Work, work, work. From the moment we finish school, until the day our first Social Security checks are cashed, our schedules are structured by the relentless rhythm of Monday-to-Friday, nine-to-five. Sure, there will be brief interruptions for things like the birth of a child or a change of employer—sometimes through layoff, sometimes by choice—but, mostly, we think of our adult lives as being all about work.

It’s a commanding image. But it doesn’t fit the facts. Instead, according to findings released Tuesday by the Bureau of Labor Statistics, from ages 18 through 48, a period which includes some of our prime working years, the typical American is employed just about three out of every four weeks. Whether or not we realize it, the average adult spends about a week each month without a job. (Note: We’re talking about employment, not “working” per se, so this measure doesn’t include time off for vacations, weekends, or other time away from the office.) It’s a far cry from always being on the clock.

Of course, not working isn’t necessarily a bad thing. When we’re 18 to 24 years old, for example, we work just two-thirds of the time, because many of us are still going to school. But even after age 30, we tend to work just four out of every five weeks—and that’s where patterns of employment that are surprising become demonstrably unfair. As it turns out, just how much time we spend not working and not earning a paycheck depends strongly on several factors that are beyond our control.

A Clearer Picture of Working America

These data come from the latest round of National Longitudinal Survey of Youth (NLSY), an ambitious research project that has tracked the employment and life experiences of a representative sample of Americans born between 1957 and 1964 since 1979. Members of the sample are interviewed every two years—Tuesday’s BLS report is based on the twenty-fifth such round of interviews—and it provides a unique portrait of how the fortunes of Americans, in the labor market and otherwise, evolve.

Because the same individuals are questioned at multiple points over their lives (thus “longitudinal”), the NLSY allows researchers to answer questions about cause and effect that more familiar surveys, such as the one used to calculate the unemployment rate in Friday’s upcoming jobs report, cannot address. The NLSY has been used in many well-known studies, including the 2008 paper in which Anne Case and Christina Paxson, both then of Princeton, found that taller people earn more because height is associated with cognitive ability.

Which brings us back to our puzzle: why do Americans work less than they think? The BLS report itself doesn’t attempt to identify causality, but it does present several instructive correlations.

Education Matters

When it comes to employment, education matters—a great deal. During the first 30 years of their careers, high school dropouts in the NLSY cohort were employed just 60 percent of the time, or roughly 18 years in total. How were the other 12 years spent? Well, for eight percent of the time (in other words, 2.4 years), they were actively looking for work and counted among the officially unemployed, which means that for the balance—9.6 years—they were out of the labor force entirely. That’s nearly a full decade spent idle.

Though people with “only” high school degree are often lumped together with dropouts, graduating actually confers a large benefit in terms of employment regularity. High school grads were employed for 77 percent of weeks between their 18th and 48th birthdays. The advantage of a college education is larger still, with bachelor’s degree holders having held a job 84 percent of the time.

To view this image larger click here

That education influences labor market outcomes is something we tend to accept as fair, on the basis that those who work harder and do better in school deserve the employment opportunities they receive. But education isn’t the only characteristic impacts job prospects—and some of these other factors enter into the employment equation less equitably than education (to say nothing of the fact that access to quality education or the home environment to support it are themselves asymmetrically distributed.)

Sex Matters

At every age, the women of the NLSY were less likely to be working than men. The gap for the cohort peaked when they were between 25 and 34 years of age, at which point men worked, on average, about nine more weeks each year than did women. Partly, this imbalance reflects the continued influence of gender-skewed child-rearing norms. These women entered the working world in the 1980s, before the rapid rise in female labor force participation reached it’s turn-of-the-century apogee; today, the gap would be smaller. Nevertheless, even as this cohort reached their 40s, which occurred between 1997 and 2012, men continued to be employed five more weeks a year than women. However much we might regard stay-at-home moms as the stuff of Life magazine features, it remains true that women’s career trajectories are more likely to be fettered by homemaker expectations.

