• Commentary

Guaranteed Basic Income not a Guaranteed Solution to Poverty

When addressing the challenging question of alleviating poverty, many are gravitating to an idea called Guaranteed Basic Income (GBI.) In practical terms, GBI would guarantee a baseline income level, with those falling short receiving a monthly stipend from the government.  

Georgia is weeks away from the launch of In Her Hands, a privately funded program that for two years will give an average of $850 per month to 650 women in Atlanta’s Old Fourth Ward, suburban Atlanta and southwest Georgia. A separate program in Atlanta will provide $500 per month to 300 residents for a year.

A guaranteed income program like this is not unique. There have been comparable programs in Mississippi; Stockton, Calif.; and Ontario, Canada, among other places. These programs varied in duration, the number of recipients and monthly amounts. 

There were some positive outcomes: Overall, recipients reported being better able to handle unexpected expenses, for example. But there were negative ones, too: A review of the Ontario program found those recipients surveyed afterward were less likely to have been employed during the program than before it.

Ultimately, the most crucial aspect of these programs is their effect on work. As with the welfare system, no guaranteed basic income can replace the benefits of having a job – not only financially, but also in terms of the intangible dignity of earning a living and contributing to the well-being of others.

These recent pilot programs, including the Ontario one, haven’t been large enough or lasted long enough to give us significant data about the effect of a long-term guaranteed income on work. There may be only one program that has.

In 1967, the Rev. Martin Luther King Jr. called for a guaranteed income in America. The next year (the year he was assassinated) the federal government actually started such an experiment. Back then it was called a “negative income tax,” as the great libertarian economist Milton Friedman had named the idea.

The experiment was done in four segments that ran from 1968 to 1980; spanned urban, suburban and rural areas in six states; varied by amount of income; lasted either three or five years per cohort; and included about 9,000 households. It’s surely the most ambitious publicly funded social experiment that no one has ever heard of. It’s also the most complete data set about the potential outcome of a Guaranteed Basic Income program.

The key revelations are about work. Across all groups, the average male recipient worked 7% fewer hours and the average female recipient worked 17% fewer hours when participating in the program. As a result, these households reduced their earned income by about two-thirds of the amount of payments they received. In other words, for every $1,000 they received in payments, the typical family reduced their earned income by $660 – a net gain of only $340. In a review of the data, the Brookings Institution’s Gary Burtless summarized the effect on earnings as follows:

“Even if the earnings reductions are taken to be modest, it is reasonable to ask whether most taxpayers would be willing to spend $3 in order to raise the incomes of poor, two-parent families by only $1.”

But while a reduction of work hours was the average outcome, some subgroups of recipients increased their earned income while receiving the payments: black men and women and Hispanic men in urban New Jersey and Pennsylvania, and married women in Gary, Ind.

One clear result, however, was that people were more likely to reduce the amount they worked if they received higher payments or knew they would be receiving payments longer. This was apparent in the variable payments and the three-year vs. five-year variables in the 1968 study. 

For many, the fact that many guaranteed basic income recipients chose to work less will not come as a surprise. But many proponents of guaranteed income claim that a guarantee of money from the government won’t disincentivize recipients from maintaining the same level of employment. That’s undoubtedly true in some individual cases, but the data suggest the opposite in the aggregate.

While there hasn’t been a study as comprehensive and long-range as the one in 1968-1980, all of the existing data points to a possible problem with a guaranteed basic income. If work is not part of the equation, a guaranteed income will not single handedly help people navigate the path from poverty to prosperity.