Roanoke Times
Copyright (c) 1995, Landmark Communications, Inc.
DATE: SUNDAY, June 25, 1995 TAG: 9506260015
SECTION: VIRGINIA PAGE: C-1 EDITION: METRO
SOURCE: PETER BAKER THE WASHINGTON POST
DATELINE: RICHMOND LENGTH: Long
Here's what he doesn't say: This windfall is no sure thing. The number is a "guesstimate" based on debatable assumptions, optimistic forecasts and outdated information, according to the Allen administration's own documents. The governor is counting on a lot of ifs: The state will save so much money only if welfare recipients react as expected, only if the economy stays strong, only if costs can be kept down.
If they don't, it doesn't or they can't, the pot of gold could disappear as quickly as a rainbow evaporates, according to specialists in the economics of poverty.
"You hear all these claims of large cost savings," said University of Wisconsin economist Robert H. Haveman, who has watched many state welfare experiments, "but my guess is, these claims would not stand up to strict scrutiny in the long run."
The financial benefit from the governor's nationally watched welfare project is key to its ultimate success. Allies and adversaries have argued about the numbers since the governor unveiled his plan to put recipients to work and cut off their benefits after two years.
Their disagreement mirrors similar debates in Washington as Congress considers a national welfare overhaul. It also underscores a fundamental truth about the current wave of experimentation: The science of welfare is imprecise at best. Since no other state has tried what Virginia plans to start Saturday, no one can predict with certainty what effect it will have on the budget's bottom line.
Allen's math sounds simple and logical: If poor people are no longer eligible for Aid to Families with Dependent Children after a certain time, the state must save money. Even some who disagree with Allen's method would nonetheless agree that it could cut costs.
"If it's really a hard time limit, it will save gobs of money," said Isabel V. Sawhill, a former federal budget official who now works as a senior fellow at the Urban Institute. "It's easy to save money by throwing people off welfare."
An examination of what stands behind the $130 million figure finds that it does not include millions of dollars in potential costs and may inflate projected savings through unrealistic assumptions. For example, the state assumes that a typical workfare participant would need only $20 for transportation each month, even though fares for public transportation in Northern Virginia, for example, cost two to five times that much.
Allen administration officials say their fiscal analysis was thorough and professional. Yet, at the same time, they portray it as informed supposition that, inevitably, will prove wrong.
"We have not cooked numbers here; I can guarantee you that," said Robert H. Lockridge, who oversees social services budgets in the Allen administration. But, he added, "if you were to ask me ... is $130 million a good figure, the answer is no. It's a good figure for right now, but there are things that are going to change."
Anne C. Pace, the budget analyst who crunched most of the numbers, agreed. "From a budget perspective," she said, "it's a best guesstimate."
Allen's welfare program will impose new restrictions on the state's 74,000 AFDC recipients, including a rule eliminating benefits for children born to mothers already receiving welfare. The two central elements - a requirement to work within 90 days of receiving a check and the two-year limit on benefits - will be phased in over four years.
Budget analysts estimated that workfare and the time limit would save $136.6 million by 2000, more than half of the state's annual AFDC budget. About $89.8 million of that would be state funds, while the remaining $46.8 million would revert to the federal government.
But that figure did not include costs of other elements of the welfare plan, estimated at $11 million for the first two years. It also rests on assumptions:
Analysts assumed that every welfare recipient thrown off the rolls after two years would have stayed on for five years, therefore crediting the state with three years of savings for each of them. In reality, however, most recipients leave welfare long before five years. The typical welfare recipient in Virginia leaves after about two years and eight months. Officials justified their assumption by noting that recipients often return to the rolls.
Analysts assumed no growth in the AFDC population over the next five years, even though the caseload jumped by 37 percent in the past five years. Officials attributed that growth to the recession and relied instead on the fact that the rolls have flattened out in the last two years.
The forecast assumed that the two-year cutoff would be phased in faster than it actually will. Analysts believed the new rules would affect 37 percent of welfare recipients in the first year, when in fact just 20 percent will be covered. That falloff is crucial because the main saving in the first five years will come from no longer paying benefits to those phased in at the beginning.
The analysis made no estimate of how many recipients will be put into community service or subsidized business jobs, or how much that will cost. Officials contend it will cost no more than is currently spent on AFDC, but they did not factor in any extra money for supervising recipients put to work, say, cleaning a state park or shelving books in the local library.
The estimate did not take into account any extra expenses for day care. An additional $9 million was allocated for child care for welfare families this year, but officials said that money would have been needed even without the Allen plan and therefore it was not included in the calculations. No additional money will be required, they maintain, even though the plan envisions putting many more parents of small children to work.
The projection did not take into account any hardship exceptions to the two-year cutoff. Under the new law, officials can decide to allow recipients to keep receiving cash payments for up to a year after the first 24 months if unemployment is high in their area, if they still can't find work despite genuine efforts, or if they need additional education or training.
No attempts were made to weigh whether the program would simply shift costs to other welfare programs, especially food stamps. Anyone who loses cash benefits after two years without a job will be eligible for more food stamps, about 30 cents for every dollar of lost AFDC. State officials didn't account for those increased costs and aren't particularly bothered if they materialize, because the federal government picks up the full cost of food stamps.
The analysis counts on saving from new policies that cut off assistance to recipients whose school-age children are truants and to teen-age mothers who don't live with a parent or adult guardian.
But how much of this will happen is unclear. The state assumed that 100 percent of those who currently don't meet those standards still won't in the future and so would lose their benefits, allowing no room for the possibility that many would modify their behavior to keep their checks. Pressed on the point, officials conceded that, in reality, some or all might comply, canceling any savings, but they point out that the $1 million a year they projected in savings was relatively small.
In an interview, Lockridge, the budget official, said quarreling with specifics misses the point. The program will save money, he said, even if it is not the amount that Allen tells everyone.
"I can tell you from experience that these numbers will not pan out, as far as implementation. There's just no way it can," Lockridge said. "I will clearly accept the fact that we really don't know. ... We cannot basically put a finger on a number and say this will be it. ... We did the best that we could with the information we had."
by CNB