All told, there are 338 means-tested federal programs spending about $240 billion a year. State governments put up about $1 of matching money for every $2 from the federal government. So in principle the federal government is considering handing over to state and local governments as much as $350 billion a year - an amount roughly equal to $3,500 for every household in America.

The goals of these reforms are laudable: transferring power back to the people and allowing local communities to find workable solutions to a welfare system that is a dismal failure. Yet serious questions remain. How much should go to each state? Should restrictions be attached? What promises should be made for future years? Let's take a look.

Comprehensive Block Grants. Some in Congress favor block grants but would restrict them to certain categories. For example, the House reform bill block-granted AFDC but not food stamps. Others suggest a swap of programs. For example, Senator Nancy Kassebaum has proposed a federal takeover of Medicaid in return for sending all other entitlement programs back to the states.

These proposals are misguided for two reasons.

First, welfare-poverty problems do not fit into neat compartments such as food, housing, health, etc. Some of the most serious health problems of the poor are related to lifestyle - eating habits, living quarters, etc. For example, by failing to get prenatal care and by smoking and drinking during pregnancy, low-income women contribute to the high mortality rate and low birth weight of their babies. And although there is no evidence that children from low-income families are hungry, by eating improperly - especially eating too few fruits and vegetables - the poor probably contribute to their higher rates of cancer and heart disease.

The key to changing behavior in ways that improve health is to package all forms of relief, taking account of interrelationships among food, housing, health care and management of a family budget.

Second, failure to block-grant all means-tested programs would create perverse incentives for the states. A state that had the opportunity to manage AFDC dollars but not food stamps would have an incentive to skimp on AFDC benefits (which would fall fully within the state's own budget) and make up the deficit by expanding food stamp benefits (which would be paid for mainly by the federal government).

Some have already accused Michigan of engaging in this type of substitution. Governor John Engler has a national reputation as a proponent of welfare reform and has abolished welfare for able-bodied, single males. But he advocates expanding the caseload of Supplemental Security Income (SSI) - which is paid for with federal funds.

No-Strings Transfers. Only two requirements should be imposed on states that receive block grants: (1) that they match a certain portion of the federal funds and (2) that funds be spent on low-income beneficiaries. Otherwise, states should have maximum flexibility to design and implement their own welfare systems.

Two violations of this principle are tempting to current members of Congress.

First, special-interest constituencies are trying to impose a requirement that health care block grants be spent only on health, housing block grants only on housing, etc. As noted above, this approach ignores the fact that the problems of poor families do not fit neatly into categories and that an integrated approach is far superior to a compartmentalized one. Compartmentalizing block grants also is a thinly veiled attempt by those wedded to the idea of entitlements to try to force state governments to continue running the very programs most people agree have failed.

Second, the House bill contained such restrictions as a five-year limit on welfare benefits to most families, a prohibition on additional benefits for AFDC women who have additional children, a prohibition against benefits for teenage mothers and a requirement that half of all AFDC recipients work or enter job training programs after five years. Such mandates would defeat the purpose and theory of the block grants, including the idea that the federal government should not impose a one-size-fits-all program.

Grants Proportional to Need. When a lump sum distribution is made to the states, what fraction of the total should each state receive? The best approach is to base each state's share on the proportion of the nation's poor who reside in that state. A much less desirable approach is currently favored by the Republican leadership in Congress and is reflected in the House bill. This approach would block-grant funds based on current federal spending, rewarding the states that currently spend the most instead of assisting those with the greatest needs.

Future Block Grants Independent of Future Needs. Under entitlement programs with shared federal-state financing, the federal government's share averages about 55 percent. State governments pay the remainder with matching funds. This structure punishes the states that reduce poverty and rewards those that fail. For example, if a program has a 50-50 match, a state gets to keep only 50 cents out of a dollar of waste it curtails. Conversely, if a state generates a dollar of waste, it pays only 50 cents. To avoid these perverse incentives, block-grant amounts should be independent of changes in poverty population over time (unless the changes are caused by immigration). The block-grant formula should be based on the initial distribution of pretransfer poverty. If a state reduces its poverty population, the state should be allowed to keep 100 percent of the resultant savings. Conversely, it should pay 100 percent of the cost of its failures.

Capping the Growth of Block-Grant Funds. As a result of welfare reform, we anticipate a steady reduction in the percent of the population living in poverty and in the number of people who qualify for relief. This assumption should be built into the structure of block grants by, say, limiting their growth to the rate of inflation.

Capping the growth of block grants would send an unmistakable signal to the states that they cannot engage in business as usual. Such caps would pressure them financially to solve their welfare-poverty problems or pay the full cost of failing to do so.

This Brief Analysis was prepared by NCPA President John C. Goodman and Robert Carlson, Senior Fellow of the Free Congress Foundation.

Note: Nothing written here should be construed as necessarily reflecting the views of the National Center for Policy Analysis or as an attempt to aid or hinder the passage of any legislation.