Race Matters

While whites (of both sexes and all education levels) were employed an average of 80 percent of the time between their 18th and 48th birthdays, Hispanics’ employment timeshare was just 72.4 percent, and that of blacks was 68.7 percent. These gaps become even more pronounced at lower levels of education: black high school dropouts worked just 47.7 percent of the time—that’s equivalent to 14.3 years out of 30—compared with 63.7 percent among white high school dropouts. The discrepancy for high school graduates is nearly as large: 65.9 percent employment for blacks versus 79.6 percent for whites.

To view this image larger click here.

If there’s a silver lining in the all-too-familiar pattern of race-based advantage, it’s that the gap attenuates with education. Among college graduates, blacks (83.2 percent) and whites (83.8 percent) spend a similar share of their time employed, and college educated Hispanics actually spend slightly more time working (84.7 percent).

Less Employment over Time Means Lower Wages

Not surprisingly, these inequalities in employment translate into earnings inequality. Among the NLSY cohort—which, remember, is representative of Americans born from 1957 to 1964 and living in the United States in 1979—those with a college degree saw their real hourly earnings grow 241 percent between ages 18 and 48. But for high school dropouts, earnings growth was just 54 percent. Similarly, men’s hourly earnings grew considerably faster than women’s (143 percent versus 113 percent), and whites’ (135 percent) grew faster than blacks’ (92 percent) and Hispanics’ (122 percent). And hourly earnings are only part of the story; given that the same groups that are paid more (whites, men, and the college educated) also work more hours, the disparities in income are larger still. (Note: These earnings growth figures are larger than the ones you’re used to hearing because we’re looking at the average changes an individual experiences over her career rather than growth in average earnings for the population as a whole, which is more typically quoted.)

To view this image larger click here.

What This Means for Progressive Policy

These patterns, accumulated from the work experiences of the people at the core of today’s labor force, present significant challenges for policy.

On key demographic dimensions—characteristics like sex and race, over which we have no control—the labor market is an unfair place. To the extent that these employment and wage gaps are the result of discrimination, we must do more to proactively combat prejudices, both overt and unconscious. To the degree they reflect childhood disadvantage, we must expand access to prized resources, such as education and career networks, that are a increasingly a prerequisite to good jobs.

And we must embrace the compassion to protect those who fall through the cracks. Dropping out of high school is a poor decision, yes, but mistakes made at 18 (and often not entirely of one’s volition) ought not to be a condemnation to a life of poverty. The less skilled deserve to work with dignity, and be paid decently.

It’s also worth noting that not all labor market challenges come down to inequality. Another lesson from the latest NLSY data is that our careers—all of our careers—are more tenuous and less predictable than we think. The average member of the NLSY generation worked 11.7 jobs in their first 30 years of work. Much of the instability came in their early-20s, but in their 30s they continued to change jobs four times, on average, as well. Indeed, even in their 40s, just a third of jobs lasted more than a year, and only half lasted more than two.

So not only do we need policies mindful of the challenges faced by the working poor; we also need a more robust safety net to help us all navigate the inevitable, yet unpredictable, headwinds our careers will encounter. The Affordable Care Act is a great example. By decoupling affordable health insurance from employment, it ensures Americans who lack jobs don’t also lack doctors.

But more remains to be done. America is the only advanced democracy that doesn’t guarantee paid family leave, a critical support for working families (and an important way to advance workplace equality for women). Our unemployment insurance and worker retraining programs also come up short, as evidenced by the still-pronounced rate of long-term unemployment we’ve seen since the Great Recession. Far too many Americans have something to contribute to our economy, but are left sitting on the sidelines. Likewise, far too much opportunity to re-engage these workers is squandered for the sake of blind allegiance to the myopic gospel of unfettered individual self-determination.

And that’s the added perspective a decades-long dataset like the NLSY gives us. It’s in our best interest to make the labor market less firm, more fluid, and. perhaps above all, more forgiving—not only because harnessing currently untapped potential will improve life for all of us, but also because, chances are, at one point or another, we’ll be the ones yearning for the exhausting familiarity of nine-to-five.