Government Programs to Assist the Poor and Paul Ryan’s Budget Proposal
In a recent interview for the Christian Broadcasting Network, Representative Paul Ryan argues that his budget proposal is entirely consistent with a primary tenet of Catholic social teaching—the preferential option for the poor. His argument rests on the Catholic notion of subsidiarity—that human problems should be addressed and solved at the lowest level of social organization possible—first the family, then the community, and then the state. In Ryan’s own words, "…the preferential option for the poor, which is one of the primary tenets of Catholic social teaching, means don’t keep people poor, don’t make people dependent on government so that they stay stuck at their station in life. Help people get out of poverty out onto life of independence."
I will not offer any critique of Ryan’s use of the subsidiarity principle in defending his budget proposal and budget philosophy. Others much more qualified than I have already done so. What I would like to address is his view, one shared by many others, that government programs to assist the poor—welfare, food stamps, etc.—create an insidious dependency that, ironically, causes their poverty to persist. Of course, this view has its roots in the "culture of poverty" proponents of the late 1950s and 1960s. Prominent among these were Michael Harrington (The Other America, 1962), Oscar Lewis (La Vida, A Puerto Rican Family in the Culture of Poverty: San Juan and New York, 1968), Walter Miller (Lower Class Society as a Generating Milieu of Gang Delinquency, 1958), and Edward Banfield (The Unheavenly City, 1970). These scholars argued that low income was only one characteristic of the poor. Their other characteristics or deviances which included dependency, illegitimacy, and instability could not be eradicated by income support. Indeed, these deviances would be reinforced by such support. Here is what Martin Anderson wrote in his 1978 book Welfare: The Political Economy of Welfare Reform in the United States:
"In effect we have created a new caste of Americans—perhaps as much as one-tenth of this nation—a caste of people free of basic wants but almost totally dependent on the State, with little hope or prospects of breaking free."
Similarly, but with a greater flare for the dramatic, a 1982 op-ed piece in the Wall Street Journal by George Gilder reads:
"In this heartbreaking harvest of liberal ‘compassion,’ all the necessary disciplines of upward mobility and small business activity have given way to the vandalism and chaos of gangs and drugs, illegitimacy, and prostitution. Thus poverty has been intensified and perpetuated by income redistribution."
The host of cuts in cash and in-kind transfers to the poor by the Reagan administration were justified on the basis of such pronouncements--that these programs reinforce low self-esteem and motivational deficits among the poor, feed the cycle of poverty, and transmits the legacy of poverty and dependency from parents to their children. Using anecdotal stories of welfare queens and three-generation welfare families, welfare was likened to a narcotic breeding long term dependency.
Here’s the crucial question: What does the preponderance of research on the incentive effects of welfare have to say about welfare dependency and the welfare trap? For the answer, I searched the scholarly literature over the last three decades.
Let’s begin with Richard D. Coe in his 1982 article in Challenge. Using data from the Panel Study of Income Dynamics (PSID) at the University of Michigan, Coe finds that between 1969 and 1978, 13.1% of all welfare recipients or 3.3% of entire population were in receipt of welfare assistance for 1 or 2 years of the 10 while 7.7% of all welfare recipients or 1.9% of entire population were in receipt of welfare assistance for 8 or more of the ten years. Given these results, Coe concludes that "Long-term dependency was the exception rather than the rule in the decade of the seventies." On the basis of these and other findings Coe reaches the following overall conclusion:
"It is difficult on the basis of these results to conclude that there is something inherently pernicious about the welfare state—that it poisons those who touch it with a debilitating dose of dependency. Quite the contrary. The welfare system would seem to be more accurately portrayed as a temporary fallback position for those individuals who suffer unexpected shocks to their more normal style of life."
