Statement of
Donald B. Marron
Acting Director

Economic Issues in the Use of Tax-Preferred
Bond Financing

before the
Subcommittee on Select Revenue Measures
Committee on Ways and Means
U.S. House of Representatives

March 16, 2006

Mr. Chairman and Members of the Subcommittee, I am grateful for the opportunity to appear before you to talk about the economic effects of financing both public projects and private activities with tax-preferred bonds.

In today's testimony, I will discuss the following four points:

• The traditional form of tax-preferred financing--exempting from federal taxation the interest income earned on state and local bonds--is not a cost-effective means of transferring resources from the federal government to state and local governments. Because of the progressive structure of the federal income tax system, the revenue loss that the federal government incurs from tax-exempt bonds exceeds the debt-service savings that accrue to states and localities. More-direct means of transferring resources--for instance, through appropriations--could deliver equal or even greater amounts of aid to the states at a reduced cost to the federal government.
  
  
• Tax-credit bonds--a relatively new development in tax-preferred financing--pay a larger share of state and local governments' borrowing costs than do tax-exempt bonds. However, tax-credit bonds could be structured to pay the same share as tax-exempt bonds at less cost to the federal government.
  
  
• The expansion of tax-preferred financing to private activities raises additional concerns. State and local governments are permitted, within limits, to use tax-exempt financing to support a variety of activities, including aid to local businesses, the financing of housing, and even the construction of sports arenas. Subsidizing such endeavors, however, runs the risk of funding investments that would be made anyway and of displacing more-productive investments with less-productive investments, thereby reducing the value of overall economic production. A key question is whether subsidized investments provide social benefits to the nation as a whole or just to local areas.
  
  
• The tax-administration system is poorly equipped to monitor compliance with the various targeting rules that the Congress has adopted to achieve social objectives. That ability could be enhanced if the Internal Revenue Service (IRS) could make greater use of the information gathered by the issuers of state and local bonds. However, a larger question would still remain: whether it is appropriate or desirable to pursue certain societal objectives through the tax code.

Tax-Exempt State and Local Public-Purpose Debt

Traditionally, the interest income earned on debt issued by state and local governments has been exempt from federal income taxation. That exemption lowers the interest rate that state and local governments must pay on their debt and encourages investment in public facilities. Purchasers of tax-exempt bonds are willing to accept a lower rate of interest than they could receive on taxable bonds because they are compensated for that difference with lower tax payments.

The exemption, which has existed since the inception of the income tax in 1913, had its origins in the belief that such income was constitutionally protected from federal taxation. Although the Supreme Court rejected that argument in 1988 in South Carolina v. Baker, the exemption has continued.⁽¹⁾

The federal government imposes some limits on the amount of such debt that is issued. For example, a government could profit by borrowing at low tax-exempt rates and then investing in taxable bonds. Anti-arbitrage rules contained in the tax code regulate and limit such opportunities. Additional limits are imposed by state and local governments themselves and by the bond markets when questions of creditworthiness result in higher borrowing rates.

In 2005, the outstanding stock of tax-exempt state and local public-purpose debt equaled about $1.3 trillion. According to the Joint Committee on Taxation (JCT), the revenue loss associated with the exemption in fiscal year 2006 amounted to about $27 billion.

As alluded to previously, tax-exempt financing is not a cost-effective mechanism for encouraging the formation of public capital. Because of the progressive rate structure of the U.S. income tax system, taxpayers with lower marginal tax rates receive lower tax savings from the exemption than do taxpayers with higher marginal tax rates. When an issuer must sell bonds to purchasers with lower marginal tax rates, the issuer must set a higher interest rate on the bond issue to compensate those purchasers for their lower tax benefits. As a result, bond purchasers with higher marginal tax rates receive an interest rate greater than they require to induce them to buy the bonds. That windfall gain causes the federal government's revenue loss to exceed the reduction in state and local borrowing costs, perhaps by as much as 20 percent.⁽²⁾ That excess tax benefit is received by bond purchasers with higher marginal tax rates.

In principle, it may be possible to deliver a higher amount of fiscal aid to state and local governments at a lower cost to the federal government if such aid is delivered as an outlay instead of as a tax preference. Such a mechanism, the taxable bond option (TBO), in which the federal government would pay a specified share of state and local borrowing costs, was reported favorably by the House Committee on Ways and Means in 1969 and 1976, and proposed by the Carter Administration in 1978. State and local governments prefer the tax exemption because it is available for any amount of borrowing they choose to undertake, making it operate more like an entitlement. By contrast, a TBO would be an outlay and subject to an annual appropriation process, which would impose a limit on its availability.
 

