Working-Age, but Not Working, 1960 to 2025
This report is an update to an earlier report on the same topic. Like our prior analysis, this report shows that the share of U.S.-born men not in the labor force — neither working nor looking for work — has roughly returned to pre-pandemic levels, but the current rate remains near a record high relative to prior economic expansions. Being out of the labor force is associated with profound social problems such as crime, overdose deaths, and welfare dependency. All of this is relevant to the immigration debate because many argue for high levels of legal immigration, or even tolerating illegal immigration, on the grounds that there are not enough workers. However, policymakers should consider encouraging work among the millions of non-institutionalized, working-age men on the economic sidelines rather than ignoring this problem and bringing in ever more immigrants.
Among those without a bachelor’s:
The primary focus of this analysis is the increase in U.S.-born men not in the labor force, though we also report figures for women and immigrants as well. Much of the prior research has not divided the population by nativity. We use the term “immigrant” in this report synonymously with “foreign-born”. This includes all those who were not U.S.-citizens at birth, including illegal immigrants.2 Our analysis is only for those of working-age, which we define in various ways, but we always exclude children under age 16 and those 65 and older. In contrast, the Bureau of Labor Statistics (BLS) reports labor force figures for the population 16 and older in its monthly employment situation reports. Including the elderly can make sense depending on the research question, but doing so conflates the deterioration in participation among working-age men with the overall aging of American society. Like other research, we find a significant increase in the share of U.S.-born men not in the labor force, particularly those without a bachelor’s degree.
Prior ResearchNumerous studies have established that there has been a decline in the share of less-educated men in the labor force, though the fact that the problem is concentrated among U.S.-born men is often not acknowledged. The issue has been extensively studied by academics, the Brookings Institution, the Obama White House, the Federal Reserve, and many others. Probably the best book on the subject is Nicholas Eberstadt’s Men Without Work. Those in the labor force are working (full- and part-time) or looking for work; those outside the labor force are neither working nor looking for work and are not counted as unemployed because they are not actively seeking a job.3 Consistent with other research, our analysis shows that there has been a very large increase in the share and number of men, particularly U.S.-born men without a four-year college degree, not in the labor force since the 1960s. Drawing more of these men into the labor force represents a potential alternative source of workers rather than continuing to bring in large numbers of immigrants.
We focus on peak years of the business cycle starting in 1960. To do this, we use decennial census data, along with CPS data. Looking back six decades, we find that the increase in the share of U.S.-born men not in the labor force is most pronounced among those without a bachelor’s degree. Like other researchers, we exclude those in institutions (primarily prisons and jails) from our analysis. We focus on men in our discussion because, unlike women, men have not been heavily impacted by the profound social, cultural, and legal changes that caused an ever-larger share of women to join the labor force in the last 60 years. Of course, we also report figures for women.
Methods and DataWe use historical census data because it allows us to look at labor force participation by nativity, which is not really possible with the Current Population Survey (CPS) before 1994.4 However, the decennial census only provides information every 10 years and it no longer asks about place of birth or citizenship. While census years do not perfectly correspond to peak years of labor force participation between recessions, which is typically the best way to look at the long-term labor market trends, 1960, 1970, 1980, and 1990 are not far off from the peak years.5 Further, the decennial census is really the only nativity data available going back decades.
We use the 1960 through 1990 decennial censuses and from 2000 onward we use the April Current Population Survey (CPS) to examine labor force participation in the peak years of 2000, 2006, 2019, and 2025.6 To supplement this data, we also report statistics for every year going back to 1962 using the Current Population Survey Annual Social and Economic Supplement (CPS ASEC), though statistics by nativity are only possible from 1994 onward. The CPS ASEC is an expanded version of the monthly CPS and is done annually. The survey oversamples minorities and also includes many additional questions; it is one of the primary sources of information about family structure, poverty, income, and health insurance coverage. The CPS ASEC is also somewhat larger than the monthly CPS, making it ideal to measure immigrants.
Why Participation in the Labor Force MattersLabor Force Participation in the Context of Immigration. The cacophony of voices calling for ever more immigration because there are not enough workers has included numerous academics, business groups, advocacy groups, think tanks, politicians, and a very large number of opinion writers in recent years. Immigration advocates typically point to low unemployment, the aging of society, or more technical measures of the number of unfilled jobs to make the case that the supply of workers is inadequate and more immigration is needed. What is perhaps most striking about virtually all of this advocacy and commentary is that the deterioration in labor force participation among the U.S.-born is seldom, if ever, mentioned. This occurs despite the fact that the increase in men not in the labor force is well documented and the number of working-age people currently not in the labor force is in the tens of millions. Any discussion of the need for more foreign workers should be conducted with an understanding of the enormous number of working-age people not working. This is especially the case because there is a clear consensus that the decline in labor force participation has detrimental effects.
