Minnesota is to be commended for having a detailed tax expenditure budget-most states do not. The report, prepared by the Department of Revenue, is a useful resource for the development of state tax and spending policy.

The Tax Expenditure Concept

The concept of "tax expenditures" originally came out of the United States Treasury Department in 1968. Stanley Surrey, Assistant Secretary of the Treasury for Tax Policy during the Kennedy and Johnson administrations, and one of the great thinkers in tax policy, is credited with the idea. Since its introduction it has achieved wide acceptance around the world as a necessary tool in budget analysis. Surrey's view, shared by many, was that an understanding that the tax system is also an instrument for government spending "is essential if an informed public is to be able to consider intelligently the total impact of government spending both on the economy and on particular groups of people or businesses."⁽¹⁾

The Congressional Joint Committee on Taxation states in its tax expenditure report that:

The essential insight provided by the "tax expenditure" concept is that a law that lowers a citizen's tax liability has no different effect than a law that requires a direct payment to the citizen. And if the law is designed to accomplish a public-policy goal distinct from the equitable collection of tax revenues, then it is better to evaluate it under the standards we evaluate spending laws than the standards by which we evaluate tax law.

Take, for example, an income tax. Economists tell us that the most efficient, least distorting, tax uses the purest, broadest, measure of income. We can also achieve lower rates with a broad tax base and no special tax breaks. Furthermore, such an income tax is more likely to impose a burden that is closely related to taxpayers' ability to pay. Finally, if all income is taxed, without exception, then taxpayers with similar total income will pay similar amounts of tax.

In other words, the most efficient and equitable income tax applies to all income with no special breaks. So why would anyone devise any other kind of income tax? Because those who make government budgets are not only concerned about making the best tax system. They are concerned with making the best budget, including both taxes and spending. Deviations from a pure income tax are introduced to accomplish spending goals, not taxation goals.

A Most Privileged Type of Spending

Tax expenditures are not just any type of government spending. They are, for the most part, subsidies. In general, subsidies are a form of government spending that require close scrutiny and constant review. The transfer of government funds to private parties is not to be taken lightly.

For most subsidies accomplished through tax expenditures, however, oversight is at a minimum. First of all, they are entitlements. Thus, they are not subject to review with each budget cycle.

Also, unlike most spending, tax expenditures are rarely paid for by an explicit tax increase or prioritized with respect to other spending. They are largely left out of the painful budget process whereby the hard choices between spending on education, roads, prisons and other competing public needs are made. For the most part, the only time tax expenditures come under review is when policy-makers have determined that a general tax increase is necessary. This is a step elected officials take as infrequently as they possibly can.

Furthermore, tax expenditures are often impregnable to attack for not achieving their purpose. An economic development agency subsidy program that granted direct payment of millions of dollars to large corporations would be carefully scrutinized. The agency would monitor the program to ensure it accomplished its objectives. During the budget process the legislature would demand, and the agency would provide, data proving the program's effectiveness.

But call this same program an "Investment Tax Credit" or "Research and Development Credit" and such scrutiny is much less likely. Economic development agencies usually do not have to defined such programs because they are not administered by the agency. Departments of revenue are not typically expected to defend these programs because they are merely charged with handling the paperwork. Frequently, no one is accountable for tax expenditures.

In fact, tax expenditures are often, in effect, given the highest priority of any spending programs. They are in the privileged position of being an entitlement, reviewed infrequently and rarely is an agency held accountable for their effectiveness.

Painless Spending

Tax expenditures are often treated as less costly than other spending. As I have alluded to, it is rare for a tax expenditure to be explicitly connected to an offsetting tax increase or a spending cut. This makes them legislatively attractive. They are also often attractive to government agencies because they are not attached to an agency budget. Thus, an agency can increase programmatic spending in its area without taking budget responsibility.

