Here are two surprising facts about welfare and poverty in the United States: (1) we are spending an enormous amount of money on people at the bottom of the income ladder and 2) all that spending does a very poor job of meeting human needs.

And here is a surprising opportunity: if we took all of the money we are currently spending on anti-poverty programs and gave it in cash to poor families, there would be no problem of poverty in this country

We might still have some homelessness—reflecting mental illness or drug abuse. But conventional poverty would be a thing of the past.

Calculating the Resources Families Have

In their book The Myth of American Inequality former senator Phil Gramm and his colleagues point out that determining who is poor and how poor they are is quite complicated. Here is how they did it.

First, they divide the population into quintiles, based on household earned income (mainly wages). They then calculate each quintile’s total income after all taxes and all transfer payments. Transfers are Medicaid, food stamps, housing subsidies, etc.; and there are 95 such programs in all. They count each dollar of transfer spending as equivalent to a dollar of earned income.

The economists were then able to determine how much is available for consumption in each of the quintiles. Because household size differs from quintile to quintile, they divided that amount by the number of people in each household to get per capita “income.”

Comparisons on the Income Ladder

The first important finding: the bottom fifth of households in 2017 had an average (after tax and after transfer) income of $33,653 per person. As I show below, almost all of this “income” is in the form of noncash welfare benefits. But if all those benefits were converted to cash, a family of four in the bottom fifth of the (earned) income distribution would have $134,652 a year to spend, after taxes!

The second important finding: The per capita income of second fifth in 2017 was $29,497; and for the middle fifth it was $32,574.

Those with the least earned income had more actual total income than those in the next two higher quintiles! The average household in the bottom fifth received 14 percent more income than the average second-fifth household and 3.3 percent more than the average middle-income household.

So, if people in the bottom fifth of the income distribution are consuming more each year than those who are supposedly better off, why is there a problem?

Why Cash Matters

Most of the income received by middle-income families is cash income that people earn in the labor market. With their after-tax income, people are able to spend in ways that maximize their well-being.

That’s not true at the bottom of the income ladder.

Since the War on Poverty started in 1965, the labor force participation of the bottom one-fifth of households has dropped from 70 percent to 36 percent. As a group, this one-fifth now receive more than 90 percent of their income from government. For this group, our welfare system has substituted in-kind benefits for labor market income.

Medicaid money goes to doctors, hospitals, and other suppliers of medical services. Food stamp money goes to agribusiness. Education subsidies go to teachers and the public-school bureaucracy. Housing subsidies go to landlords. Throughout most of its history, War on Poverty programs gave very little cash to poor people.

It doesn’t take too much imagination to understand that if you give money to impersonal bureaucracies, it is more likely to be spent in ways that benefit bureaucrats rather than in ways that benefit poor families.

Studies show that Medicaid enrollees value Medicaid as little as 20 cents on the dollar. That means that they would in principle sell their Medicaid insurance if you offered a cash payment equal to as little as one fifth of the program’s actual cost.

Falling through the safety net

The other day I Googled “emergency medical care near me” and got two dozen or so nonhospital options near my home in Dallas, Texas. Not one of them had the word “Medicaid” in their initial message. I understand that some of these facilities will take Medicaid. But none of them are located in the part of the city where most Medicaid enrollees live. Obviously, they are not anxious to attract Medicaid patients.

About the same time as my Google search, an article in the Dallas Morning News described South Dallas as a “medical desert,” despite the fact that this is the area of the city with the highest prevalence of chronic disease. In a normal market, venders go where the need for their product is greatest. But neither private nor government medical service providers have any interest in South Dallas.

To meet the health needs of this area, a nonprofit entity has been established.

There are an estimated 1,400 federally funded community health centers in the United States, serving the health needs of one in every 11 people in the country. These are mainly, people who are uninsured or on Medicaid. Moreover, at the very same time that the federal governing was expanding community health centers, states were also expanding Medicaid enrollment as a result of the Affordable Care Act.

Although it was predicted that Medicaid enrollment would reduce the use of emergency room use, the opposite has been true. Medicaid insurance causes people to increase their emergency room visits by 40 percent.

Nationwide, the average time a person can expect to spend in an emergency room is a little more than 3 hours. At a safety net facility like Parkland Hospital in Dallas it is almost 6 hours.

The number of community health centers is roughly equal to the number of private walk-in clinics, such as MinuteClinic in CVS pharmacies. Yet, if MinuteClinics can meet the needs of the nonpoor, why can’t they do the same for the poor? And if ordinary insurance can give the non-poor all kinds of options in the medical marketplace, why can’t Medicaid do the same for the poor?

Another article in the Dallas Morning News announced that one in five Texas children experienced hunger, and bemoaned the fact the state was missing out on a $450 million federal grant for summer-time food for kids because of bureaucratic inertia.

But why do we even need that program? Remember, the family of each of these kids is supposed to have the equivalent of $33,653 per year to spend on the child's needs—with 90 percent of it provided by government!

A Better Way

Writing in the New York Times, MIT economist Amy Finkelstein proposed an alternative. Take the $600 billion we spend on Medicaid every year and give it to poor families in the form of cash.

I would require the recipients to contract with a primary care doctor of their choosing to provide 24/7 “concierge care,” put the remainder in a Health Savings Account (HSA), and allow any remaining HSA funds (after a period of time) to be withdrawn for nonmedical spending.

We could do something similar for food stamps, housing subsidies and a raft of other government programs. If we did all that poverty in the United States would vanish in a heartbeat.