The list of negative outcomes for individuals, their families, and society associated with being out of the labor force is so long that it can barely be summarized here. The issue has been extensively studied by academics, the Brookings Institution, the Obama White House, the Federal Reserve, and many others. Probably the best book on the subject is Nicholas Eberstadt’s Men Without Work. There simply is no question that having a large share of working-age men not in the labor force contributes to a long list of profound social problems.
Impacts at the Individual Level. It should be obvious that being in the labor force significantly increases income and dramatically reduces the risk of being in poverty.7 There is also good evidence that not working is associated with a deterioration in mental health. In addition, being out of the labor force is linked with declining physical well-being. Research shows that not working has a negative effect on the incidence of obesity and the associated health problems this creates. Perhaps most important, there seems to be a strong association between not working and so-called “deaths of despair”, including suicide, drug overdose, and destructive levels of alcohol consumption and death from poisoning.
Societal Impacts. Research by the Federal Reserve shows that low labor force participation reduces economic growth. Not surprisingly, both the Congressional Budget Office and the Obama White House pointed out that deterioration in participation in the labor force has negative fiscal implications, as fewer people pay taxes and more make use of public benefits.8 Men not in the labor force also make relatively unattractive marriage partners, so non-participation in the labor force hinders family formation. Additionally, more than one study has found that not being employed is associated with social isolation. In contrast, being in the labor force leads to greater political participation and civic engagement. There are a significant number of studies showing a link between not working and crime. The bottom line is that not working is bad not only for those out of the labor market – it has significant negative implications for their families and the broader American society.
FindingsIn the analysis that follows, we discuss the results shown in the figures, which summarize our findings. However, we also provide an Excel spreadsheet that reports more detailed labor force statistics by nativity, age, gender, and education level in the peak (or near-peak) years of 1960, 1970, 1980, and 1990 based on the census and the April CPS for 2000, 2006, 2019, and 2025. The census and April CPS data are the sources for Figures 1 through 9 in this report. The decennial census and the April CPS are the sources for the detailed information in the Excel file as well. We also use the CPS ASEC to report some additional information for every year in Figures 10 to 13 for U.S.-born and immigrant men from 1962 to 2025.9 We focused on working-age natives without a bachelor’s degree for simplicity, rather than the various education levels that comprise this group; however the Excel sheet does report more detailed educational breakdowns. We often refer to those without a bachelor’s degree as the “less-educated” in this report. The Excel spreadsheets included in this report and our prior research show that the sub-groups of U.S.-born men without a bachelor’s such as those with only a high school education, have seen a similar deterioration in participation in the labor force.
Share of U.S.-Born Men Not in the Labor Force. Figure 1 reports the share of working-age, U.S.-born men not in the labor force, with working-age defined broadly as 16 to 64. Though the share has returned to roughly pre-Covid levels in 2019, the figure still shows that the share not in the labor force has increased for nearly six decades for U.S.-born men and remains at a near-record high for an economic expansion. Overall, the share of men born in the United States not working or looking for work is 10.4 percentage points higher in 2025 than in 1960. Even compared to 2000, it is 4.8 percentage points higher. For U.S.-born men without a bachelor’s degree, the rise is an enormous 15.5 percentage points from 1960 to 2025. Since 2000, the increase for those without a bachelor’s degree is 7.4 percentage points. All of these changes are very large because they represent millions of people on the sidelines of the economy. Figure 1 also shows some increase in the share of men with at least a bachelor’s degree not in the labor force, though not so much recently.
 |
Looking at Different Age Groups. One might imagine that the increase in men not in the labor force is only due to American teenagers working less. The increase in the number of Americans attending school, particularly college, may also explain the rise. Or perhaps the results reflect early retirement. But this is not what has happened. Figure 2 excludes teenagers and shows that the deterioration from 1960 to 2025 is still 14.3 percentage points for U.S.-born men 20 to 64 without a bachelor’s degree. It is true that a larger share of Americans are in school, particularly those under age 25, today than in 1960 or even 2000. Of course, students then and now can and often still do work. Even part-time work a few hours a week for pay still counts as being in the labor force. But even if we restrict the analysis to only those 25 to 54, what economists often call the “prime-age” for work, we still find the same pattern.
 |
Figure 3 shows that in 1960 the share of U.S.-born men (25 to 54) not in the labor force was only 4 percent. This was roughly true for all educational levels. U.S.-born men were almost all working regardless of their education level. But this has changed significantly since then. At 11.2 percent overall in 2025, the share of all prime-age, U.S.-born men not in the labor force is more than 2.5 times larger than in 1960. The increase for those without a bachelor’s (10.6 percentage points) is very large. The decline in less-educated, prime-age U.S.-born men not in the labor force makes clear that the increase in American men on the economic sidelines cannot be attributed to teenage males or college-age males not working, nor can early retirement explain it. Even when those age groups are excluded, the decline is decades long; and the share today not in the labor force remains extremely high by historical standards, though it has returned to the level it was before Covid in 2019.