Tax expenditures do, however, cost every bit as much as any other government spending. One way to think about this is to treat the payment of taxes and receipt of tax expenditures as separate transactions. Instead of thinking of a taxpayer paying less in taxes because of a tax expenditure, think of the taxpayer as paying its full taxes and then receiving a government subsidy. It is merely an administrative matter that the tax obligation and the subsidy are being netted to lower the taxpayer's payment to the tax collection agency. Viewing tax expenditures this way makes clear that they are every bit as costly as any other form of spending.

Another way of demonstrating the costliness of tax expenditures is to show how they are just as validly viewed as "tax penalties." The most famous tax penalty today is the "marriage penalty" in the federal income tax. An interesting footnote to the debate going on in Washington is that most of the current federal marriage penalty came into being legislatively through a cut in taxes for the unmarried. In other words, it was passed as a tax break, not a penalty. It has, however, become viewed as the marriage penalty instead of the single persons bonus. Nevertheless, both characterizations are equally valid.

This concept is exceedingly important in the context of tax expenditures. For example, Minnesota adopts the higher federal standard deduction for the blind in its personal income tax. It may sound nonsensical, but this can, with equal validity, be viewed as a penalty on all sighted people. Tax liability would be exactly the same if it were presented on the tax form as a higher tax on sighted people than on the blind. Of course, sighted people may be willing to pay higher taxes to subsidize the blind. The perception of this tax expenditure would, however, be different if it were presented in this way.

The same can be said for most tax expenditures. I opened the 1995-1997 Minnesota tax expenditure report at random and found a provision (p. 82) called "Seven-Year Amortization of Reforestation Expenditures." It is a subsidy for timber growers. It costs $100,000 per-year and I don't know if it's a good use of Minnesota tax dollars or not. But when evaluating this or any other tax expenditure we should not forget that non-recipients pay higher taxes because of them.

The bottom line is that for every dollar in revenue lost through tax expenditures, someone else has to pay a dollar more in tax, or lose a dollar in other spending.

Tax Expenditures, Bad or Good?

The fact that a spending program is administered through the tax code does not make it good or bad. Many tax expenditures are, however, problematic in ways that would be unlikely in a direct spending program. The following is a list of some of the problems that are frequently seen.

• High Priority. As discussed, tax expenditures are given an exalted status that they do not necessarily warrant. They are not subject to the normal prioritizing that takes place in government budget-making.
• Distorts the Debate. New tax expenditures are often treated in a budget, and rhetorically, as a tax cut. Removing a tax expenditure is treated as a tax hike. This contributes to the growth of tax expenditures at the expense of other spending without well-considered prioritization.
• Poorly Understood. Placing spending programs in the tax law often hides their function. The tax expenditure for accelerated depreciation on the corporate franchise tax, for example, costs about $75 million per-year. This program, incorporated from federal tax law, essentially gives companies interest free loans when they make capital investments. The public and policy-makers would have a clearer view of this program if it were structured as such through a direct spending program.
• Limited Oversight. Tax expenditures are generally entitlements available to those who meet the qualifications set out in law. There is typically only minimal review of applications. Tax expenditures are not routinely evaluated to determine if they are efficiently accomplishing their goals. If, for example, ninety percent of a research and development tax credit on a corporate income tax goes to companies for work they would have done in the state anyway, it is, arguably, an inefficient subsidy. Yet, structured as a tax expenditure, it is unlikely that policy makers would be aware of this. If the program was structured as a direct spending program it is more likely that its effectiveness would be subject to careful review. And it would be more likely to be structured in a way that would target state dollars to where they are most needed.
• Agency Involvement. Government agencies with expertise in the area covered by a tax expenditure are typically not involved in their administration. Thus, the agency does not routinely assess whether the tax expenditure is the best way to achieve the agency's policy objectives.
• Simplification. Tax expenditures complicate the tax system. The complications go beyond the initial form-filing. Taxpayers, businesses particularly, often restructure their affairs and relationships with other businesses to take advantage of lucrative tax expenditures. The inevitable legislative responses to excess utilization of tax expenditures (the Alternative Minimum Tax, for example) and even more complexity.
• Wasteful. Corporations and individuals sometimes go to great lengths to take advantage of tax expenditures. For example, those who have more tax expenditures than they can use in a year enter into transactions to transfer their tax benefits to those who can make use of them. The most commonly cited example of this is the trading of the benefits of accelerate depreciation and other investment incentives through leasing deals. The costs of implementing tax avoidance schemes can end up absorbing much of the tax break, making it a very inefficient tool for achieving its original purpose. A substantial portion of what are reported as large tax compliance costs in the United States would probably be more accurately reported as "tax avoidance costs."
• Illogical. Many tax expenditures are limited to the amount of tax liability and are worth more to those at higher incomes. The additional standard deduction for the blind, for example, is of no use to a blind person who does not owe personal income tax. Furthermore, the higher deduction is more valuable for those in higher tax brackets. A direct aid program for the blind would never be designed to help the better-off more than middle- and low-income people. This result follows naturally, however, from making the program part of the personal income tax.
• Federal Deductibility. Reducing liability in state taxes deductible on federal tax returns causes federal taxes to be higher. Some forms of direct spending are not taxable under federal law. Thus, using a state tax system to administer a spending program can cause a portion of the spending to leave the state in the form of higher federal taxes.

