The Heritage Foundation

Center for Data Analysis Report #08-04 on Jobs, Jobs and Labor Policy

June 17, 2008

June 17, 2008 | Center for Data Analysis Report on Jobs, Jobs and Labor Policy

A Good Job Is Not So Hard to Find

Many believe that the era of good jobs is slipping away. Especially now with the economy weakening, the popular perception is that Americans are less and less able to find jobs that will allow them to support their families. Some analysts argue that so few workers can find jobs with health insurance that America now needs nationalized health care.[1]  Members of Congress believe that the government needs to make it easier for unions to recruit new members-even by depriving workers of the right to vote on joining a union-to prevent the disap­pearance of good jobs.[2] 

These fears are unfounded. There is a natural human tendency to look at the past with nostalgia, but the economy before the 1980s was not a golden era when every worker had a fulfilling, high-paying job. Many of the jobs of a generation ago paid low wages and offered little job security. The economy is constantly in flux, creating new jobs and destroying old ones. Computer programmers were as rare in the 1970s as typewriter repairmen are today. Some of the jobs that have disappeared were good jobs, but many were not.

The data show that job opportunities have grown the most in occupations with the highest wages while shrinking in most low-wage occupations. Employers need more workers with education and skills, and this increases the opportunities for upward mobility. Today, the typical job involves less mindless repetition or physical exertion and more mental activity than in the past. It is also safer and more comfortable than a generation ago.

Workers' benefits are not disappearing. After accounting for the effects of unskilled immigration, American workers are just as likely to have health insurance today as a decade ago. Pension coverage is stable, but employers are switching toward defined contribution pensions that give workers more control. Access to sick leave has expanded. Workers have greater job security than in the past: They are more likely to move between employers but less likely to be fired than in the 1970s.

The opportunities available to American workers have expanded over the past generation. While Congress is understandably concerned by the cur­rent economic weakness, Congress should not pass legislation based on the mistaken belief that work­ing conditions have deteriorated.

Economic Transitions

Many Americans have heard of friends or acquaintances who have lost good jobs and could not find a comparable new job. These anecdotal sto­ries reveal an economic reality: Such hardships do occur. Similarly, many workers begin their career in undesirable jobs and move up the career ladder, although these stories attract less media attention. The economy is constantly creating new jobs and destroying old ones as technology and consumer preferences change.


Economists call this "creative destruction." Busi­nesses that serve consumers more effectively and more competitively will expand and add jobs, and less efficient firms will go out of business. Employ­ers created 57.8 million new jobs in 2007 while ending 54.6 million positions.[3]  This is essential to a healthy economy. It directs employees to work where their skills are most valuable in the economy and ensures that their talents are not wasted at obsolete firms. Thirty years ago, many companies employed workers to repair broken typewriters. Today, those businesses and jobs have disappeared, but tens of thousands of workers repair personal computers that did not exist then. Without this con­stant churning, the economy would stagnate.

The Changing Nature of Work

Many workers lose good jobs every year, and many workers start good jobs. Determining overall changes in the job opportunities available to work­ers requires looking at changes in the overall distri­bution of jobs in the economy, not at individual cases. Chart 1 shows the change in the distribution of jobs in the economy from 1980 to 2006.

Contrary to the popular impression that workers had better job opportunities in the past, the econ­omy now requires more workers in high-wage occupations than a generation ago. The fastest-growing occupations have been in professional spe­cialty fields, which include positions such as engi­neers, computer scientists, and registered nurses. These employees work in the second most highly compensated sector of the economy. The most highly compensated sector-executive and mana­gerial occupations such as financial managers, marketing specialists, and human resource profes­sionals-has also expanded. Collectively, the occu­pations with above-median pay expanded their share of jobs in the economy by 8.2 percentage points from 1980 to 2006.


The economy shed jobs in all but one of the occu­pations that pay below-median wages. The low-wage operators, fabricators, and laborers sector (e.g., printing machine operators, solderers, and garbage collectors) experienced the largest drop in employment, falling by 5.7 percentage points. Administrative support and clerical workers, another low-wage sector, also declined as voice mail and computers reduced the need for their services. The exception among low-wage sectors was service occupations, which rose 3.1 percent.

Accounting for Immigration.However, this analysis does not account for the many new workers who have joined the economy since 1980. The economy has added jobs in high-wage sectors and the service sector, but this does not mean that the typical worker employed in the 1980s is now more likely to work in one of those sectors. Nor do changes in average wages necessarily mean that individual workers are better or worse off.

Immigration-both legal and illegal-has increased dramatically since 1980. The number of immigrant workers as a proportion of the workforce has more than doubled in the past 26 years, rising from 7.2 percent in 1980 to 17.4 percent in 2006.[4] 

Immigrants come to America looking for better opportunities than they could find in their home countries. In many cases, they succeed. However, many immigrants who come to the United States are low-skill workers. Chart 2 shows the educa­tional attainment of foreign-born workers and native-born workers in the United States in 2006.

While many highly educated workers migrate to the United States, an immigrant is more than four times as likely as a native-born worker to lack a high school education. Unsurprisingly, given their lower skills, the median immigrant earns less ($14.88 per hour) than the median American-born worker earns ($18.45 per hour). Yet this is well beyond what many immigrants could earn in their home countries. Although most immigrants improve their material circumstances by moving to America, many start out at the bottom of the U.S. economic ladder.

