Planning For the Best Of TimesBy Daniel P. McMurrer and Isabel V. Sawhill
Monday, August 18, 1997; Page A19
The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 required states to place 25 percent of all welfare recipients in jobs or other work-related activities by Sept. 30. The requirements for two-parent families are higher, and some households are exempt. But for most households, the percentage required to work rises by five points each year until 2002, when it hits 50 percent.
The formula sounds simple enough, but it's not set in stone. States can adjust the participation rates downward if their caseloads fall below where they were in FY 1995. If, say, a state's caseload dropped by 10 percent between 1995 and 1996, it would have to find work for 15 percent of its recipients instead of 25. It's this adjustment factor, driven by a good economy, that's helping states breathe easier as they think about how to move large numbers of recipients into jobs. What they don't seem to have noticed is that this formula can backfire if bad times follow good (as often happens).
Consider two scenarios. Under the first, the economy keeps growing at a modest pace, and the unemployment rate remains at about its current level. Under this scenario, the number of additional recipients required to work each year between now and 2002, our analysis shows, is no more than 140,000 per year, on average. These workers would make up no more than 10 percent of the roughly 1.4 million additional people predicted to join the labor force each year under this scenario. Although finding jobs for the most disadvantaged portion of the caseload will be extremely difficult, the numbers at least don't overwhelm. A good economy, it turns out, helps in three ways. First, it reduces the size of the caseload by limiting the number of people applying for welfare. Second, it lowers the proportion required to work (because these requirements are tied to previous caseload declines). Finally, it provides more job opportunities for those subject to the requirements. In short, with a strong economy, the optimistic predictions of welfare reform advocates make some sense.
Under a second scenario, the economy enters a recession in 1998 similar to the one experienced during 1990, and caseloads rise. Then the three factors just mentioned combine to produce exactly the opposite effects. States would be faced with requirements to increase the proportion of the caseloads in work activities not by five percentage points a year, as envisaged in the new law, but by as much as 10 to 15 percentage points a year. The number of people required to work would double to about 280,000 a year, on average. In some years, it would exceed half a million. If thrown into a job market in which others were being laid off, these would-be workers would find themselves competing for already-scarce jobs. Under these circumstances, states would be hard pressed to meet the higher participation requirements. Indeed, most probably would fail.
Unfortunately, Congress cannot repeal the business cycle. But it could consider a modest correction to the welfare reform bill next year that at least would eliminate the tendency of work requirements to vary perversely with the state of the economy. The job of moving welfare recipients toward self-sufficiency is hard enough without imposing unrealistic demands on states that, thus far, appear to be coping quite well with the new law.
Just as investors in the stock market seem to have forgotten that what goes up must come down, those banking on the success of welfare reform need to remember that caseload declines are often followed by equally large increases. Welfare reform may indeed work. But let's not declare victory prematurely.
Daniel P. McMurrer is a research associate at the Urban Institute; Isabel V. Sawhill is a senior fellow at the Urban Institute.
© Copyright 1997 The Washington Post Company