Hardly any idea about the United States economy is as entrenched as the one about how old-fashioned, long-term jobs have disappeared. No longer can the new college graduate step into his wingtips and expect to wear them out by climbing the career ladder at one solid company, or at a steadily growing small business.

The disappearance of these reliable jobs is one of the bedrock political crises of this election cycle. “The old model of work, where you could expect to hold a steady job with good benefits for an entire career, is long gone,” Hillary Clinton said in an economic-policy speech last week. “People in their twenties and thirties have come of age in an economy that’s totally different.”

It’s commonplace to say that the sepia-toned America of those better days never really existed. Still, it bears saying: it never existed. Clinton gives voice to a nostalgia that misrepresents how eager the young people of the past were to start “stable” jobs.

Moreover, our collective nostalgia misrepresents historical job security so completely that it gets it close to backward. We imagine a past where everyone had thirty-year careers (or, less pretentiously, jobs), tapering off into a work twilight and then retirement. This memory is surprisingly at odds with the data: the typical worker now stays at a job six months longer than the average worker did a decade ago. Taking an even longer-term view, the typical worker has stayed at the same job for more than four and a half years, versus just three and a half years in 1983.

Whether that increase in stability is wholly positive is arguable. In roaring economies, workers switch jobs more often, looking for higher pay or better bosses. The length of time spent at one job goes up in times of economic stress (such as the mid-aughts), when workers hang on for dear life. But whatever the cause, it’s clear that younger workers switch jobs less often than in the past. For women, also, the length of time at the average job has gone up markedly. (It’s now almost the same as for men.) To the extent that there was security in the past, it didn’t apply to women.

One group, however, has suffered in terms of job stability. You can probably guess which one: men in the later stages of their careers. The share of men older than fifty-five who have been at their jobs for twenty years or more has plummeted. In 1983, almost forty-five per cent of men in their late fifties and early sixties had been at their jobs for twenty years or more. Now that number is about thirty-one per cent, meaning that fewer than a third of men keep jobs that they got in their prime into their late working years. If you leave out government jobs, the numbers would likely look worse. Meanwhile, retirement comes later, if at all, and the share of Americans who stay in the workforce after the age of fifty-five is increasing. The pay raises that at one time may have been rewards for a job well done are now a reason to reëvaluate whether a worker is worth his salary, or might better be replaced with someone who comes with lower costs and newer skills.

Job security enters the national discussion as a problem of youth, a perception that opportunities of the past have been lost and that it is impossible to find a long-lasting job in the age of the “gig economy.” But the real problem of job stability is conspicuously a problem of later age, and its image is not really the younger worker looking for a good start but the older worker trying to re-start a failing career or hopping though Amazon warehouses. Talking about job insecurity is a way of talking about the fear—especially among men—of ending up useless and discarded, with too many years of work behind them and too many years of emptiness ahead.

There is reason for this fear. Security in older age is a recent luxury. Before the Second World War, the spectre of old age was closely tied to the fear of abject poverty. In the second half of the twentieth century, increasing national wealth, Social Security, and pensions allowed those fears to be bottled up and mixed into a more appealing cocktail of retirement. Now those fears of becoming obsolete—or, worse, spent—have metastasized and returned for American men in their later years. The experience of job loss in late middle age is a common one, and often difficult to recover from. Reinvention clearly happens a fair amount. If you look at the data, there are plenty of sixty-five-year-old men, for instance, who have been at their jobs for between ten and fifteen years, meaning that they found something new to do after the age of fifty. But even in such cases many probably give up the expectation of an income that rises steadily until retirement.

Our national response to this is much like the masculine response to many things. On the right, it takes the form of a largely inchoate anger that looks for a screamer-in-chief. Donald J. Trump has responded by calling for ripping up international trade deals. Clinton’s response is that we’ll power through and get the economy working at greater speed.

Without a doubt, better economies are better for everyone, for old workers as well as young. But right now unemployment is at less than five per cent, the median length of time at a job is probably as high as it has been in fifty years, and we are still talking about a crisis of stability. The numbers on job tenure show that we are not facing the crisis that everyone seems to think we face. There is no disappearance of long-term jobs, no crisis of opportunity for the young, no replacement of regular jobs with short-term stints. What has substantially diminished is the expectation that workers, and especially men, can coast from the fifteenth year of a job to the twenty-fifth year. That shift in social norms elicits an intense, and largely justified, anxiety in the middle-aged. Older voters talk about their fears about what will become of their children just graduating into the workforce, but it’s also a way of saying that they fear what will become of themselves.

Mark Gimein is a New York-based writer and editor.