This means that comparing snapshots of working conditions from two points in time creates a mis­leading picture of how the welfare of U.S. workers has changed. As workers move up the economic ladder, millions of low-skill immigrants arrive and enter on the bottom rung. Adding millions of low-skill immigrants lowers average wages and expands the relative size of low-skill occupations.


Low-skill immigration will lower average wages even if every worker in a country earns a raise. The declining average age of a country with a high birth­rate illustrates this point. Because many children are born each year, the average age of the country's pop­ulation falls. Yet each individual in the country is growing older. Although the average age is falling, arguing that individuals are growing younger is sim­ply ridiculous.

Similarly, adding millions of low-wage immigrant workers to the economy lowers the average wage, even if every American earns a raise and every immigrant earns more in America than in their home country. Therefore, immigrants must be excluded from the calculations to gauge accurately how job opportunities available to American work­ers have changed.

Table 1 reveals that the quality of jobs in the econ­omy has improved even more substantially when looking at the jobs of American-born employees. The table shows the change in each occupation's share of employment between 1980 and 2006, both including and excluding foreign born workers.

The proportion of American-born workers employed in the three most highly paid occupations has expanded by 9.9 percentage points since 1980. The decline of low-wage operators, fabricators, and laborers positions and farming, fishing, and forestry jobs was even more pronounced, while the low-wage service sector, excluding immigrants, grew by only 2.1 percentage points over that time, com­pared to an increase of 3.1 percentage points if immigrants are included. In other words, immi­grant workers account for about one-third of the relative expansion of the service sector.

Many immigrants in the United States take low-skill jobs. In many cases, these jobs would not exist without immigrant workers.[5]  This does not mean that most positions available to American workers are low-skill jobs. On the contrary, American work­ers are much more likely to work in a quality job than was true a generation ago.

More Fulfilling Jobs.The number of jobs in high-wage occupations has increased because the nature of work has fundamentally changed. The economy needs fewer people performing repetitive tasks or doing manual labor. Chart 1 and Table 1 show that employment has declined the most among operators, fabricators, and laborers and in precision production, craft, and repair occupations. Robots and computers now do much of this work.


Chart 3 shows that manufacturing employment has declined as output has increased over the past two decades. Technology has made manufacturers more productive, producing more goods with fewer workers.

However, machines cannot think for people. Employers need skilled workers, and the economy is shifting toward jobs that require greater mental labor and less physical labor than was required in the past.[6]  Office work is replacing the assembly line.

This means better opportunities for American workers. Jobs that require more mental labor tend to pay higher wages, have more flexible hours, and offer better benefits than jobs that require manual labor.[7]  Workers also tend to prefer these jobs. Research shows that employees work more now because the nature of work has shifted and because the quality of their jobs has increased.[8]  The assem­bly line is physically demanding and monotonous. Most workers find working in an office more enjoy­able than fitting into place one widget after another in a factory.

More Educated Workers. The increased need for brainpower and the decreased need for man­power mean that employers need workers who are more educated. As a result, both the average educa­tion level and the return to education in the econ­omy have grown dramatically.

Chart 1 and Table 1 reveal that employers are creating more jobs in occupations that require the most education and fewer jobs in occupations that need the least education. Other research confirms this. Chart 4 shows the change in distribution of U.S. employment among occupations for 1980- 1990 and 1990-2000, with occupations ranked by the education required.[9]  In both decades, occupa­tions that required the most education grew the fastest, while those requiring less edu­cation either decreased or stayed rela­tively constant.

American workers have signifi­cantly more education today than they had a generation ago. (See Chart 5.) Today, 95 percent of American work­ers have at least a high school diploma, versus 79 percent in 1980. The proportion with at least a bache­lor's degree has risen by 50 percent. Today, almost two-thirds of workers have at least some college education.


This shift benefits American work­ers. The increased need for highly skilled and educated workers means expanded opportunities for upward mobility. Jobs that require more educa­tion pay higher wages than those that require less schooling. This has always been the case and has become espe­cially pronounced in recent years. The wage premium on a college education has increased over the past quarter-century. In 1980, the median worker with at least a bachelor's degree earned 44 percent more than the median high school graduate. In 2006, that differ­ence was 76 percent.[10] 

The jobs being created for Ameri­cans today are not predominantly "burger flipping" positions that re­quire few skills or little education. Instead, businesses need more highly skilled workers who can perform complex tasks. These jobs require more education and pay higher wages. Workers have more quality jobs available to them now than they did a generation ago. Skills and education provide an increasingly accessible pathway to success.

Service Occupations. However, not every occu­pation requiring less education or fewer skills has experienced declining employment. Service occu­pations (e.g., housekeepers, security guards, cooks, health aides, and janitors) are a notable exception to this trend. These occupations require manual labor or interpersonal interactions that cannot easily be automated. In 1980, 9.2 percent of employees worked in these jobs. By 2006, that figure had risen to 12.3 percent.

Service jobs are typically less skill-intensive than other jobs. Chart 6 shows the educational back­ground of workers in the service occupations and the overall economy in 1980 and 2006. Although the education level of workers has increased in all sectors of the economy, service occupations still require below-average skills. Of service workers, 56 percent have a high school education or less, com­pared to 38 percent in the overall economy.

The increase in service jobs reflects increased immigration and technological changes. Many immigrants work in service occupations precisely because those occupations require fewer skills. Table 1 shows that foreign-born workers account for one-third of the relative growth of the service sector.