 |
The Possibility of a “Creaming” Effect. The growth in the share not in the labor force for the less-educated could be caused by the so called “creaming” effect. That is, the most academically oriented and employable part of the population now has college degrees, something that relatively fewer people had in the past. In 1960, 10 percent of men 20 to 64 had a bachelor’s, by 2025 it was 35 percent; and among prime-age men it increased from 12 percent to 39 percent. So perhaps when we only look at those without a bachelor’s over time, we are really looking at the increasingly less employable part of the population. However, Figures 1 through 3 show that the share not in the labor force for the entire U.S.-born male working-age population, regardless of education, is much higher now than in the past, though it is slightly better than in 2019 before Covid. There can be no creaming when all working-age people are considered. Further, although not as pronounced, those with a college degree also show some increase in not being in the labor force. So while the very pronounced increase in the share not in the labor force among less-educated men specifically might be partly explained by creaming, it certainly cannot explain the overall increase. Perhaps most important, whether the creaming phenomenon exists or not, it does not mitigate all the negative effects on society caused by the increase in those not participating in the labor force.
It is worth adding that the significant increase in the overall share of working-age, U.S.-born men with a college degree over the decades should have significantly reduced the overall share not in the labor force as more educated people tend to be much less likely to be out of the labor force, but the opposite happened. While increasing the education level of the U.S. population might be a good idea for any number of reasons, based on the last six decades it is unlikely to solve the high rate of U.S.-born, working-age men neither working nor looking for work.
Number of U.S.-Born Men Not in the Labor Force. Figure 4 shows the total U.S.-born working-age (16 to 64) male population from 1960 to 2025 relative to the increase in the number in the labor force. Because labor force growth did not keep pace with population growth, the number of U.S.-born men out of the labor force increased significantly. In April 2025, 18.4 million U.S.-born men 16 to 64 were not in the labor force, up from 12.7 million in 2000 and 5.3 million in 1960.10 Figure 5 shows the number of men who would have been in the labor force if the participation rate had remained at the 1960 level or the 2000 level. If the same share of U.S.-born men (16 to 64) were in the labor force in April 2025 as in 1960, there would have been 75.4 million U.S.-born men in the labor force — 8.9 million more than was actually the case. If the share in the labor force only returned to the 2000 rate, it would result in 70.6 million U.S.-born men in the labor force, which is still 4.1 million more than were actually working or looking for work.11 Figure 6 shows that the growth in the number of men not in the labor force is substantial even if the age range is restricted.
 |
 |
 |
Increase in Immigrant Men in the Labor Force. The number of foreign-born men in the labor force (16 to 64) grew by 14.6 million from 1960 to 2025. As already noted, over this same period the number of working-age, U.S.-born men not in the labor force increased by 13.1 million. It would be a mistake to assume that every job taken by an immigrant represents a loss by the U.S.-born, but it would also be a mistake to assume that increasing the supply of workers so dramatically has no impact on the labor market outcomes of the U.S.-born.
The Situation for U.S.-Born Women. Figure 7 shows the share of U.S.-born women ages 16 to 64 not in the labor force from 1960 to 2025. The revolution in social norms and legal protections for women is clearly evident in the figure, which shows a dramatic decrease in the share not in the labor force since 1960. In 2000, it reached a low of 27.6 percent not in the labor force. Although the 30.5 percent in 2019 is somewhat higher than the 28.9 percent in April 2025, it is still higher than the 27.6 percent in 2000. (Using other age ranges shows the same pattern; see tables in Excel sheets.) If the same share of U.S.-born women (16 to 64) were in the labor force in April 2025 as in April 2000, it would add 1.1 million more women to the labor force.
 |
The percentage of U.S.-born women with a bachelor’s in 2025 not in the labor force is at a record low. However, the same is not true for their less-educated counterparts. At 36.8 percent, the share of U.S.-born women without a bachelor’s out of the labor force is somewhat lower than in 2019. But it is still 2.6 percentage points higher than it was in 2006 before the Great Recession and 5.9 percentage points higher than in 2000. This compares to a somewhat larger 7.4 percentage-point increase for less-educated U.S.-born men since 2000. But the share of both women and men born in the United States who do not have a bachelor’s not in the labor force remains extremely high in 2025 by historical standards.
