It is true that there is voter resistance, but this occurs because it is very difficult for States and local governments to get out of line with the tax levels in neighboring communities.

If, for example, the State of Maryland enacted very high income taxes, some people might move out of Maryland. Similarly, if it enacted higher corporate taxes, businesses might not come into the State.

In a national system like ours, it is very difficult for some States to make more use of the tax sources which grow with the growth of the national economy. That is why the Federal Government, which can levy such taxes, should allocate some of them to the States.

I do not believe that there is voter apathy at the State-local level. On the contrary, they have made a remarkable effort. They have increased property and sales taxes and they need more help.

Mr. WEINBERG. A large part of the local opposition comes from the regressive nature of the taxes involved at the local level-school bond issues have been defeated in many areas because existing sources of funds-essentially profit taxes are so burdensome on lower income groups that they vote against the bonds.

This does not mean they do not recognize the need but they recognize that they are already shouldering more than their fair share of the cost of meeting that need.

In Michigan we have a 4 percent sales tax that covers food and drugs as well as other goods, and we have no corporate income tax and no personal income tax. We have property taxes that have reached the outermost limits.

This is the kind of situation that makes for the defeat of proposals at the State and local levels to meet the needs.

MONEY FOR THE STATES*

BY JOSEPH A. PECHMAN

First broached in 1964 by Walter W. Heller, the idea that the Federal Government should share some of its revenues with the 50 States and over 90,000 local governments has an unusual degree of support from moderates in both political parties. The opposition comes mainly from conservatives who always prefer tax reduction to better public services, and from liberals and bureaucrats who believe that major decisions about public services should be made in Washington. Revenue sharing is intended to allocate to the States and local governments on a permanent basis a portion of the very productive and highly "growth-elastic" receipts of the Federal Government. Most Federal revenues come from income taxes that rise at a faster rate than income as income grows. By contrast, State-local revenues barely increase in proportion to income. One reason is that the Federal Government has virtually preempted the fruitful income tax. Ninety percent of the taxes that are levied on incomes of individuals and businesses in this country goes to the Federal Government, though 33 of the 50 States tax incomes. States are generally reluctant to increase taxes on incomes, for fear of losing business to other States. State-local needs have outstripped the potentialities of their revenue system at constant tax rates, so that the rates have been pushed steadily upward throughout the postwar period and many new taxes have been added. But essential public services are starved by Governors, mayors, and legislators who naturally try to avoid the politically distasteful and sometimes politically suicidal-choice of increasing taxes. Furthermore, State-local taxes are on balance regressive; they impose unnecessarily harsh burdens on low-income recipients.

Stripped to essentials, the revenue-sharing plan would operate as follows:

A portion of Federal revenues would be automatically set aside each year in a special trust fund on the basis of a predetermined formula.

Disbursements from the fund would be made primarily on a per capita basis, a method that automatically helps the poorer States relatively more than the richer States.

The funds would be turned over to the States, with the understanding that a major share would go to the local governments.

Constraints on the use of the funds would be much less detailed than those applying to conditional grants. However, the funds would not be available for highway construction, since there is a special Federal trust fund with its own earmarked revenue sources for this purpose.

*Reprinted from New Republic, April 8, 1967.

An audit of the actual use of the funds would be required, as well as certification by the appropriate State and local officials that all applicable Federal laws, such as the Civil Rights Act, have been complied with in the activities financed by the grants.

The per capita method of distributing the grants was chosen because it is the best available index of State fiscal capacity and need. It allocates more money to the populous States; at the same time, it automatically distributes relatively more to a poor State than to a rich State. For example, a $25 per capita distribution would amount to 10 percent of the budget of a State that can afford to spend $250 per capita and only 5 percent of the budget of a State that can afford to spend $500 per capita. More equalization could easily be provided if desired; for example, a small part of the fund, say, 10 percent, could be allocated to the poorest third of the States. Tax effort could also be given some weight in the formula, thereby encouraging States to maintain or increase tax collections out of their own sources, and penalizing those which might yield to the temptation of reducing State taxes.

On the other hand, it would be totally inappropriate to allocate the funds in proportion to the amounts collected from each State. This would give disproportionately larger shares to the wealthiest States, and would widen rather than narrow differentials in State fiscal capacities.

The same criticism holds for the various types of Federal income tax credits for State income taxes, which are often proposed as a substitute for revenue sharing. The tax credit is a method to coerce States to adopt income taxes (and is needed for that reason), but it is not a good device for achieving the equalizing objectives of revenue sharing. So long as State shares of Federal money depend on income, wealthy States will do better than the poorer ones.

Several methods can be used to calculate the amounts to be set aside annually for revenue sharing. The two most important criteria are that (a) the amounts should grow more than in proportion to the growth of the economy, and (b) the changes that might be required with the passage of time should be held to a minimum. The first criterion would be satisfied by any one of a number of growing basesfor example, total Federal revenues, total income tax revenues, and the individual income tax base. The second would be satisfied best by the individual income tax base (i.e., taxable income) which is changed only rarely. Actual income tax receipts could also be used, provided Congress is prepared to change the revenue-sharing formula when tax rates change. On balance, the income tax base is preferable.