Tax-Credit Bonds

Tax-credit bonds are a new tax-preferred bond option. They are available as Qualified Zone Academy Bonds, adopted in 1997; Clean Renewable Energy Bonds, adopted in 2005; and Gulf Tax Credit Bonds, recently authorized as part of the Gulf Opportunity Zone Act of 2005. A number of other applications have been proposed, almost all of which are for activities that would have been eligible for tax-exempt financing.

Current tax-credit bond programs provide more-generous subsidies than do tax-exempt bonds. The purchaser of a tax-credit bond receives a taxable tax credit set by the Treasury that yields tax savings equivalent to the interest that would have been earned on a taxable bond. For example, if the taxable-bond interest rate was 7 percent, the bond purchaser would receive a taxable tax credit every year from the Treasury Department equal to 7 percent of the face value of his or her bond holdings. In essence, the federal government pays 100 percent of the financing costs on the bond issue through the tax system. By contrast, a tax-exempt bond pays only about 25 percent of borrowing costs. Nonetheless, the tax-credit bond is more cost-effective than the tax-exempt bond--every dollar of revenue loss is used to reduce state and local borrowing costs.

A variation on the tax-credit bond could be used as a cost-effective alternative to tax-exempt financing. Bond purchasers would receive two payments: taxable interest income equal to their current tax-exempt interest income, and a taxable federal tax credit equal in value to the tax benefits that a tax-exempt bond would have provided to the purchaser with the lowest marginal tax rate. Since the credit rate would be the same for all bondholders regardless of their tax bracket, there would be no windfall gain to taxpayers and the full revenue loss to the federal government would be received as a subsidy by state and local governments.⁽³⁾
 

Private-Purpose Tax-Exempt Bonds

Prior to 1968, the Congress imposed few restrictions on the type of capital facilities that state and local governments could finance with tax-exempt bonds. Over time, state and local officials began to use such funding to finance more than just public capital investment. In essence, they began to perform commercial banking functions, relending borrowed funds to private entities for various purposes. As a result, the share of bonds used to finance business investments and loans to individuals grew. The Congress responded by imposing limits on the issuance of bonds for those "private activities"--restrictions that have gradually been relaxed since 1986.

Currently, the outstanding stock of private-purpose tax-exempt debt totals about $315 billion. According to the JCT, the revenue loss associated with the exemption--including state and local funding for housing (rental and owner-occupied), student loans, industrial development, transportation, nonprofit institutions, energy, and waste disposal--amounts to about $7 billion for fiscal year 2006. The Congress set the ceiling on the annual volume of private-activity bonds to rise gradually to a maximum of $80 per state resident in 2006. In addition, the Gulf Opportunity Zone Act of 2005 provided for increases in that ceiling for the areas affected by Hurricanes Katrina and Rita.

The expansion of tax-exempt financing to private activities raises additional concerns besides excess lost revenue. Private-activity bonds subsidize some investments that would be made without the subsidy--in effect, transferring resources to private investors. Private-activity bonds also distort the allocation of capital investment and thereby reduce the nation's economic output. They do so by subsidizing investments that would otherwise not be made, channeling scarce private savings into investments that have a relatively low rate of return.

Companies will not undertake investment projects unless they expect a return that is at least equal to the next best alternative use of their funds. If they can obtain bond financing at a lower rate, the profits (net of tax) that may accrue to the owners are increased. Thus, if they have a choice between two investments, one that can be financed with tax-exempt bonds and one that cannot, the one with tax-exempt funding does not have to be as profitable or productive. Because the tax-exempt subsidy does not increase the supply of funds in capital markets, investment in the economy may flow from activities that yield a higher private return to those that yield a lower return. As a result, the value of total economic output may decline unless the tax-subsidized activity has sufficient social or public value to compensate for the lower private return. Given financial returns in today's economy, a manufacturing firm that invests in a project made profitable by substituting a small-issue industrial-development tax-exempt bond for taxable bond financing might impose annual costs on the economy that average more than $22 per $1,000 bond.⁽⁴⁾

Most social benefits can be measured qualitatively, at best, so making judgments about whether such subsidies are worthwhile is difficult. Restrictions on private-activity bonds were implemented as a means to control the loss of federal revenue and national income from private projects lacking social benefits.

