Why Most Government Antipoverty Programs Aren't Effective
In Part 1 Diogenes discussed in economic terms what it means to be poor in America and reviewed federal programs for those in need. After 50 years of effort and spending at a rate of about $600 billion/year, more Americans are poor than when the programs began. An old adage states that “insanity is repeating the same mistakes and expecting different results.” Here we will look at a policy alternative to current approaches.
The primary reason for the failure of governmental poverty programs is an “agency problem.” The antipoverty bureaucracy has itself often become an impediment to those who would try to climb out of poverty because of disincentives for working. We have agencies that try to insure that those receiving aid it use it for “correct” purposes (milk, not beer). Such attempts always fail because nobody spends somebody else’s money as carefully as their own. And then there is the cost of the agency. For example, the food stamps program (SNAP) has program costs of about 12%.
Add in waste and poor management, and today America cumulatively spends more on poverty programs than it would cost to raise the income of every American above the poverty level. Even assuming NO income for the 46.2 million Americans in poverty, and multiplying by the $11,170 per capita that is the level below which per capita poverty is defined, would cost about $514 billion. The costs would be considerably less if we gave this aid on a household income basis.
The only enduring cure for poverty is increasing production of goods and services which result in greater demand for labor, because ultimately jobs are needed to pull people out of poverty no matter how good government antipoverty programs are. In other words, free markets are the ultimate antipoverty programs.
What We Should Do
Diogenes believes that America should implement a Guaranteed Minimum Income (GMI). Also known as a Negative Income Tax (NIT), a GMI would provide baseline support to which all would be entitled. Diogenes would define this as the federally defined household income poverty level of $23,050 for a family of four, plus $1. Adjustments up or down would be made for additional children and for each state’s cost of living. The GMI would be implemented through the tax code utilizing IRS infrastructure.
The GMI is not a new idea. Milton Friedman advocated for a NIT in the early 1960s. Legislation was first proposed during the Nixon administration, and had support from both liberals and conservatives along with a long list of recipients of Nobel Prizes in Economics and other economists, but ultimately did not pass Congress.
Almost all other antipoverty programs should be phased out in order to pay for the GMI. The GMI would be far more efficient than welfare, food stamps, subsidized housing, unemployment benefits, and the 47 different federal job training programs. The GMI should be progressively phased out at a rate of 50% reduction of benefits per dollar of income, so as to incentivize Americans to find any sort of work at any wage.
The GMI would also necessitate a repeal of minimum wage laws, as all workers would already be guaranteed a base income just above the poverty level. Any work income would incrementally reduce government assistance even as the workers’ household income rose to median levels. Reducing the costs of unskilled labor would encourage business to hire more workers, providing crucial entry level jobs that are priced away with existing minimum wage laws.
At first blush, the move to a GMI might appear to be a radical shift away from the status quo. However, the GMI would be a dramatic expansion of perhaps the single most efficacious federal antipoverty program, the Earned Income Tax Credit (EITC). Congress first implemented the EITC in 1975 to offset the burden of social security taxes for the working poor, in order to provide greater incentive to work. When the EITC exceeds the amount of taxes owed, it results in a tax refund to those who claim and qualify for the credit. According to the IRS, the total cost of the EITC was $61 billion in 2011, and program costs were less than 1%. It helped move 6.6 million Americans, half of them children, out of poverty. In 2013, the maximum credit to a married couple filing jointly with three children is $6,044. The credit phases out as income increases to about the national median income level.
Why This Hasn’t Been Done Before
Diogenes proposes to dramatically expand the single best program we have and replace lots of programs that do not work. The incremental cost would be nothing because the change would redirect current antipoverty funding. And if the program succeeds, it’s cost would decline. So who would disagree? For starters, the entire federal antipoverty bureaucracy who would lose their jobs. Unlike the private sector, government is rarely reorganized when the results yield fewer jobs.
Historically, government antipoverty programs have seemed reluctant to give cash aid to the poor. For example, food stamps have value restricted to consumable products excluding alcohol, as if aid recipients didn’t know how to spend what funds they have. The moral legitimization of the welfare system requires that recipients use aid to support lifestyles that comport in some rough sense with the idea of a good life held by taxpayers who provide the funds. In the minds of many, it’s one thing to provide a safety net, and another to support those who just don’t want to work. The counter argument is that by keeping a GMI at a less than “comfortable” level for able bodied Americans, incentives to work would dramatically increase.
Others might argue that there are other problems with a GMI:
- The incentive to work is reduced, however marginally, by providing any level of guaranteed income.
- If government lowers the minimum guaranteed level below an absolute safety net, we don’t fulfill our societal imperative to provide for all who are unable, for whatever reason, to provide for themselves.
- By having only a gradual phaseout of support as income grows, we invariably will pay (some) benefits to those above the poverty level.
- Whatever the initial level of a GMI, politicians will continue to raise the amount until it eventually bankrupts us. Just look at Social Security, welfare and the Income Tax for examples.
The rebuttal to these arguments is that living at the no work GMI base level is pretty tough in America. No one would want that kind of life if they could reasonably easily augment it. Raising lower than median incomes to the median level (about $51,000 for a family of four), would be an excellent use of American resources. Because of income inequality, this measure is far below the mean income, and we are already spending at the cost of a GMI with far less effective programs. A limit to future spending largesse could be enshrined in the legislation by defining the base income to $1 over the poverty level and by pegging the phaseout of the GMI to an upper limit just below the median income.
A GMI would not be a perfect program, but the perfect is not to be found. It would be a significant improvement over current antipoverty programs and would not cost taxpayers a dime. Isn’t that something virtually everyone wants?