In addition, there are some arguments made for tax expenditures that are not universally valid:

• Simplification. Simplicity is often used as an argument for tax expenditures. The claim is that by making the program part of the tax system an involved application process and complicated review system can be avoided. In many cases, however, a direct spending program could be developed that also had these characteristics. On the other hand, seen as spending programs, the ease of application and lack of scrutiny over tax expenditures might not be acceptable.
• Free Enterprise. Tax expenditures are sometimes justified on the grounds that they "leverage" private sector investments of some sort-whether it's charitable contributions, or job creation. In any given case, the claim may or may not be true. The administration of the spending program as a tax expenditure, however, doesn't make it any more market-oriented than if it were done as a direct spending program-even if it feels that way to the recipient.

None of the problems of tax expenditures are inherent. For some tax expenditures the issues raised here are not a problem. For many tax expenditures, however-particularly in the corporate area-the record of federal, state and local governments overcoming these problems is not encouraging.

Policy Options Regarding Tax Expenditures

My testimony already has implicitly suggested several ways in which review of tax expenditures could be improved.

The most extreme step would be to have tax expenditures treated exactly as other government spending. Tax expenditures could be made subject to the budget process, scored in the budget as spending and assigned to agency budgets. One objection to this is that it would make taxpayer planning difficult. This can, however, also be said of other spending programs. Besides, it is unlikely that the legislature would take any abrupt actions that caused major disruptions.

Some of the objectives of treating tax expenditures as spending could, however, be accomplished without making this substantial change in the budget process:

• An item-by-item review of tax expenditures could be commenced to determine whether particular provisions would be significantly better administered as direct spending programs.
• Agencies could be assigned responsibility for periodically evaluating tax expenditures in their program areas.
• Greater disclosure by companies receiving tax expenditures could be required. This would allow the public and policy-makers to know where their tax dollars are going and to evaluate whether the tax expenditure programs are accomplishing their goals.
• The Department of Revenue could be required to gather and tabulate more information on tax expenditure recipients. New York State's Department of Taxation and Finance periodically produces its "Analysis of Article 9-A General Business Corporation Franchise Tax Credits." I have attached copies of pages from the most recent such report. Among other things, this report provides information on which industries are receiving credits and shows the amount of credits received by companies in different net income ranges.

Conclusion

Tax expenditures are not good or bad. They are, however, dangerous. They are usually open-ended, poorly understood, little-scrutinized entitlement programs. Any spending program with those characteristics is cause for concern.

There is, however, no requirement that tax expenditures have these characteristics. Steps can be taken, some modest, some significant, that greatly increase a state's grasp of the value of these programs and where they fit in the overall budget framework.