Immigrants. Figure 8 shows the share of immigrant men ages 16 to 64 not in the labor force. (See additional information for immigrant men and women in Table 2 in Excel sheets.) is worth pointing out again that the decennial census and CPS data include both legal and illegal immigrants. There was a rather significant increase in the share of foreign-born men not in the labor force from 1960 to 1980. But it declined thereafter, reaching a low in 2006 before the Great Recession, which is very different than the sustained climb for U.S.-born men from 1980 to 2006. The rise from 1960 to 1980 should be interpreted with an understanding that the immigrant population can change more quickly than the U.S.-born population, which it did from 1960 to 1980. In 1960, there were still some working-age immigrants who entered during the so-called great wave of immigration, prior to 1924. Perhaps more importantly, the immigrants in the immediate post-WW II period from Europe tended to be relatively modest in number and relatively higher-skilled.12 As that population fell as a share of the total foreign-born population after 1960, the share not in the labor force might have been impacted.
 |
Figure 8 indicates that the overall share of immigrant men 16 to 64 not in the labor force has roughly returned to pre-Covid levels, though it remains somewhat higher than in 2006. Turning to the less-educated, we see that the percentage of immigrant men without a bachelor’s not in the labor force is 2.1 percentage points higher in 2025 than when it hit a low point in 2006. While less than the 4.3 percentage-point deterioration for U.S.-born men without a bachelor’s over this same period (Figure 1), it is still an indication that immigrant men without a bachelor’s have not been entirely immune from the rise in the share neither working nor looking for work.
A Huge Pool of Potential Labor. Figure 9 shows the number of working-age (16 to 64) immigrants and U.S.-born people of both sexes not in the labor force. The 52.9 million non-institutionalized U.S. residents out of the labor force in April 2025 was down from 2019 but remains enormous. Compared to the peak in 2006, the total number not in the labor force in 2025 is nearly five million larger, and relative to 2000 it is 12.2 million larger. Two additional things should be kept in mind about these numbers. First, as is the case with all of the figures reported in this analysis, those in institutions, primarily prisons and jails, are not counted. If the working-age institutionalized were included it would add significantly. Second, the 6.6 million unemployed people in April of this year are also not included in the total numbers as they are actively looking for work and are thus considered to be part of the labor force.
 |
Those Not in the Labor Force. Of course, the enormous number of non-institutionalized working-age people who are out of the labor force does not mean all or even most of these individuals can, should, or would work. Many are in school, others care for family members, some are disabled, others have retired early, and to be sure some could legitimately be described as lazy and completely unwilling to ever work. What these numbers do mean is that there is a very large pool of potential workers available, some of whom could be brought into the labor force if businesses recruited, offering more competitive wages, benefits, and working conditions. As we have made clear throughout this report, not so long ago a much larger share of working-age U.S.-born men and even to some extent women were in the labor force. If just 12 percent of working-age (16 to 64) U.S.-born men and women currently not in the labor force join/rejoin it, it would bring the rate nearly back to what it was in 2000 and add 5.2 million workers to the labor force. This is roughly equal in number to all of the growth in the number of immigrants (legal and illegal), including the recent surge, added to the labor force since 2016.13
Labor Force Trends Every Year. Rather than just looking at peak years, Figure 10 shows the share of men 16 to 64 not in the labor force for every year from 1962 through 2025 based on the CPS ASEC. (See the introduction of this report for a discussion of the CPS ASEC.) We know from decennial census data that 92 to 94 percent of the male labor force from 1960 to 1980 was U.S.-born, so the overall increase in the share of all working-age men not in the labor force in the 1960s and 1970s shown in Figure 10 primarily reflects the deterioration among the American-born. As we have seen (Figure 8), an increasing share of foreign-born men were not in the labor force from 1960 to 1980 as well. But immigrants were a relatively small share of the labor force back then, so the deterioration in labor force participation for working-age men overall in those decades primarily reflects the increasing share of native-born men not working or looking for work.
 |
Figure 10 shows that the share of U.S.-born men not in the labor force typically grows after each recession, which is to be expected since, like unemployment, it is a lagging indicator of the economy. But the figure also shows that participation in the labor force generally did not make it back to pre-recession levels for the U.S.-born once the economy recovered. Figure 10 also shows that in 1994, the first year for which the CPS ASEC has nativity data, there was only a modest difference in rates between immigrants and the U.S.-born, but that changed thereafter with the foreign-born improving and the U.S.-born deteriorating.14 We see the same general trend in Figure 11, which shows the share of prime-age men 25 to 54 not in the labor force based on the CPS ASEC for every year. In fact, prime-age immigrant men were actually more likely to be out of the labor force in 1994 than their U.S.-born counterparts, which is completely reversed today. Like all the other figures we have examined, looking at the CPS ASEC shows a very long-term deterioration in the share of U.S.-born men who participate in the labor force.