Technological changes have also reduced the number of labor-intensive jobs. Some former fac­tory workers have found jobs in high-wage sectors of the economy, but others, especially low-skill workers, now work in service occupations. Chart 7 shows the occupations of non-immigrant employ­ees with a high school degree or less.


Employment as operators, fabricators, and labor­ers and in precision production, craft, and repair occupations has fallen by 4.8 percentage points among workers with a high school degree or less. Employment in service occupations for these work­ers has risen by 4.3 percentage points. Jobs in ser­vice occupations have replaced other low-skill jobs in the economy.


Service jobs require fewer skills and pay lower wages. Workers in service jobs earn the second least of any occupation, behind only farm workers. The economy has generally created more high-skill, high-paying jobs, but some low-wage occupations have expanded. The stories about the factory worker who loses his job and takes a job as a short-order cook are not so much inaccurate as incom­plete. Such things do happen, but they do not rep­resent how the economy as a whole has changed.


Rising Wages and Compensation

With the economy shifting toward better jobs in high-wage occupations, the income of the typical worker has risen. Table 2 shows hourly earnings and annual wage and salary incomes of American workers. Foreign-born workers are excluded from this table to prevent immigration from distorting the picture of how the welfare of American workers has changed over time.

Earnings data from the American Community Survey (ACS) and the decennial Censuses are not directly comparable because their survey questions are worded differently.[11]  Consequently, the data in Table 2 can be used to determine wage growth from 1980 to 2000 and from 2000 to 2006, but not from 1980 to 2006.

The wages of American workers have risen over the past generation. Average annual wage and salary income rose 27 percent between 1980 and 2000 and has recovered from the collapse of the dot-com bubble, rising 4 percent since 2000. Median incomes increased less quickly but still rose 15 per­cent between 1980 and 2000 and have risen 6 per­cent since 2000.

Contrary to popular wisdom, the largest income growth has occurred among the poorest workers. The annual incomes of workers in the bottom 10 percent by wage rose 39 percent between 1980 and 2000-more than in any other income group. The incomes of workers at the 25th percentile rose 21 percent during this period.

Chart 8 shows the percent change in hourly wages and annual wage and salary income for dif­ferent earnings levels from the Current Population Survey, which provides comparable data over time after 1993 and allows examination of changes in earnings over the past decade.[12] 


Chart 8 shows that hourly and annual earnings of American workers have increased notably since 1993. Wages grew rapidly in the late 1990s and slowed sharply after 2003. However, the typical American worker is substantially better off now than in 1993. Between 1993 and 2006, average hourly earnings rose 28 percent, and average annual earn­ings rose 18 percent. During the same period, median hourly earnings rose 17 percent, and median annual earnings rose 18 percent.

Contrary to the conventional wisdom that the rich get richer and the poor get poorer, the data show that incomes have risen at all income levels. The incomes of workers in the top 10 percent have risen 20 percent since 1993. The incomes of work­ers at the bottom 10 percent are up 18 percent.

Minimum Wage Work Rare. A common mis­perception is that while some good jobs exist, mil­lions of Americans are trapped trying to provide for their families on minimum wage incomes. This is not the case. Very few Americans work for the min­imum wage. The federal minimum wage increased to $5.85 per hour in July 2007. Only 1.3 percent of American workers earn this wage.[13] 


Most of these minimum wage workers do not rely solely on their wages. Half are between the ages of 16 and 24, and 61 percent work part time. The ste­reotypical single parent working a minimum wage job fulltime to provide for his or her family is very rare. Less than one in 15 minimum wage earners fall into this category.[14]  Although it does occur, very few workers in today's economy rely on a minimum wage income to support their families.

Upward Mobility. Workers are not trapped in fixed income classes. The workers earning the min­imum wage in 1993 are not the same workers earn­ing the minimum wage today. Two-thirds of minimum wage workers earn a raise within a year.[15]  Many formerly low earners have moved up into the middle class and higher. Looking only at how average wages have changed ignores the sub­stantial upward mobility that characterizes the American economy.

Table 3 shows the results of a Treasury Depart­ment study following individual wages between 1996 and 2005. Most workers' situations improved markedly. The average worker's income rose 38 per­cent. Counterintuitively, the wealthy experienced much less wage growth than was experienced by low-wage workers. The average income of the top 1 percent rose only 13 percent. Workers in the bot­tom 20 percent experienced the largest increases, with their median income increasing by 90.5 per­cent and mean income increasing by 232.5 percent.


America is not becoming a nation of low-wage fast-food workers. Job opportunities are expanding, with the typical job paying more than it did a gener­ation ago. American workers, especially low-income workers, experience great upward mobility. The wealthy are getting wealthier, and so are the poor.

Health Insurance.Wages are only part of work­ers' earnings. Over 30 percent of the typical worker's total compensation is provided in bene­fits.[16]  Many workers consider a "good job" one that provides health insurance and/or a pension. The popular perception is that finding a job with these benefits has become increasingly difficult.


This conventional wisdom is also wrong. The overwhelming majority of American-born employ­ees working regular hours have employer-provided health insurance, either as the policyholders them­selves or as the dependents of other workers with employer-provided insurance. Chart 9 shows health insurance coverage rates since 1995.[17] 

Health insurance rates have barely changed over the past decade. In 1995, 80.7 percent of Ameri­can-born workers had employer-provided health coverage, which is statistically identical to the 80.5 percent rate in 2006. The health insurance rate did rise slightly to 83.2 percent during the dot-com bubble of the late 1990s before returning to its previous level after the bubble burst, but these were minor fluctuations. Now, as a decade ago, more than 80 percent of employees working regu­lar hours have employer-provided health insur­ance. Employers have also increased the amount they spend on health benefits. The average employee today earns 44 percent more in health insurance benefits per hour of work than in 1994.[18]  The conventional wisdom that employer-provided health benefits are becoming a relic of the past is simply wrong.