ANOTHER $6 BILLION

At this year's expected income levels, that base is in the neighborhood of $300 billion and the allocation would amount to $3 billion for every percentage point of the base. If the Vietnam war were to end soon, the Nation could easily afford to allocate two points of the income tax, or $6 billion, for revenue sharing, as an addition to the $15 billion a year that the Federal Government already sends the States in grantsin-aid. This extra $6 billion would be enough to finance general grants

average $30 per capita. At this rate, California would receive about $580 million, New York $560 million, Massachusetts $160 million, Wisconsin $125 million, Maryland $110 million, Mississippi $65 million, and so on. Equally important as the amounts, the grants would automatically grow with the tax base-in a full-employment economy, they would double every 9 years.

The plan is often criticized because the States and local governments might be in trouble if the revenue-sharing funds declined during a recession. But it turns out that this is not a matter of great concern. The tax base has declined only twice since the end of World War IIby 4 percent in 1949 and by less than one-tenth of 1 percent in 1958. These are within the range of fluctuations that State and local governments are accustomed to in some of their own tax sources. But even if a deep recession occurred, Congress could easily add to the statutory amounts to prevent State-local distress. Few antirecession measures would satisfy both the efficiency and stabilization objectives as well as revenue sharing. Rather than reducing the Federal Government's flexibility to combat recessions, as some allege, the plan would provide another useful outlet for Federal funds in these circumstances.

Some people have embraced the revenue-sharing plan as a method of undercutting the present Federal conditional grant system: a few would even replace the present grants by unconditional or generalpurpose grants. But the two types of grants have very different functions and these cannot be satisfied if the Federal system were limited to one or the other.

Conditional grants-for example, for urban renewal and public assistance are justified on the ground that the benefits of many public services "spill over" from the community in which they are performed to other communities. Expenditures for such services would be too low if financed entirely by State-local sources, because each State or community would tend to pay only for the benefits likely to accrue to its own citizens. States have a well-developed system of conditional grants to local governments for this reason. Additional assistance by the Federal Government is needed to raise the level of expenditures closer to the optimum from the national standpoint.

General purpose or block grants-for example, for health, education, and welfare as a block-are justified on different grounds. In the first place, all States do not have equal capacity to pay for local services. Even though the revenue effort of the poorer States is average, they are unable to match the revenue-raising ability of the richest States. Second, Federal use of the best taxes (i.e., on income) leaves a substantial gap between State-local need and State-local fiscal capacity. Moreover, no State can push its rates much higher than the rates in neighboring States for fear of placing its citizens and business enterprises at a disadvantage. This justifies some Federal assistance even for purely State-local activities, with the poorer States needing relatively more help because of their low fiscal capacities.

For these reasons, the general-purpose grants are intended to supplement the conditional grants, not to replace them. Considering the large unmet needs throughout the country for public programs with large spillover effects (education, housing, et cetera), adoption of revenue sharing should not be the occasion for reducing conditional

grants. It is a well-known axiom of logic that two objectives cannot be satisfied by using only one instrument.

The most serious criticisms of revenue sharing come from those who have lost faith in the State governments. On the whole the States have been doing a good job, although there are exceptions. Without central direction or coercion, they have actually used most of their scarce resources for urgently needed State and local programs. Between 1955 and 1965, general expenditures of State governments rose steeply by $23 billion, to around $40 billion. Of this increase, about 60 percent went for education, health, welfare and housing-more than 40 percent went to education-most of it through grants to local governments. This evidence suggests that, if the States were to receive unencumbered funds from the Federal Government, they would spend them on urgently needed services whether the particular services were stipulated in the legislation or not. To be specific, if the Federal Government allocated $6 billion for revenue sharing, there is little doubt that about $3 billion of this money would be spent on teachers' salaries, school buildings, and other educational needs.

The Federal Government would, of course, expect the States to pass the funds through to their local governments in an equitable manner, but this is much less of a problem than most people might suppose. All States give aid to local units and most give significant amounts. As a matter of fact, the State grant-in-aid system for local governments is much more highly developed than the Federal grant system. In the aggregate, transfers from State to local governments account for more than a third of State expenditures and about 30 percent of local general revenues. By contrast, Federal grants amount to only 17 percent of State-local revenues. Thus, even without any specific requirements, the local governments would receive at least a third of any general funds the States might receive from the Federal Government.

SAFEGUARDING LOCAL UNITS

But there is no reason why the Federal Government should not write a "pass-through" formula into the plan to be sure that the States will turn over to their local units an even larger share of the revenuesharing receipts than they might otherwise allocate. This can be done. in two ways:

1. The revenue-sharing legislation might provide that all States must pass along a certain percentage of the grants to their local governments. In view of recent trends, the minimum should be at least 40 percent and might even be as high as 50 percent. This would prevent any State from short-changing its local governments (although it might be difficult to detect offsetting reductions in existing grants if the State legislature was of a mind to do so). The disadvantage of a fixed percentage is that the extent to which the States delegate responsibilities to, and share revenues with, local governments varies greatly. In some States, the appropriate percentage may well exceed the 50percent mark, and in others it may be below it. The danger is that any minimum percentage is likely to become a maximum, so that stipulating the percentage may do more harm than good in some States.