 |
Less-Educated Men for Every Year. Figure 12 reports the share not in the labor force for foreign- and U.S.-born men without a bachelor’s degree ages 20 to 64 for every year using the CPS ASEC. Less-educated immigrants and U.S.-born American men were about equally likely not be in the labor force in 1994. But that changed thereafter, with the immigrant rate decreasing and the U.S.-born rate climbing so that a large gap exists in 2025 in the share not in the labor force for both groups even though the share out of the labor force has returned to the 2019 pre-Covid levels. Figure 13 shows the same information as Figure 12 except the figures are only for prime-age men, ages 25 to 54. Both prime-age immigrant and U.S.-born men have roughly made it back to pre-Covid levels by 2025, but the share of U.S.-born men not in the labor has increased for decades, while the situation for immigrants has been much better. The CPS ASEC, like decennial census data and the April monthly CPS data reported in prior figures, all show significant growth in the share of less-educated, U.S.-born men not in the labor force for roughly six decades.
 |
 |
There are clearly a huge number of working-age people on the economic sidelines due to the profound long-term decline in the labor force participation of working-age, U.S.-born men, and to some extent women. However, the causes of the decline are much debated. Below, we briefly summarize some of the explanations offered for the falloff in participation. It seems certain that there is no single cause for the decline. Rather, many factors likely contribute to the problem, and these factors almost certainly interact with each other in complex ways.
Weak Labor Demand. There is no question, as Figures 10 to 13 show, that the share not in the labor force often rises during recessions when demand for labor is slack. But it is less clear why the rate does not return to pre-recessions levels when the job market recovers. The very brief Covid recession, which prompted the massive stimulus spending in response, seems to be the exception. The total number (immigrant and U.S.-born) of men and women without a bachelor’s holding a job increased from 59.2 million in 1960 to 100.4 million in 2006, right before the Great Recession. Demand for labor was certainly not weak over this 46-year period. The number of workers without a bachelor’s holding a job is 95.8 million in 2025, down from 2006. This reflects the general steady increase in the share of the U.S. population with more education. Still, it is very hard to argue that demand for less-educated workers somehow collapsed, especially before 2006 when demand seemed to steadily increase.
Nonetheless, both the Obama White House and the Brookings Institution emphasized weak labor demand for less-educated workers and the resulting decline in wages for the less-educated as an important cause of the increase in those leaving the labor force. Other researchers have come to similar conclusions. Globalization, particularly trade and technological innovations, sometimes referred to as “skill-biased technological change”, are typically identified as key causes for reducing demand for less-educated workers.15 Clearly, low wages are a disincentive to work and must have played some role in the growth of those not in the labor force. But if demand for less-educated labor is weak, and this is causing a growing number of less-educated American men to leave the labor force, then current immigration policy is profoundly misguided.
In the April 2025 CPS, 11.9 million immigrant workers (legal and illegal) without a bachelor’s degree indicated they came to the United States in 2000 or later.16 Immigration has certainly added enormously to the supply of less-educated workers. That said, we know that the majority of jobs in nearly every occupational category, as defined by the Department of Commerce, are done by the U.S.-born, so the argument that the migrants only do jobs Americans don’t want is false. Adding so many new less-educated workers to the economy when demand did not keep pace with the size of the non-college-educated population seems very unwise if we want to improve the share of working-age men in the labor force and avoid all the negative consequences caused by the increasing share of men not in the labor force.
Labor Demand Relative to Immigration. It is reasonable to argue that demand for less-educated male labor has been weak relative to the increase in the supply of less-educated workers, but only once immigration is considered. From 1960 to 2006, the number of 16- to 64-year-old, less-educated men in the labor force (both immigrant and U.S.-born) increased 15.3 million, but only 8.1 million of that increase went to the U.S.-born, while the rest went to foreign-born men. The 8.1 million increase represented a 21.5 percent increase in the number of less-educated, U.S.-born working-age men in the labor force. However, the total number of working-age, less-educated, U.S.-born men (in and out of the labor force) increased by 39.3 percent (16.7 million) over this same time period, hence the huge increase in the share and number not in the labor force from 1960 to 2006. The overall number of U.S.-born men without a bachelor’s degree ages 16 to 64 (in and out of the labor force) was 1.7 percent (one million) lower in 2025 compared to 2006.17 However, the number in the labor force fell by 7.2 percent (3.3 million) over this time period, so their participation is lower in 2025 compared to 2006. As for less-educated, working-age, male immigrants, the number in the labor force actually increased by one million (10 percent) from 2006 to 2025. It is clear that demand for less-educated male workers has not kept pace with the total population of less-educated, working-age men. But the weak demand argument only makes sense in the context of adding so many immigrant workers.