This misconception is partially the result of the massive influx of unskilled immigrants into the economy over the past decade. These immigrants, many of whom are illegal, often work in positions that do not offer benefits. Employer-provided cov­erage among all workers, including foreign-born workers, has fallen 2.0 percentage points since 1996 and 3.8 percentage points since its high in 2000.[19]  However, this does not mean that the job opportunities for American workers have declined. It means that there has been an influx of unskilled foreign-born workers.

Pensions Not Disappearing. Coupled with con­cerns over the disappearance of health insurance are concerns that fewer employers provide pension ben­efits. These fears are also exaggerated. Company pensions are not disappearing from the economy.

Chart 10 shows that a large majority of workers are offered employer-provided pensions and that a majority have such pensions. Current Population Survey (CPS) data show a decrease of about 1 per­centage point in pension coverage since 1993. However, other government surveys show that employer pension coverage increased by a few per­centage points during this same period.[20] 


Chart 11 shows pension coverage rates for Ameri­cans looking at four different data sources: the Current Population Survey, the Survey of Income and Program Participation, the Survey of Consumer Finances, and employer 5500 tax forms filed with the IRS. This chart looks at private-sector wage and salary workers between the ages of 25 and 64 but does not restrict attention to American-born workers. As the chart shows, the availability of employer-pro­vided pensions has been stable in recent decades.[21] 


The Shift to Defined-Contribution Pensions. Although the availability of employer pensions has not changed in recent years, the type of pension that employers offer has changed dramatically. A gener­ation ago, most employers provided defined-benefit pension plans. Today, most employers offer defined-contribution plans.

In a defined-benefit pension, the employer guar­antees a set benefit to the worker after retirement, such as 1.5 percent of the worker's final salary multiplied by the number of years that he or she worked. In a defined-contribution plan, the employer contributes a set amount to a pension plan owned by the employee-for example, by depositing an amount equal to 6 percent of the employee's earnings into a 401(k).

Chart 12 shows this shift toward defined-contri­bution pensions. In 1980, almost two-thirds of workers with an employer-sponsored pension had defined-benefit plans. By 2003, the ratio had more than reversed, with almost 70 percent of workers with pensions in defined-contribution plans.

Employers have increased, not decreased, the amount they spend providing retirement benefits during this transition. The average employee now earns 32 percent more per hour worked in retire­ment benefits than in 1991.[22]  Counterintuitively, workers today earn more generous retirement ben­efits than in the past.

The Advantages of Defined-Benefit Plans. Organized labor frequently criticizes the shift from defined-benefit pensions to defined-contribution pensions.[23]  This criticism of the disappearance of defined-benefit pensions helps to explain why many Americans mistakenly believe that employer-provided pensions are disappearing.


Part of the labor movement's opposition to defined-contribution plans stems from institutional self-interest. Unions use the stocks that they man­age in collectively bargained defined-benefit plans to pressure corporate directors. A recent study of one AFL-CIO pension found that the union's proxy votes "partly reflect union worker interests rather than the objective of maximizing equity value alone."[24]  Unions use workers' savings in defined-benefit pensions to achieve their own objectives. They could not do this if the workers controlled their own pensions in defined-contribution plans.

Other concerns do not stem from institutional self-interest. Under a defined-benefit pension, the company owes the worker the promised benefit. If the corporate investments do poorly, the company must still pay the promised benefits unless it has gone bankrupt. Under a defined-contribution plan, the company has no liability after its initial contri­bution. The employee bears the risk of poor invest­ment performance and is not guaranteed any fixed retirement benefit.

However, research shows that defined-contribu­tion plans are better for workers on balance.[25]  While workers bear the risk of investments per­forming poorly in a defined-contribution plan, this risk is small over the span of an entire working life. Furthermore, defined-benefit plans are not risk-free. The company operates the pension plan, and if the company goes bankrupt, workers may receive only a fraction of the promised benefits from the government. This happened at United Airlines and Bethlehem Steel. Defined-benefit plans are low-risk only if the employer stays in business.


Defined-contribution plans protect workers from this risk. In a defined-contribution plan, workers own their pension and do not lose any of it if their employer goes bankrupt. Employer contributions cannot be taken back once they have been made. Defined-contribution plans also protect workers if they change jobs because 401(k) and other defined-contribution pension plans are perfectly portable between employers. Workers do not lose any of their benefits when they start working for a new company. Under a defined-benefit plan, workers can lose a substantial portion of their pension ben­efits if they change jobs.[26] 

The advantage of higher returns makes a defined-contribution pension a better deal for all but the most extremely risk-averse employees. The typical worker will retire with more money with a defined-contribution pension than with a defined-benefit pension.[27]  The steady shift toward defined-contri­bution pensions means that more workers have pensions that give them greater control over their working lives and will provide them with more income for retirement.

Better Jobs

Job quality has also improved in non-mone­tary ways.

