In this 2011 photo, Sacramento County Deputy Sheriff Chris Carroll opens a cell at a formerly closed housing unit at the Rio Cosumnes Correctional Center, in Elk Grove, Calif. (Rich Pedroncelli/Associated Press)

Nicholas Eberstadt holds the Henry Wendt Chair in Political Economy at the American Enterprise Institute.

The United States today is home to two huge but essentially invisible populations. Each of them is widely stigmatized and largely composed of people living in the shadows. The government does not know who they are, where they are or how well they are doing.

The first of these invisible tribes — illegal immigrants — at least has attracted more than passing comment in politics. By contrast, America’s second invisible caste is almost never mentioned. Yet this group is far larger than the unauthorized immigrant population, and it is made up almost entirely of U.S. citizens.

I refer to our vast underground army of released felons — adult men and women convicted of serious criminal offenses for which they have been punished with prison time or probation, and who now form part of the general population. So hidden from public sight is this vast army, indeed, that many Americans are unaware of its existence.

Most well-informed readers know that the number of Americans behind bars has soared since the early 1980s and that the United States has a higher share of its populace in jail or prison than virtually any other country. But only a tiny fraction of Americans who have been convicted of a felony are incarcerated. Perhaps 90 percent of all sentenced felons are out of confinement and living more or less among us.

How can that be? To begin: Few felons are sent away for life. According to the Justice Department’s Bureau of Justice Statistics, the average time that imprisoned first offenders serve in state penitentiaries is just more than two years. More than 600,000 convicts are released from prison every year, and despite high rates of recidivism, many do not return. In addition, many convicted felons are never confined in the first place; instead, they undergo “community supervision” (such as probation). Taken together, correctional release, parole and probation guarantee a steady annual flow of convicted felons into society.

What sort of totals are we talking about? Curiously, there seem to be no official estimates. Some researchers, however, have attempted to determine the approximate dimensions of this invisible population — and their findings may astonish.

In two studies on the demography of what they call our “criminal class,” professors Christopher Uggen, Melissa Thompson and five colleagues estimate that the cohort of incarcerated and released felons in the United States had reached nearly 20 million by 2010 — four times larger, in their estimate, than just 30 years earlier.

If this estimate is even roughly accurate, and if the United States’ total felon population has continued to grow at more or less the same tempo the researchers cited for 2004 to 2010, we would expect the number of convicted felons to surpass 23 million people this year. That would be roughly twice as high as the number of illegal immigrants in the country. And since the combined U.S. jail and prison population is about 2.2 million (including some non-felons sentenced to jail or awaiting trial there), these figures would suggest the number of non-institutionalized Americans with a felony conviction will almost certainly exceed 20 million by the end of the year.

If America’s non-institutionalized felon population today were a state, it would be the third largest in the country — about the same size as Florida, and larger than New York. The adult population of this “state” would be the country’s second largest — nearly tied with Texas. And its adult male population would be by far the nation’s biggest — at least 5 million ahead of California. By the same token: If released felons were regarded as a minority, their numbers would well exceed the size of our Asian American population.

Given this reality, one might think policymakers would have an interest in knowing at least a little about this major segment of our population. Wrong: To judge by the data our democracy collects, the circumstances of this ex-con population are a matter of almost complete indifference to the rest of us. These individuals show up only in our statistics on crime and punishment — in other words, when they run afoul of the criminal justice system.

We don’t know how many children they have, their marital status, who they live with, their housing situations. We don’t know their mortality rates or life expectancy, their disease and disability profile, their mental-health status. We do not know their labor force participation rates, unemployment rates, jobs by sector or wages. Apart from broad generalities, we know roughly nothing about their education patterns, skills or training.

The irony here is not that felons who have paid their debt to society have need of a largely indifferent public: It is that this same public needs them, too. We need them to succeed: as fathers and mothers, as breadwinners, as citizens — as people who make the most of a second chance. Our society can’t hope to flourish with 20 million modern-day outcasts in our midst.

Given its sheer scale, the task of reintegrating reformed felons has never been more important than it is today. But thanks to officialdom’s statistical neglect, we haven’t a clue about how well this task is working. And we can’t gather evidence to learn what we could do to make “re-entry” work better either.

Our government is perfectly capable of compiling key facts and figures about conditions for Americans with felony convictions. Some of this could be done easily, quickly and at very low cost; other aspects would take more time, money and technical effort. But it’s all eminently doable. All we need is the political desire — and social compassion — to see it done.