, 1997).
(52.) DiPrete and Grusky, "Structure and Trend in the Process of Stratification" (see note 11); Hout, "More Universalism, Less Structural Mobility" (see note 2).
(53.) James S. Coleman, "Social Capital in the Creation of Human Capital," American Journal of Sociology 94, supplement (1988): S95-120.
(54.) Blau and Duncan, The American Occupational Structure, chap. 5 (see note 11); Hout, "More Universalism, Less Structural Mobility" (see note 2).
Emily Beller is a Ph.D. candidate in sociology at the University of California, Berkeley. Michael Hout is a professor of sociology and demography at the University of California, Berkeley.
Table 1. Intergenerational Occupational Mobility of Men Born
between 1950 and 1979
Destination: son's occupation
Origin: Upper professional Self-
father's occupation professional and clerical employed
Upper professional 42 24 7
and clerical 29 27 7
Self-employed 29 18 16
Technical and skilled 17 19 6
Farm sector 14 11 8
Unskilled and service 16 17 6
Destination: son's occupation
Origin: Technical Farm
father's occupation and skilled sector
Upper professional 12 0
and clerical 17 0
Self-employed 19 0
Technical and skilled 30 1
Farm sector 17 13
Unskilled and service 22 1
Destination: son's occupation
father's occupation and service Total
Upper professional 15 100
and clerical 20 100
Self-employed 18 100
Technical and skilled 26 100
Farm sector 37 100
Unskilled and service 38 100
Source: General Social Surveys, 1988-2004.
Table 2. Intergenerational Income
Mobility: Probability of Son's Quartile
Given Parent's Quartile
Destination: son's income quartile
income quartile First Second Third Fourth
First 42 28 19 12
Second 26 29 27 19
Third 18 24 29 29
Fourth 15 18 25 40
Source: Adapted from H. Elizabeth Peters, "Patterns of
Intergenerational Mobility in Income and Earnings,"
Review of Economics and Statistics 74, no. 3 (1992): 460.
Table 3. Intergenerational Wealth Mobility, 1979-2000
Origin Poorest Second Third Fourth Richest Total
Poorest 45 27 11 9 9 100
Second 24 35 20 14 7 100
Third 11 20 35 21 13 100
Fourth 7 11 23 33 25 100
Richest 5 6 9 25 55 100
Source: Lisa A. Keister, Getting Rich: America's New Rich
and How They Got That Way (Cambridge University Press, 2005),
Table 4. Amount and Direction of Men's
Occupational Mobility, by Year of Birth
Year of Upwardly Downwardly
birth Immobile mobile mobile Total
1930-39 26 49 25 100
1940-49 28 45 27 100
1950-59 31 39 30 100
1960-69 31 35 33 100
1970-79 33 35 32 100
Source: General Social Surveys, 1988-2004.