Safer Jobs. Jobs are safer and more enjoyable today than they were two decades ago.[28]  The shift away from manufacturing and manual labor-inten­sive jobs means less physical hardship and fewer workplace injuries. Workers are far more likely to be injured moving parts on an assembly line than they are photocopying documents in an office. New technologies have also made work safer, even in the more dangerous jobs.


Chart 13 shows that the rates of workplace inju­ries, illnesses, and fatalities since 1992 have decreased. On-the-job injuries and deaths have fallen sharply since 1992. Technology and the changing nature of work have made jobs less dan­gerous and less unpleasant than in the past. This is another way that the quality of jobs available to workers has increased, even if it cannot be mea­sured in dollars and cents.


More Sick Leave. Another misconception is the belief that employers have reduced the paid leave they provide their employees. This is another case of nostalgia painting a misleading picture of the past. Table 4 shows the percent of workers with access to paid holidays, paid vacation days, and paid sick leave since 1991. Now as two decades ago, 90 percent of full-time workers have paid holidays and vacation days.[29] 

Workers' access to paid sick leave has expanded since 1991. Then, 57 percent of full-time employees had paid sick leave. Today, 68 percent have paid sick leave. Workers now earn 16 percent more paid sick leave per hour of work than in 1991. All paid leave benefits, including vacations and holidays, have increased 17 percent per hour since 1991.[30] 

Increased Mobility and Increased Job Secu­rity.Even if the quality of jobs in the economy has not declined, many analysts worry that job security has decreased.[31]  This conventional wisdom holds that workers in good jobs can no longer count on keeping them. In the past, most workers stayed with the same employer throughout their career. Today, the argument claims, out­sourcing and corporate greed have put many jobs at risk. Businesses rou­tinely lay off senior employees, and the lifetime job is a relic of the past. Tens of millions of American workers could face layoffs at any time.

This belief is widespread, but it is contrary to the facts.

Like most myths, the belief in declining job security contains a grain of truth: Some measures show that employees are slightly more likely to leave their jobs now than they were during the mid-1980s. [32]   But such studies do not forecast the end of longterm jobs. Chart 14 shows the average length of time that men near retirement age spent with their longest employer.[33]  This number has fallen since 1980, but only to its 1969 level. This metric indicates that employee mobility has increased modestly over the past few decades, but not to historically unusual levels. Lifetime jobs have not disappeared from the economy. Other measures show that employees are modestly less likely to leave their job from one year to the next. [34]

A change in employee mobility does not necessarily mean reduced job security. Workers can leave their jobs voluntarily or involuntarily. Leaving a job to take a better one does not constitute reduced job security. Rather, job security concerns the likelihood that workers will find themselves unexpectedly unemployed. Neither surveys of workers nor empirical studies bear out this fear. When asked "How likely do you think it is that you will lose your job or be laid off over the next year?" workers have expressed essentially the same levels of concern since 1977, although it has fluctuated with the business cycle. (See Chart 15.)


Workers are more likely to believe that they could find an equally good job if they left their current employer. When workers were asked how easily they could find another job with the same income and fringe benefits as their current jobs, over two-thirds said that it would be very easy or somewhat easy-up sharply from the 1970s. (See Chart 16.) Workers themselves do not believe that their job prospects are less secure.


In this, workers agree with empirical studies that show that workers' job security has increased.35 Heritage Foundation research shows that men were almost 6 percent less likely to leave their jobs in 2006-2007 than in 1975-1976, and women were about 10 percent less likely to leave their jobs. Workers' attachment to their jobs has increased modestly. Interestingly, job security has increased even more dramatically; Men are 46 percent less likely and women are 33 percent less likely to leave their jobs and become unemployed than in the 1970s. (See Table 5.)


The conventional wisdom is clearly wrong: Workers' jobs are more secure now than they were a generation ago.

However, workers are significantly more likely to move between employers today than in the past, and such movement is usually voluntarily. Men are 39 percent more likely and women are 29 percent more likely to leave their current jobs and immediately start working for new employers than they were in the 1970s. Increased mobility benefits workers, as greater mobility gives workers more opportunities to move up. Workers rarely switch jobs without a better offer.

Pension Changes Facilitating Mobility. Work­ers have become more willing to switch jobs partly because of the increase in defined-contribution pen­sions. Defined-contribution pensions are perfectly portable between jobs, while defined-benefit pen­sions often penalize workers for switching employ­ers. Many workers would like to switch jobs, but they would lose thousands of dollars per year in pen­sion benefits. Defined-benefit pensions often act like golden handcuffs chaining workers to their jobs.

Legal changes in the 1980s encouraged compa­nies to start offering defined-contribution pensions, and as employers started doing so in large numbers, workers became more willing to switch jobs when they found better opportunities.[35]  Workers have become slightly less attached to their jobs because they now have more choices in employment, not because greedy chief executive officers casually lay off employees.


Projected Continued Growth of Good Jobs. Forecasting the future is difficult, but the best estimates show that the number of good jobs in the United States will continue to increase over the next decade. Table 6 shows projected job growth, broken down by earnings and educational requirements. Jobs in occupations with the highest earnings are pro­jected to grow the most rapidly, increasing 14 percent by 2016. Every lower-paying job quartile is expected to grow less rapidly than the econ­omy as a whole. Jobs requiring more education will also increase signifi­cantly faster than those requiring less education. Occupations requiring at least some college education will increase twice as fast as those requir­ing only a high school degree.