Changing Social Norms. Nicholas Eberstadt argues that slack labor demand for less-educated workers is an inadequate explanation because foreign-born men with modest levels of education have not experienced the same decline as U.S.-born men. Perhaps even more important, the decline for less-educated, unmarried men is much more pronounced than for married men of the same age and skill level. He believes that changing expectations about men as providers explains much of the decline due in large part to the decline in institutions that reinforced the value of work, particularly marriage. This makes it increasingly acceptable for men to rely on family members, such as parents or partners, for support. He also argues that the welfare state undermines work and enables those not in the labor force to at least have some income. These arguments echo those made by anti-poverty researcher Lawrence Mead, who has long argued that “culture” and “social isolation” are key reasons why an increasing share of working-age men do not work and not so much economic conditions.
The Welfare State. Some researchers focus specifically on what they believe is an overly generous and easily accessible welfare and disability system as a cause of the increase in the share leaving the labor force. Growth in use of disability is particularly striking, even if we cannot fully measure it because the government does not track all the sources of disability payments, public and private. We do know that use of the largest disability program, Social Security Disability Insurance (SSDI), has grown dramatically. The total number of workers in America was 2.5 times larger in 2021 than in 1960, but the number of people on SSDI who qualified as workers was 17 times larger in 2021 than in 1960.18 From 2000 to 2021, the total number of workers was up about 13 percent, but the number on SSDI was up 56 percent, though the number is down somewhat since the height of the Great Recession. Eberstadt observes that disability programs are, “increasingly used as income-support mechanisms for men on a work-free life track”.
Immigration. Direct competition with immigrants likely has also contributed to the rising numbers leaving the labor force. In a prior analysis, we found that the fall-off in immigration in the first three years of the Trump administration, before Covid, coincided with an increase in wages and labor force participation for less-educated, U.S.-born Americans relative to the higher period of immigration in the prior two decades. The National Academies’ comprehensive 2016 study found that immigration reduces wages for some U.S.-born workers, which almost certainly undermines work incentives. A 2019 CIS analysis of EEOC discrimination cases found that immigrants can be used to replace U.S-born workers. Other research has found a negative impact on the employment of young, U.S.-born workers, while more than one study has found a negative impact on the employment of Black Americans from immigration.
Several recent academic papers show that the arrival of immigrants in an area tends to cause the U.S.-born to move out, likely to avoid competition. Perhaps as important as direct competition between the U.S.-born and foreign-born is that the availability of immigrant labor allows policy-makers, business groups, and American society in general to ignore all those on the economic sidelines and the resulting social problems.
Other Possible Causes. It has long been known that having a criminal record significantly reduces one’s employment prospects. There is good reason to think the dramatic increase in the share of less-educated men with a criminal record in recent decades has significantly impeded the willingness of employers to hire them, particularly African-American men. There is also evidence that in addition to being caused by dropping out of the labor force, addiction itself reduces labor force participation. One factor that can be ruled out as a cause of the deteriorating participation in the labor force is population aging. Some research includes the elderly in its analysis and then identifies population aging as a cause of declining participation. But aging cannot really explain the increasing share of working-age people not in the labor force. We also know that less-educated men are not increasingly assuming household responsibilities, such as childcare, rather than working. Fewer men are fathers now than in the past, and fewer fathers live with children. Moreover, the time utilization survey shows that men not in the labor force, unlike women, spend relatively little time caring for others. Rather, these men report they spend much of their time watching TV, socializing, relaxing, and other leisure activities.
ConclusionThe findings of this analysis confirm prior research showing a six-decade-long increase in the share of working-age men not in the labor force. Unlike most research on this issue, we break out the U.S.-born and foreign-born separately going back decades. The share not in the labor force — neither working nor looking for work — rose for both U.S.-born and foreign-born men from 1960 to 1980, but since 1980 the rate has not continued to climb for immigrant men in the way that it has for the U.S.-born. The dramatic increase in U.S.-born men not in the labor force means there is now a huge number of potential workers on the economic sidelines who are out of the labor force entirely but are not counted among the unemployed because they have not looked for a job in the four weeks prior to being surveyed. The total number of U.S.-born men and women ages 16 to 64 not in the labor force was 43.4 million in April 2025, 8.9 million larger than in April 2000. This does not include the 5.4 million U.S. natives unemployed who are not working but are considered part of the labor force. It also does not include the roughly 11 million working-age immigrants not in the labor force or unemployed.