Using a more sophisticated methodology, Mark W. Plant in a 1984 paper published in The American Economic Review uses data from the Seattle and Denver Income Maintenance Experiments (SIME/DIME), to determine whether welfare persistence over time is due to the generosity of the system ("welfare trap") or to the serial correlation in earnings. He finds strong evidence that persistence is explained by the correlation in earnings and very weak evidence of a welfare trap. He concludes the following:
"This study supports the notion that policymakers should worry about the social structures that entrap the poor into persistent low earnings rather than portraying short-term disincentives to work inherent in the welfare system as the culprit."
Like Coe, Corcoran, et al. in their 1985 paper published in the Journal of Policy Analysis and Management, use the PSID data to examine the issues of welfare dependency and intergenerational transmission. Here’s what the authors write, explicitly addressing the aforementioned assertions in the context of the Reagan welfare cuts that such programs reinforce low self-esteem and motivational deficits among the poor, feed the cycle of poverty, and transmit the legacy of poverty and dependency from parents to their children :
"…there is no evidence that such a motivationally-driven cycle exists…. And most importantly, there is little evidence of strong intergenerational transmission of poverty and dependency."
In a 1989 paper published in the Journal of Public Economics, Rebecca M. Blank, one of the leading authorities on welfare impacts and reform, used a variety of sophisticated duration dependence models to examine the extent to which program induced welfare dependence occurred among participants in Aid to Families with Dependent Children (AFDC). She finds that the statistical evidence for duration dependence in welfare spells is weak. Even for the group of women that rely on welfare as a long-term source of income support she concludes that "…these long spells are not created nor lengthened by the use of AFDC itself."
In a 1997 discussion paper from the Institute for Research on Poverty, sociologists Gary Sandefur and Steven Cook use data from the National Longitudinal Survey of Youth (NLSY) and offer some evidence of welfare dependency but report greater influences from education, marital history, number of children, and work experience.
Now, lest one accuse me of "cherry-picking", I also looked at two extensive reviews of the literature. The first by Duncan and Hoffman (1988) published in the Social Service Review states the following:
"A review of research on the patterns of receipt and behavioral effects of the Aid to Families with Dependent Children (AFDC) program shows that many widely held ideas are not supported by findings based on nationally representative data. Among the key findings are the following: short-term and long-term welfare use are equally common; long-term dependence does not appear to be induced by the AFDC program; the majority of women growing up in heavily welfare-dependent homes are not themselves heavily dependent as young adults; and social-psychological factors have not been persuasively shown to either cause or be affected by welfare receipt."
A second and even more detailed review by Robert Moffitt (1992) in the Journal of Economic Literature states that "…the available research unequivocally shows that the AFDC program generates nontrivial work disincentives." But, "…very little of the labor supply reduction arises from initially ineligible female heads who lower their hours of work to become eligible for AFDC. That is, virtually all AFDC recipients would have sufficiently low hours of work even if they were off the program to retain eligibility for benefits. …This finding implies that the work disincentives of the program have very little effect on the size of the caseload itself. …Thus the problem of welfare ‘dependency’…cannot be ascribed to the work disincentives of the program." Additionally, Moffitt concludes that "…the evidence does not support the hypothesis that the welfare system has been responsible for the…growth in female headship and illegitimacy."
Even more pointedly, in his paper published in the Yale Law and Policy Review, Moffitt writes the following:
"Most research by economists has focused on several popular beliefs about behavioral effects of the welfare system:
(1) that it serves as a severe disincentive to work
(2) that it encourages long-term dependency on welfare
(3) that it encourages marital breakup and illegitimacy
(4) that it encourages state-to-state migration to take advantage of higher benefits
(5) that welfare is "passed down" from generation to generation, as children of welfare families go onto welfare themselves.
Probably the most important, and perhaps surprising, overall conclusion from the research is how little evidence exists to support any of these beliefs, with the possible exception of the first."
So, as far as I can tell, the preponderance of empirical evidence from the research on the effects of welfare programs does not even come close to supporting the "culture of poverty" based views of Representative Ryan and his supporters towards cash and in-kind government assistance programs. Apparently, to those who continually assert such "welfare trap" and "welfare dependency" arguments, this fact is either unknown or immaterial. I will not speculate on which it might be.