Table 7 shows the 25 occupations with the largest projected growth over the next decade. Not every rapidly growing occupation pays high wages, but most do. Of the 25 fastest-growing occupations, 16 are either in the highest or second-highest earnings quartiles. Most require above-average levels of edu­cation. The jobs of the future are not predominantly burger flippers, but network systems analysts, com­puter engineers, medical assistants, and database administrators.


The media and policymakers worry that good jobs are disappearing from the economy. These fears have gained heightened attention during the cur­rent economic slowdown and since the housing bubble burst.

However, these concerns are unfounded. Some workers have lost high-paying jobs and have been unable to find comparable new jobs, but these work­ers are the exception, not the rule. Throughout the economy, jobs paying high wages in fields requiring more education are more available today than they were a generation ago, while low-wage, low-skill jobs are decreasing. Technology and automation have increased the value of mental labor and reduced the need for manual labor. Office jobs are replacing jobs on the assembly line, causing work­ers' incomes to rise. The government expects these trends to continue, with high-wage, high-education occupations growing more quickly than the econ­omy as a whole.

Contrary to the prevailing wisdom, employer-pro­vided health insurance and pensions are as available now as they were in the mid-1990s. Worker pension plans have improved significantly, with most employ­ers shifting to defined-contribution pensions that provide workers with more money for retirement and do not penalize them for switching jobs.


Work has also become safer and less unpleasant, with on-the-job injuries down sharply. Access to sick leave has expanded. Workers have both more job security and more mobility than in the past. Workers are more likely to switch employers but less likely to be fired than a generation ago.

Good jobs are at least as common today as they have been in the past. Congress should not take rad­ical steps to address a non-existent crisis. For exam­ple, Congress should not deprive workers of the right to vote before joining a union to make it easier for unions to organize workers. Nor should Con­gress nationalize health care in the mistaken belief that employer-provided health coverage has shrunk. Rather, Congress should continue policies like tax relief that promote economic growth and the creation of more good jobs.

James Sherk is Bradley Fellow in Labor Policy in the Center for Data Analysis at The Heritage Foundation.

Appendix A

Charts 1, 2, 5, 6, and 7 and Tables 1 and 2 are based on Heritage Foundation calculations using data from the 1 percent samples of the 1980, 1990, and 2000 Censuses and the 2000 and 2006 Amer­ican Community Surveys. These data were obtained from the Integrated Public Use Microdata Series (IPUMS).[36] 

Occupations were separated according to the occupation classification system used in the 1990 Census. The IPUMS project codes this variable to provide consistent comparisons across years. The managerial and professional specialty occupations sector was divided into an executive and managerial sector and a professional specialty sector. The tech­nical, sales, and administrative support sector was divided into a technical and sales sector and an administrative support and clerical sector.

Individuals between the ages of 25 and 65 who worked at least 1,250 hours in the previous year (25 hours per week for 50 weeks per year) were included in this analysis. Today, a common criticism is that increasing numbers of workers cannot find full-time jobs and must turn to part-time employ­ment. The threshold of 1,250 hours per year was chosen to capture the effect of any shift toward or away from part time employment without examin­ing the wages of employees who do not rely on their jobs for their incomes. Changing employment pat­terns for a spouse in a dual-income couple who works part-time for only 10 hours per week shouldnot be conflated with an employee who cannot find full-time work and works only 30 hours per week.

Hourly wages were calculated as the reported wages and salary income for the previous year divided by the total number of hours worked in the previous year. The total number of hours worked was calculated as the usual number of hours worked per week multiplied by the number of weeks worked in the previous year. Wages are expressed in 2007 dollars using the personal con­sumption expenditures chain-type price index. Workers reporting earning less than $3.50 per hour or more than $500 per hour in inflation-adjusted dollars were omitted from the analysis.

Educational achievement was measured using the EDUCREC variable created by IPUMS to allow comparisons of education levels across different Census years. EDUCREC is a composite variable combining the educational codes used by the 1980 Census with the educational codes used in the 1990 and 2000 Censuses. In this paper, a high school dropout is a worker reporting less than a 12th grade education. A high school graduate is a worker reporting exactly a 12th grade education. Workers with some college education are those reporting one to three years of college. Workers with college degrees or higher are those reporting four or more years of college education.

Workers born outside the United States were considered immigrants.

Appendix B

Charts 8-10 are based on Heritage Foundation calculations using data from the March supplement to the Current Population Survey. These data were obtained from the Integrated Public Use Microdata Series-Current Population Survey.[37] 

As with the Census data, workers between the ages of 25 and 65 who worked at least 1,250 hours per year in the previous year were included in the analysis. Total hours worked were calculated by multiplying the usual hours worked per week in the previous year by the total number of weeks worked in the previous year.

Workers born outside the United States were considered immigrants.

All income and earnings variables were adjusted for inflation and expressed in 2007 dollars using the personal consumption expenditures chain-type price index. Hourly earnings are total wage and sal­ary income divided by the total number of hours worked in a year. Workers reporting earning less than $3.50 per hour or more than $500 per hour in inflation-adjusted dollars were omitted from the analysis, as were workers reporting earning less than $100 per week or more than $5,000 per week.

A worker is classified as covered by employer-provided health insurance if the worker is a policy­holder for employer-provided insurance or is a dependent covered by employer-provided insur­ance. To make data gathered since 2001 comparable with earlier data, workers were excluded from health coverage analysis if their insurance status was identified using the verification question intro­duced in 2001.