When considering limiting the scope of private-activity bonds, it is important to distinguish between local and national social returns. For example, bonds issued for a nonprofit hospital may have a presumption of providing social benefits to the community that can arguably be said to extend to the nation, such as contributions to the control of communicable disease and basic research in teaching hospitals. But some activities that are financed with tax-exempt bonds may lack such presumptions. That is particularly true when benefits are strictly local rather than accruing to a broader population.

For example, small-issue industrial-development bonds are used to finance investments by manufacturing companies. Since no presumption exists that those companies are providing goods that are materially different from other unsubsidized manufacturing competitors, nationwide social benefits of a conventional nature are unlikely. State and local officials' desire to subsidize those investments is based on their belief that the investments are effective tools to stimulate local economic development. However, the success of the bonds in achieving that goal is not necessarily beneficial to federal taxpayers. The subsidy might make the community where the subsidized firm is located better off than it otherwise would have been, but other communities may be made worse off. Federal taxpayers as a whole would not necessarily gain. In effect, the social benefits may not be adequate to offset the loss of national income and the reduction of the federal tax base, unless federal taxpayers' objective is to reallocate investment within the United States.

Trying to restrict the use of tax-exempt borrowing authority for private activities may not prove successful in all instances, however. Even with limits on or elimination of tax-exempt private-activity financing, states and localities may find ways to continue funding those activities through their regular public-purpose bond issues. For example, the Congress prohibited the issuance of private-activity bonds for professional sports stadiums in 1986. Yet some communities consider the funding of those stadiums to be so important that they are willing to finance them with general-obligation debt, pledging their taxing power as security for the bonds. Because one community's successful acquisition of a franchise comes at the expense of all remaining communities without a franchise, the federal tax dollars provide no benefits to federal taxpayers as a whole. Similarly, states and localities can circumvent the limits on financing private activities by undertaking the activities themselves in partnership with private firms.
 

Administering Public Policy Through the Tax System

From an administrative perspective, much of the complexity in tax law that relates to tax-preferred financing stems from the use of that funding for private activities. The Congress limits the issuance of tax-preferred bonds by restricting ("targeting") private use to those selected activities and users that are enumerated in sections 141 to 150 of the Internal Revenue Code. For example, the issuance of mortgage revenue bonds and rental housing bonds requires that numerous provisions relating to income eligibility and housing prices be satisfied. Similarly, rules governing the issuance of small-issue industrial-development bonds require that the use of such bonds be restricted to companies with limited amounts of capital investment. Virtually every type of private-activity bond has similarly detailed targeting criteria.

Private legal counsel must certify that a bond issue complies with federal tax law. After issuance, most monitoring of a bond issue's tax-law compliance takes place at the state and local level. The extent of monitoring among state agencies that issue mortgage revenue bonds, hospital bonds, higher education bonds, small-issue industrial-development bonds, and so on, varies widely. No requirement exists for bond issuers or their support organizations to report on their compliance with targeting rules, and state and local information is not shared systematically with the IRS.

As a result, the extent to which compliance with federal eligibility rules is maintained over the life of a bond is unknown. For example, mobility and the changing income characteristics of tenants may render a rental housing project ineligible for continued use of multifamily rental housing bonds. Recipients of mortgages financed with owner-occupied housing bonds may sell the house at a time that triggers a requirement to repay the subsidy. And manufacturing companies that use small-issue industrial-development bonds may be acquired by firms whose capital-acquisition history makes them ineligible to use such bonds. Many other requirements could be cited.

To determine whether compliance problems exist, the IRS has established a program to sample bond issues for a particular private activity. The program is not comprehensive, however. Compliance could be enhanced if state and local organizations were required to monitor compliance and report their findings to the IRS.

The discussion of administrative difficulties associated with private-activity bonds raises a larger question, one that applies to tax preferences in general. It is not always clear from the perspective of public administration that the tax system is the best way to pursue certain social objectives. For some objectives--such as those that are means-tested--the tax system may lend itself to fulfilling social goals because of the information it compiles on taxpayers' income status. But in general, a bureaucratic apparatus designed to collect revenue may be poorly suited to administer what are essentially spending programs.

There are two reasons for that. First, the administration of social programs may serve to divert the attention of tax administration from its principal purpose. Goals as divergent as collecting revenue and regulating state and local support of certain private activities may be difficult to pursue simultaneously.

Second, many government programs are subject to periodic review and evaluation to determine how well they achieve their objectives and whether their benefits exceed their costs. That effort requires coordination within the executive branch to provide economic analysis and performance evaluation and provides a basis for regular Congressional oversight. Such efforts may be more effectively undertaken in the context of similar programs and by agencies with specific programmatic missions.