With welfare reform legislation set to expire by the end of the month, Congress is preparing to revisit the issue. Earlier in the year the House and Senate finance committees each passed separate welfare reform bills. The two bills differed on work requirements, Medicare coverage, and funding for abstinence programs. Those differences are important. In particular, strong work requirements have been a key component in the success of welfare reform.

However, in the future the most important debates about welfare policy will take place not in Congress, but at the state level. In fact, it is innovative state level policies that are primarily responsible for the dramatic decline in welfare caseloads during the last five years.

Indeed, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) signed into law in 1996 differed from prior reform bills partly because it gave states more responsibility for both structuring and implementing welfare reform. Therefore, it should come as no surprise that some states have enjoyed more success than others in moving people from welfare to work. Between August 1996 and August 2000, Idaho reduced its welfare caseloads by 94 percent. Conversely, Rhode Island reduced its caseload by 21 percent over the same time-span.

Many critics have argued that the economy is largely responsible for the welfare caseload declines. However, a recent study by the Cato Institute refutes that notion. The study examines the magnitude of welfare caseload declines in all 50 states between 1996 and 2000. It finds that the strength of the economy has only had a marginal effect on the number of individuals receiving welfare. The model predicts that the decline in welfare caseloads would only be 4 percentage points greater in a state with above average economic growth than in a state with a below average economic growth.

Additionally, historical evidence also shows that the economy has only had a limited impact on the number of people receiving welfare. For instance, between 1983 and 1989 the economy was growing at a brisk rate. However, the number of AFDC recipients actually increased during that period of time. Likewise, the economy was booming during the 1960s, but welfare caseloads soared largely because benefits became more generous.

If the strong economy is not responsible for the decline in welfare caseloads, then what is? The Cato study provides some insights. One policy that is having a substantial impact on the size of state-caseload declines is the strength of state-sanctioning policies. PRWORA gave states the flexibility to sanction welfare recipients who were not complying with mandatory work activities. Some states proceeded to adopt tough sanctions that made welfare recipients ineligible for benefits at the first instance of noncompliance. Conversely, other states imposed weak sanctions that allowed welfare recipients to keep a substantial portion of their welfare benefits regardless of their conduct.

The results of the Cato study indicate that sanctioning policies played a major role in explaining state caseload declines. Holding other factors constant, a state that adopted a strict sanctioning policy for four years would experience a welfare caseload decline that is over 20 percentage points greater than a state that implemented a weak sanction for four years.

Additionally, the Cato study also examines the effect of welfare-benefit levels. Regrettably, benefit levels have been neglected in many of the current debates over welfare policy. This is unfortunate because there is evidence that benefit levels are playing a large role in caseload fluctuations. In his groundbreaking book Losing Ground, Charles Murray convincingly argues that increases in welfare benefits during the Great Society fostered greater dependence on welfare. When welfare benefits began to exceed the minimum wage, collecting welfare suddenly became economically advantageous for many women. Not surprisingly, this led to increases in both welfare caseloads and the number of single-parent families.

Indeed, the results of the Cato study are consistent with Murray’s arguments. The results show that states with low benefits enjoyed larger caseload reductions than states with high benefits. This is no surprise. In states with low benefits it will be easier for caseworkers to convince welfare recipients to leave welfare for work. Additionally, states with low benefits will have fewer people clamoring to get on the welfare rolls.

The passage of a welfare reform bill with strong work requirements may well be necessary for the continued success of welfare reform. However, it will not be sufficient. During the past six years, evidence indicates that the success of welfare reform is largely determined by policies that are adopted at the state level. While the upcoming debate over the reauthorization of welfare reform merits close scrutiny, upcoming debates in the state legislatures will carry even greater importance.