People who pay attention to poverty, including the poor themselves, know one thing all too well: Over the past few decades, anti-poverty policy in this country has evolved to be “pro-work.” This means that if you’re a low-income parent who’s well connected to the job market, the government will help you in a variety of ways. But, if you’re disconnected from the job market, public policy won’t help you much at all.

How do people in that second group survive?That’s a question that Kathryn Edin and H. Luke Shaefer, a sociologist and a social-work professor, answer in their new book, $2.00 a Day: Living on Almost Nothing in America. It is, as the title suggests, a devastating portrait of families struggling to get by on impossibly low incomes.

A few of their strategies: availing themselves of charities and public spaces (like libraries), selling food stamps for cash (illegal, and they typically get just 60 cents on the dollar on the street), relying on relatives (who can be as hurtful as helpful), selling scrap metal or aluminum cans, selling plasma (which involves considerable angst as to whether a person’s blood’s iron levels are sufficiently high, especially difficult around menstruation), receiving some public support (housing vouchers, nutritional support, disability payments), occasionally holding a job, and—the most common strategy of all—just going without.

It is important to recognize that what Edin and Shaefer call $2-a-day poverty doesn’t mean that their subjects really survive for long periods on nothing but $2 a day, and I fear that too many readers will be thrown off by this distinction. Do not go there, as it’s an unnecessary distraction. The authors explicitly acknowledge that no one could survive in this country if that was all they had to live on over an extended period. What they call “$2-a-day poverty” means spells of scraping by on almost no regular, reliable income, though many may be able to access the dicey income sources just noted above.

Edin and Shaefer calculate that “1.5 million households with roughly 3 million children were surviving on cash incomes of no more than $2 per person, per day in any given month” in 2011. That’s about 4 percent of all families with kids, though once you start adding other resources, like the value of SNAP (Supplemental Nutrition Assistance Program, more commonly known as food stamps), that percentage declines. Though blacks and Hispanics are disproportionately likely to be deeply poor, half are white.

The book is also a policy critique, one which carefully follows the role of President Bill Clinton’s 1996 welfare-reform initiative in the growth of deep poverty. Far from a dry recounting, Edin and Shaefer turn this history into a powerful story of the collision of politics, economics, and the real lives of poor families with kids.

Both for political and substantive reasons, candidate Clinton made the reform of cash welfare, the receipt of which wasn’t much conditioned on work, a key plank in his platform. As Edin and Shaefer stress, this was not a wholly conservative or even centrist position. Progressive scholars recognized that work had to be a ladder out of poverty, and were thinking about ways to facilitate that upward climb. Notably, this view is widely held by the working-age people in Edin and Shaefer’s sample. They want decent, steady jobs, and not just because the recognize work as an economic necessity but because of the dignity they believe it will bring to their lives as people and as parents.

There are numerous things that need to be in place for that to happen, i.e., for the deeply poor to be able to work their way up. First, there have to be enough jobs paying livable wages. Second, work supports, including affordable child care, are necessary. And third, any skill deficits must be closed. As one might expect, many of the subjects in the book have few marketable skills, but not all. One woman, a cashier at Walmart, handily memorized four-digit bar codes on dozens of common items. Based on her numerical acumen, she’s heading for a management position when she gets fired because one of the people she lives with used up all the gas and there was no money left to fill the tank.

Another worker they profile had a job cleaning foreclosed homes for the city of Chicago. Clean water is necessary for cleaning houses, but the city had shut the water off in many of these properties. So the woman and her co-workers come up with various crafty schemes to get and transport water from neighbors, gas stations, and restaurants. In other words, we’re talking about some people who are willing to do some hard work for not a lot of money.

Fortunately for those with jobs, work-based, time-limited welfare reform was accompanied by some real gains in the way of work supports, particularly through the expansion of the Earned Income Tax Credit, a wage subsidy that can add over $5,000 dollars to the annual income of a full-time working parent with two kids, though the average is a bit less. Unfortunately, policy makers completely dropped the ball on the jobs part of the equation. Essentially, and amazingly, to anyone who knows the market failures that pepper the low-wage labor market, the new welfare law assumed the jobs would be there.

Edin and Shaefer argue that “although the 1996 welfare reform pushed millions of low-income single moms into the workforce, it did nothing to improve the conditions of low-wage jobs.” They further note that the EITC expansion partially exacerbate the problem, by further increasing the supply of labor, thus putting downward pressure on wages.

But they miss the other critical development of those years, one that many of us who raised these labor-supply concerns didn’t see coming: the strong increase in labor demand, as the late 1990s ultimately became a unique period of truly full employment. For the first time in decades, real low-wages grew at the rate of productivity. In the early 1990s, poverty among black mother-only families with kids was around 60 percent. By the end of the 1990s it was around 40 percent, a massive decline over such a short time period.

That’s still far too high, of course—the lasting disadvantages to children growing up in poverty are a key theme in $2 a Day—but it is a telling guidepost of what it will take to address a huge flaw in welfare reform, i.e., the insistence on work without regard to job availability. In the latter 1990s, the increased supply of low-wage workers, many of whom were poor single moms, was induced by welfare reform. But this supply shock was met with an even sharper increase in quantity and quality of low-wage jobs, the likes of which we haven’t seen since.

And what happened? Many of these women went to work, their earnings rose, and their poverty rates fell. But this period of a welcoming low-wage labor market turned out to be but a brief oasis in what, as $2 a Day reveals, is once again a desert of diminished opportunity, especially for the least advantaged with little in the way of work supports, most notably, child care.

I cannot overemphasize the importance of this fundamental flaw in poverty policy, i.e., the assumption that there is an ample supply of perfectly good jobs out there that poor people could tap if they just wanted to do so. To this day, this misguided notion underlies the conservative policy agenda that views anti-poverty policy as a narcotic that weans people away from the jobs awaiting them. Kill the programs, and they’ll get out of their hammocks (Rep. Paul Ryan’s term for the safety net) and get to work.

Edin and Shaefer, informed by reality as opposed to such ideology (or just wishful thinking), propose to correct that flaw as the first principle in their three-pronged approach to ending $2-a-day poverty. The second principle regards housing: “parents should be able to raise their children in a place of their own.” The third recognizes that “not every parent will be able to work … all of the time, but” family well-being must “nonetheless be ensured.”

Clearly, if America’s anti-poverty policy framework is founded on work in the paid labor market, and if that labor market doesn’t provide the necessary quantity and quality of jobs, public policy must make up the difference. If this sounds radical, consider the following: Policy makers are quite comfortable spending literally trillions of dollars to reflate credit markets when they fail, based on the notion (a sound one, I’d argue) that the economy cannot function without adequate credit flows. Well, neither can it function without enough jobs, so if the Federal Reserve is the “lender of last resort,” then the government must be “the employer of last resort.” Edin and Shaefer smartly propose to scale up a useful employment program that worked well during the Recovery Act.

Edin and Shaefer teach us that we in America live amid deep poverty, privation of a level that would be perfectly at home in an emerging economy with a per-capita income that’s a fraction of our own. Yet we also live in an era of political dysfunction, where a large and dominant part of our politics insists that reduced government is better government, and any expenditures to help the poor merely blocks them from picking themselves up and taking advantage of the fulsome set of opportunities they’d face if only they sought them out.

It is simply impossible to read $2 a Day and maintain such naïve thinking. What happens next is up to the American public and, given the lives to which Edin and Shaefer have introduced us, doing nothing should not be an option.