Appendix C

Data on the characteristics of minimum wage workers comes from August 2007 through April 2008 Current Population Survey outgoing rotation groups downloaded from the Census Bureau. Workers were defined as being at or below the min­imum wage if they reported that they were paid hourly wages and that their usual weekly earnings divided by their usual hours worked per week was less than or equal to $5.85 an hour. This measure of earnings includes tip income, which raises the wages of many restaurant employees above the min­imum wage.

A part-time worker is one who reports that he or she usually works part time or who is currently part time, despite usually working fulltime. A single par­ent is defined as a worker who reports that he or she has one or more of his or her own children present in the household and who is either widowed, divorced, or separated or has never married.

[1] Elise Gould, "The Erosion of Employment-Based Insurance: More Working Families Left Uninsured," Economic Policy Institute Briefing Paper No. 203, November 1, 2007, at  (May 12, 2008).

[2] Senator Barbara Boxer (D-CA), "Floor Statement on the Employee Free Choice Act," June 20, 2008, at  (May 12, 2008).

[3] News release, "Job Openings and Labor Turnover: January 2008," U.S. Department of Labor, Bureau of Labor Statistics, March 12, 2008, at  (May 12, 2008). This is not identical to the number of new workers over this time because the survey measures new hires and job separations, not the status of indi­vidual workers. A student who worked part time as a waiter, left that position to become a tutor, and then worked as an engineer after receiving his degree would represent three new hires and two job separations in this survey.

[4] Heritage Foundation calculations based on data from U.S. Census Bureau, United States Census, 1980, 1990, and 2000, and American Community Survey, 2006. Figures include all workers ages 25 to 65 who worked at least 1,250 hours in the previous calendar year. For more details, see Appendix A.

[5] In 1980, 342,000 people ages 25 to 65 were working at least 1,250 hours per year as housekeepers, maids, butlers, stew­ards, or lodging quarters cleaners. By 2006, that number had risen to 860,000, a net increase of 518,000 workers. A net of 378,000 workers of that increase were born outside the United States. Many immigrants work as housecleaners for wages below what most Americans will accept. This makes domestic services affordable for middle-class Americans and improves the quality of life for many busy two-earner families. Without immigration, most of those jobs would not go to Americans. They would not exist, and those families would not be able to afford housekeepers.

[6] Brahima Coulibaly, "Changes in Job Quality and Trends in Labor Hours," Federal Reserve Board International Finance Discussion Paper No. 882, October 2006.

[7] Ibid., Table VII.

[8] Ibid.

[9] David H. Autor, Lawrence F. Katz, and Melissa S. Kearney, "The Polarization of the U.S. Labor Market," American Economic Review, Vol. 96, No. 2 (May 2006), pp. 189-194, Figure 3.

[10] Heritage Foundation calculations based on data from U.S. Census Bureau, United States Census, 1980, 1990, and 2000, and American Community Survey, 2006. The calculations included all workers ages 25 to 65 who worked at least 1,250 hours in the previous calendar year and excluded workers who reported being born outside the United States. The col­lege wage premium is defined as the ratio of the median wage of workers with a bachelor's degree or greater to the median wage of high school graduates. For more details, see Appendix A.

[11] For example, the Census requests information on a worker's earnings in the previous calendar year, so a worker inter­viewed for the 2000 Census reported his income in 1999. The ACS asks about a worker's earnings over the past 12 months, so a worker interviewed in July 2000 reported his income from July 1999 to July 2000.

[12] The CPS was redesigned in the early 1990s, and the changes went into effect in 1994, affecting responses to questions about income in 1993. For many questions, data from before the redesign are not strictly comparable with data from after the redesign.

[13] Heritage Foundation calculations based on Current Population Survey, August 2007-April 2008. For details, see Appendix C.

[14] Ibid. Of the workers earning the federal minimum wage, 6.3 percent are single parents working fulltime. This represents 95,300 of 146 million American workers.

[15] James Sherk, "Minimum Wage Workers' Incomes Rise When the Minimum Wage Does Not," Heritage Foundation WebMemo No. 1181, July 28, 2006, at

[16] Heritage Foundation calculations based on data from U.S. Department of Labor, Bureau of Labor Statistics, Employer Costs for Employee Compensation, Table 1. In the fourth quarter of 2007, benefits comprised 30.2 percent of the average employee's total compensation.

[17] The Census and ACS do not ask about health insurance coverage. Current Population Survey (CPS) health data after the redesign are not strictly comparable with data from before the redesign. The redesigned survey utilizes computer-assisted interview techniques that allow interviewers to probe respondents more thoroughly. This reduced the proportion report­ing no health insurance by several percentage points, creating a spurious upward trend in insurance rates. Consequently, this paper uses only CPS data gathered after the redesign.

[18] Heritage Foundation calculations using data from U.S. Department of Labor, Bureau of Labor Statistics, Employer Costs for Employee Compensation (ECEC), compiled by Haver Analytics. Data cover all civilian workers from the first quarter of 1995 to the first quarter of 2007. Data were adjusted for inflation using the PCE chain price index. The ECEC starts in 1991. From the first quarter of 1991 to the fourth quarter of 2007, average employer spending on health benefits per hour of work rose by 52 percent.

[19] Heritage Foundation calculations using data from U.S. Department of Labor, Bureau of Labor Statistics, Current Popula­tion Survey, 1994-2007, March supplement. For more details, see Appendix B. Health insurance coverage rates for all workers were 78.8 percent in 1996, rose to 80.7 percent in 2000, and fell to 76.9 percent in 2006.