The rise in those not in the labor force is primarily confined to those without a college education. The data also shows an increase for these less-educated, U.S.-born men no matter how “working-age” is defined — 16 to 64, 18 to 64, 25 to 64, or even only those of “prime-age”, 25 to 54. Confining the analysis to only the working-age means the increase in the percentage not in the labor force has nothing to do with the growing share of the U.S. population that is over age 65. Among less-educated, U.S.-born women, the share not in the labor force has also gone up some since 2000, though it has been better post-Covid. In addition, the share of less-educated, immigrant men not in the labor force has also increased some since 2006, though the increase is not as large as for less-educated, U.S.-born men.
We summarize in this report the much-debated reasons for the increasing share of less-educated people not in the labor force. It is likely that many factors have contributed to this situation, including low wages, prior criminal convictions, ease of access to welfare and disability programs, competition with immigrants, and changing values and norms about the importance of work. While there is no agreement on the relative importance of these individual causes, there is agreement that the decline in work is associated with a host of serious social problems. These include, but are not limited to, substance abuse, crime, overdose deaths, suicide, obesity, welfare dependence, and social isolation, to say nothing of the fiscal and economic costs of having a larger share of working-age people not in the labor force.
The growth of those not in the labor force means there is huge pool of working-age people from which employers could potentially draw to fill jobs. If labor force participation for U.S.-born men and women together returned to the peak years of 1960 for men and 2000 for women, it would add a total of 10 million people to the labor force. Even if the participation rate only returned to the 2000 rate, it would still add 4.1 million U.S.-born men and 1.1 million U.S.-born women to the labor force. This is germane to the immigration debate because one of the primary justifications for large-scale legal immigration, and even tolerating illegal immigration, is that there are not enough people to fill all of the jobs available. But to make that argument one has to ignore or at least dismiss concern over the enormous number of working-age, U.S.-born men and women not in the labor force and all the problems this creates for society.
Trying to draw significantly more working-age men, and to some extent women, into the labor market will certainly not be quick or easy. It will involve the difficult task of reforming our welfare and disability systems, combating the opioid and mental health crises, improving job training, and reexamining our approach to globalization. Allowing wages and benefits to rise, partly by reducing immigration, is almost certainly a key part of the solution. Perhaps most challenging, we will need to figure out how to re-instill the value of work.
We are certainly much less likely to undertake needed reforms if we continue to keep the level of legal immigration very high and tolerate widespread illegal immigration. The business community, policy-makers, and society in general have much less incentive to deal with the huge share of the working-age population out of the labor force if we can continue to simply turn to immigrants to fill jobs. We have a choice as a country: We can either adopt policies designed to get more working-age Americans currently on the economic sidelines into jobs or we can ignore the problem and continue to bring in ever more immigrants and then try to manage all the pathologies associated with our higher rate of the working-age not in the labor force.
End Notes1 This report uses the April data because historical censuses are collected in April of each year and the decennial census is the only source of data before 1994 that provides information on employment by nativity on a consistent basis. Using April CPS data essentially controls for seasonality in the data. In April 2025 there were 6.6 million unemployed immigrants and natives (seasonally unadjusted) based on Table A-7. The official unemployment numbers include a relatively small number of people 65 and older actively looking for work.
2 The term “immigrant” has a specific meaning in U.S. immigration law, which is all those inspected and admitted as lawful permanent residents. In this analysis, we use the term “immigrant” in the non-technical sense of the word to mean all those who were not U.S. citizens at birth. Typically, the government refers to these individuals in surveys such as the CPS or decennial censuses as the “foreign-born”. While it may surprise some, the BLS is clear that illegal immigrants are included in the survey, and the Census Bureau is equally clear that illegal immigrants are included in the census every 10 years, though in both cases some fraction are missed. Although there is some undercount, immigrants, or the foreign-born, in the CPS and decennial census include naturalized citizens, legal permanent residents (green card holders), long-term temporary visitors (e.g. guestworkers), and illegal immigrants.
3 The standard unemployment rate using the monthly CPS is calculated by dividing the number of people actively looking for work in the last four weeks by the number in the labor force (working or looking). Because of the way it is calculated, those not in the labor force are not included in the official unemployment rate. This means that the number of working-age people not working at any one time is dramatically larger than the number who are officially considered unemployed.
4 Because the CPS only began to ask about citizenship in 1994 on a regular basis, we cannot use the survey to measure employment by nativity before that year. In order to measure long-term trends in employment by nativity, we use the decennial census in earlier years. While this allows us to divide the population by nativity back to 1960, we are limited to the years of the census.
5 Looking at the monthly data, it seems employment peaked in 1970, 1979, and 1989. But fortunately 1980 and 1990 are not far off the peak. Using the census still gives a pretty clear picture of the long-term decline at the peak of each business cycle along with nativity data.
