Work . . . Or Else

            In the Conservative Reform Network’s latest report on alleviating poverty through work, Michael Strain, director of economic policy studies and resident scholar at the American Enterprise Institute for Public Policy Research writes:

A job is about more than a paycheck.  It is about more than productivity and adding to the national income.  Working is central to the flourishing life.  One of the things sound public policy does is to help provide the conditions under which our fellow citizens can flourish, realize their whole human potential, and lead lives of dignity.  Public policy, then, is properly interested in helping to create a vibrant labor market.

            Strain also notes a decline in male labor participation that began in the 1950s.   “If fewer men are participating in the workforce because relatively more men would rather spend their time on other activities than in the past,” he says, “then the nature of our concern will be different than if fewer men are participating because firms don’t want to hire as many.  In the first case, we are witnessing a decline of labor supply; in the second, falling labor demand.”

            According to Strain, “A decline in the number of workers accompanied by an increase in wages implies that we are seeing a reduction in labor supply.  A decline of workers accompanied by a decline in wages implies that the demand for workers has fallen.”

            It’s a familiar fact by now that inflation adjusted wages for men without college degrees have stagnated or declined relative to the late 1970s.  That implies to Strain that our unemployment problem is demand driven. 

            For Strain, the likely causes for this inadequate demand for labor include “the changing nature of technology and its ability, through machines and software, to replace certain types of workers, along with the impact of globalization on the U.S. labor market.”  In other words, unemployment arises, not from lazy workers who prefer to do something other than work with their time, but from a paucity of appropriate work for the people looking for jobs.

             Strain presented his findings on a panel hosted by the American Enterprise Institute. He offered several solutions to unemployment, but, oddly, aside from “work-based learning programs” such as apprenticeships designed to give workers the skills employers need and will pay for, most of his other proposals are ultimately solutions tailored to labor supply problems.  Strain suggested, for example, cutting payroll taxes to increase wages and thus, the incentive for people to go to work.

            If low labor demand is the key problem, it puzzles me that co-panelists Kiki Bradley of Chartwell Policy Solutions and Robert Doar of AEI argued that the best way to reduce poverty is to include the same kinds of work requirements that the federal Temporary Assistance For Needy Families (TANF) program has.  TANF is what replaced “welfare as we know it” in 1996.

            Though it is undoubtedly a way of reducing the number of people receiving public assistance, cutting off cash assistance payments to people who fail to meet statutorily imposed work requirements is not an appropriate solution to a flagging demand for labor. Eliminating public assistance to people who do not meet statutorily imposed work requirements is a punitive solution to problems on the labor supply side of the poverty equation.  It’s a way of incentivizing work, not of producing jobs beneficiaries can actually do.

            According to Bradley and Doar, TANF was wildly successful in moving people from welfare to work.  Figure 1, a chart prepared by the Congressional Research Service, shows why they think that the work requirements in TANF reduced unemployment and thus why all public assistance programs should emulate it.

Figure 1

            Figure 1 shows the number of families receiving “welfare” from 1959 until 2014.  That number peaks at around 5.1 million families in 1994 and then fell sharply until 1999 when the U.S. experienced a relatively brief recession.  After 1999, the number of welfare families continued to decline, but at a much slower pace.

            TANF replaced “welfare as we know it” in 1996, and it wasn’t fully implemented until 1998.  But, as Figure 1 shows, the sharp decline in families receiving cash assistance under the TANF program actually began in 1994, at least two years before TANF’s work requirements could have produced any effects.  And the steep downward trajectory of the line didn’t resume after the recessions, despite the fact that the TANF law required the federal government to begin penalizing any state that failed to continue making progress in moving TANF recipients off public assistance rolls and into work or work preparatory activities.

            Doar and Bradley see in Figure 1 what they want to see.  They assume that the big decline in the number of welfare families is the result of TANF simply because it occurs after TANF's work requirements were adopted.  That’s an example of  a logical fallacy known as post hoc ergo propter hoc ("after this, therefore because of this").

            Participation in TANF is sensitive to a number of factors.  Economist Jeffrey Grogger, for example noted that the overall level of participation depends not just on how many people leave the welfare rolls, but also on the number of people who sign up for it at any one time.  In an interesting research paper, Grogger shows that about half of the decline in the level of TANF participation flows from a reduction in the number of people who applied for it.  Grogger says that this reduction had far more to do with the availability of the Earned Income Tax Credit and the general improvement of the labor market that took place between 1994 and 2000.

            I don’t think that anyone disagrees with Strain when he identifies work as crucial to individual flourishing.  I also don’t think that anyone would disagree with the proposition that even if work weren’t important for individual flourishing, society is better off when people are engaged in economically self-sustaining labor rather than dependent on others.  That’s why I agree with Strain’s argument that “public policy . . . is properly interested in helping to create a vibrant labor market.”

            But, if we’re going to use public policy for this purpose, we ought to use the kinds of policies best suited to accomplish our objective as well as metrics that tell us whether we have succeeded.   Having fewer people receiving public assistance isn't the same thing as having more people gainfully employed.  To reduce poverty (and move people “from welfare to work”), we’re probably far better off investing in education and training as well as in jolting the economy with federal spending on appropriate projects than we are in threatening people with starvation or homelessness for their failure to pursue pointless activities that lead to non-existent jobs.