[20] Geoffrey Sanzenbacher, "Estimating Pension Coverage Using Different Data Sets," Boston College, Center for Retirement Research Issue in Brief No. 51, August 2006, at  (May 12, 2008).

[21] "While pension participation has risen and fallen over short intervals, all four data sets show stable participation rates over the period 1991-2004." Ibid., p. 3.

[22] Heritage Foundation calculations using data from U.S. Department of Labor, Bureau of Labor Statistics, Employer Costs for Employee Compensation, compiled by Haver Analytics. Data cover all civilian workers from the first quarter of 1991 to the fourth quarter of 2007. Data were adjusted for inflation using the PCE chain price index.

[23] AFL-CIO, "Retirement Security," Resolution 30, 2005 convention, at /static/reportimages/30D732930957C132903142117C9AFB4F.pdf (May 12, 2008), and AFL-CIO, "Defined-Benefit Pensions," at (May 12, 2008).

[24] Ashwini Agrawal, "Corporate Governance Objectives of Labor Union Shareholders: Evidence from Proxy Voting," University of Chicago, Graduate School of Business Working Paper, March 10, 2008, p. 28, at /static/reportimages/90C113034A2953CE18B32E10AB71CABB.pdf (May 12, 2008).

[25] See Andrew A. Samwick and Jonathan Skinner, "How Will 401(k) Pension Plans Affect Retirement Income?" American Economic Review, Vol. 94, No. 1 (March 2004), pp. 329-343, and James Poterba, Joshua Rauh, Steven Venti, and David Wise, "Defined Contribution Plans, Defined Benefit Plans, and the Accumulation of Retirement Wealth," Journal of Public Economics, Vol. 91, No. 10 (November 2007), pp. 2062-2086.

[26] For example, two workers at a firm receive a defined-benefit pension that pays out 1.5 percent of final wages per year of service. One worker stays with the company for 40 years and retires, earning $60,000 per year in his final year. His defined-benefit pension would pay him $36,000 per year (0.015 * 40 * $60,000). The other worker leaves his job after 20 years to work for another company that has the same pension structure. The wage profile is identical at both firms. When he left the first company, he earned $40,000 per year, and he retires earning $60,000 in his final year. However, this worker would receive only $30,000 per year (0.015 * 20 * $40,000 + 0.015 * 20 * $60,000). Switching employers halfway through his career to take a job paying the same amount reduces the second worker's pension by 17 percent.

[27] Samwick and Skinner, "How Will 401(k) Pension Plans Affect Retirement Income?" and Poterba et al., "Defined Contri­bution Plans, Defined Benefit Plans, and the Accumulation of Retirement Wealth."

[28] For a discussion of working becoming less unpleasant, see Coulibaly, "Changes in Job Quality and Trends in Labor Hours."

[29] The Current Population Survey does not provide data on paid time off work. These data come from the Employee Bene­fits Survey, a separate survey conducted by the Bureau of Labor Statistics. These data survey establishments, not individ­uals, so it is not possible to separate out the effects of increased immigration on paid leave trends in this chart.

[30] Heritage Foundation calculations using data from U.S. Department of Labor, Bureau of Labor Statistics, Employer Costs for Employee Compensation, compiled by Haver Analytics. Data cover all civilian workers and were adjusted for inflation using the PCE chain price index.

[31] For example, see Louis Uchitelle, The Disposable American: Layoffs and Their Consequences (New York: Alfred A. Knopf, 2006).

[32] See David Neumark, "Changes in Job Stability and Job Security: A Collective Effort to Untangle, Reconcile, and Interpret the Evidence," National Bureau of Economic Research Working Paper No. W7472, January 2000, and David A. Jaeger and Ann Huff Stevens, "Is Job Stability in the United States Falling? Reconciling Trends in the Current Population Survey and the Panel Study of Income Dynamics," Journal of Labor Economics, Vol. 17, No. 4 (October 1999), pp. S1-S28.

[33] Men are examined to focus on changing economic conditions, not broader social forces that have increased women's attachment to the workforce.

[34]  James Sherk, "Job-to-Job Transitions: More Mobility and Security in the Workforce," Heritage Foundation Center for Data Analysis Report No. 08-06, September 2, 2008, at

[35]  See also Steven Allen, Robert Clark, and Sylvester Schieber, "Has Job Security Vanished in Large Corporations?" in David Neumark, ed., On The Job: Is Long-Term Employment a Thing of the Past? (New York: Russell Sage Foundation Publications, 2000).

[36] Alicia H. Munnell, Kelly Haverstick, and Geoffrey Sanzenbacher, "Job Tenure and the Spread of 401(k)s," Boston College, Center for Retirement Research Issue in Brief No. 55, October 2006, at /static/reportimages/BB430D5AB8B5D24F79F651B3D6C10AB5.pdf  (May 12, 2008).

[37] Steven Ruggles, Matthew Sobek, Trent Alexander, Catherine A. Fitch, Ronald Goeken, Patricia Kelly Hall, Miriam King, and Chad Ronnander, Integrated Public Use Microdata Series, version 3.0, Minnesota Population Center, at (May 16, 2008).

[38] Miriam King, Steven Ruggles, Trent Alexander, Donna Leicach, and Matthew Sobek, Integrated Public Use Microdata Series-Current Population Survey, version 2.0, Minnesota Population Center, 2004, at  (May 16, 2008).

About the Author

James Sherk Research Fellow, Labor Economics
Center for Data Analysis