6 We use the April CPS to create continuity with the decennial census, which is also collected in that month. While the figures in this report are not seasonally adjusted, we are looking at the same month each year, which should do a good job of controlling for seasonality. We use unadjusted numbers because such numbers are conceptually simpler and easier for other researchers to replicate. Moreover, we are comparing CPS data and decennial census data in this report and the census is not seasonally adjusted. Also, the limited number of statistics on the foreign-born published in the BLS monthly "Employment Situation" (Table A-7) reports are, like those reported here, not seasonally adjusted. The counts and percentages, particularly for the sub-populations we report based on the CPS will, in some cases, be slightly different from published BLS reports. The reason is that the Census Bureau adds "perturbations" to the public-use microdata to protect respondent confidentiality. Fortunately, the Census Bureau states that any differences between published figures and those calculated from the public-use data will be so small that they "fall well within the sampling variability associated with CPS estimates".
7 Average total personal income (from all sources) in the 2025 Current Population Survey ASEC for men 20 to 64 who were in the labor force was $92,332, compared to only $26,548 for men in this age group not in the labor force. Moreover, 5 percent of men 20 to 64 in the labor force lived in poverty, based on this same data, compared to 27 percent who were not in the labor force.
8 For example, using the 2025 CPS ASEC we find that households in which all persons are under age 65, but no one works, pay on average $4,326 in federal income tax, compared to an average of $16,911 in households where all persons are under age 65, but at least one person works. Further, if we use the same data to look just at the biggest welfare programs (SSI, TANF, SNAP, public/rent subsidized housing, and Medicaid) and again confine the analysis to only households where everyone is under 65, it shows that, for households with no workers, 56.9 percent receive at least one of these programs compared to 22.1 percent for households where at least one person works.
9 We use 1962 as the starting date because that is the first year that the Ipums website has data for the CPS ASEC.
10 It should also be pointed out that of the 18.4 million working-age, U.S.-born men not in the labor force in April of this year, 87 percent did not have a bachelor’s degree. If we confine the analysis to only prime-age men, it is still the case that 81 percent of those out of the labor market do not have a bachelor’s degree. The huge pool of potential male labor in the country is disproportionately less-educated.
11 When thinking about this number, it is important to note both the change in the share not in the labor force and the overall growth in the U.S.-born population. That is, an increasingly larger share of an ever-larger total number is not in the labor force.
12 Unlike the U.S.-born population, which grows from within the country in a more predictable fashion, the immigrant population does not reproduce itself in the United States — all children born to immigrants are by definition native-born. For example, in 1960 the average male immigrant aged 16 to 64 (in and out of the labor force) was 45.8 years old, compared to 36.7 years old in 1980 and 36.2 years old in 1990. Second, 48.4 percent of working-age male immigrants were 50 and older in 1960, compared to 21.1 percent in 1980 and 17.3 percent in 1990. Age can impact participation rates. However, it should also be pointed out that the more detailed analysis reported in the Excel file (Table 2) shows that when the analysis is confined to only male immigrants 25 to 34, we still see the same decline from 1960 to 1980. Other possible explanations for the decline include the change in the primary sending countries, from Europe to Latin America and Asia, and the resulting relative decline in educational attainment of immigrants. Additional analysis would be necessary to determine what factor or factors caused the increase in the share not in the labor force among immigrant men from 1960 to 1980. Of course, perhaps no immigrant-specific explanation is really needed since the increase of those not in the labor force occurred for U.S.-born men over this time period as well. Rather, perhaps the real question is why the increase did not continue as it did for the U.S.-born.
13 Compares number of immigrants in the labor force in 2016 to the number in 2025. All of these figures are for the working-age, ages 16 to 64, and do not include the modest number of immigrant workers over age 64.
14 In 1994 there seems to be a break in the continuity of the data, with an increase in the share not in the labor force that does not correspond to a recession, though of course employment and related statistics are always a lagging indicator of a downturn. The BLS states that, “The 1994 redesign was to improve the overall quality of labor market information through extensive question changes and the introduction of computers into the collection procedures.” The survey redesign and the new questions may help to explain the seemingly sudden change.
15 Skill-biased technological change can be defined as any economic change resulting from new technologies that increase demand for more-skilled or more-educated workers relative to less-skilled, less-educated workers.
16 This includes the relatively small number of workers over age 64.
17 There was a recovery in the overall number of less-educated men in the labor force so that there were somewhat more in the labor force by 2013 than there had been in 2006, but by the time of the next peak in the business cycle in 2019, that number had declined and was below the peak in 2006. This reflects the general increase in education of the population and the increasing number and share of jobs done by those with a bachelor’s or more.
18 These figures are only for those who themselves qualified for the program because they worked at some point in their lives. Some people on disability do work, but most do not because they are unable to do so and because there are significant limits on how much they are allowed to earn and still qualify for benefits.