CRS LOOKS AT ECONOMIC EFFECTS OF TAX INCENTIVES FOR ENTERPRISE ZONES.

                       CRS REPORT FOR CONGRESS

SUMMARY

The tax incentives for enterprise zones (EZs) proposed in six bills introduced in the 101st Congress could stimulate substantial increases in EZ economic activity. EZ bills with the most generous capital subsidies would reduce the cost of capital for EZ investments by almost 20 percent (assuming 50 percent tax-exempt financing). All bills but one provide labor subsidies that would reduce wage costs almost 14 percent if 50 percent of total wages is devoted to economically disadvantaged workers. Firms whose production functions are similar to the assumptions used in this report might increase EZ investment between 20 and 8.5 percent, EZ employment between 14 and 6.5 percent, and EZ output between 15 and 7.5 percent.

Even taking into account these potential increases in economic activity in EZs, the EZ program is NOT a Federal revenue raiser. Analysis suggests that tax-induced increases in investment, employment, and output in EZs would be largely offset by the investment, employment, and output foregone in other industries or locations. As a consequence, the revenue increase associated with the increased EZ economic activity would be offset by non-EZ revenue decreases. If some of the EZ increase in economic activity did represent a net increase for the economy, it could have been obtained through some other subsidy program.

Absent a revenue gain, the EZ program must be justified on the basis of its social benefits. The theory behind these bills is that investment and employment directed to geographic areas that have had a history of low levels of economic activity provide benefits to society in excess of what would be obtained if the investment and employment were directed to areas that experienced healthy economic growth. In effect, the EZ program assumes that society values the distribution of the investment and employment growth.

Estimates in this report indicate that EZ investments could earn up to 39 percent lower pre-tax rates of return than are available from alternative investments. This represents a reduction in national income and can be viewed as the cost society pays for the social benefits of EZs.

Congressional EZ proposals present several difficulties in attempting to maximize social benefits per dollar of Federal revenue loss. First, very little of the direct benefits from the capital subsidies (money spent to build equipment and structures) is likely to accrue to EZ residents. Second, much of the labor subsidy is provided to employees who are neither economically disadvantaged nor EZ residents. And third, EZ residents' benefits that accrue through expansion of the State and local tax base will be reduced because higher land values increase the cost of living, State and local governments must themselves provide EZ tax and spending incentives, and gentrification drives some residents to areas not experiencing enhanced tax bases.

                              CONTENTS

The Federal Enterprise Zone Program was created by Title VII of the Housing and Community Development Act of 1987 (P.L. 100-242) and modified by the Stewart B. McKinney Homeless Assistance Amendments of 1988 (P.L. 100-628). The Department of Housing and Urban Development (HUD) is to designate up to 100 zones, 33 of which must be in rural areas. To be eligible for Federal assistance, a zone must receive financial and other support from its State and/or local government. A zone may exist for a period not to exceed 25 years, unless terminated earlier by the Secretary of HUD or the State. The object is to attract new rather than relocated businesses to these zones. ¹

The current Federal enterprise zone (EZ) law provides no Federal tax incentives to stimulate investment and economic activity in the zones. Six bills have been introduced in the 101st Congress to provide the missing tax incentives: Senator Boschwitz (S. 58); Senator Danforth (S. 35); Representative Armey (H.R. 69); Representative Ireland, (H.R. 193); Representative Rangel (H.R. 6); and Representative Stangeland (H.R. 1221).

Proponents of tax incentives for EZs cite two reasons for their use. The first relates to social benefits -- that the economic lives of the people who live in or near geographic areas experiencing declining employment and investment are improved. Second is the possibility of increased Federal tax revenues -- that the increased investment and employment stimulated by EZ tax incentives generate more than sufficient revenue increases to offset the revenue losses caused by the tax preferences.

The first section of this report describes the Federal tax incentives contained in the six enterprise zone proposals. The second section provides estimates of the effect of these provisions on the cost of capital and the after-tax wage cost. The third section assesses the potential effect of the incentives on economic activity in the enterprise zones. The fourth section describes in more detail and reviews the two justifications for enacting enterprise zone tax incentives -- increased Federal revenues and provision of social benefits. An appendix explains how the capital and labor subsidies were estimated.

PROPOSED LEGISLATION

The tax incentives these bills provide are summarized in Table 1. As a group, the bills provide six types of capital subsidies and two types of labor subsidies. The Ireland and Boschwitz bills are almost identical and are grouped together.

CAPITAL SUBSIDIES

INVESTMENT TAX CREDIT: Current law provides no investment tax credit (except for the phase-out of the pre-1986 credit). All EZ bills except for Danforth provide an investment tax credit. The Ireland/Boschwitz, Rangel, and Stangeland bills establish a 10 percent credit for both equipment and structures. The basis of the asset (the percentage of asset cost that can be depreciated) is reduced by the full amount of the credit. The Armey bill establishes a 5 percent credit for equipment and a 10 percent credit for structures.

R&D CREDIT: Current law provides a 20 percent credit for research and development expenditures. The Ireland/Boschwitz, Rangel, and Stangeland bills provide an increased credit for research and development expenditures. The Ireland/Boschwitz and Rangel bills set the credit at 37.5 percent, and Stangeland sets the rate at 30.0 percent.

ORDINARY LOSS DEDUCTION FOR WORTHLESS STOCK: The Ireland/Boschwitz, Rangel, and Stangeland bills allow losses on worthless EZ stock to be deducted as ordinary losses, rather than requiring them to be limited as offset against capital gains.

STOCK PURCHASE DEDUCTION: The Rangel, Danforth, and Stangeland bills provide for a maximum $100,000 deduction against current income for the purchase of stock in EZ companies. The basis of the stock is reduced by the amount of the deduction.

CAPITAL GAIN PREFERENCE: All bills except for Armey provide for deferral of capital gains on the sale of property when the proceeds are reinvested in EZ property within a year. The basis of the new property is reduced by the amount of nonrecognition of gain. The Armey bill excludes capital gains on EZ property from taxation.

TAX-EXEMPT BONDS: All bills allow the portion of an asset financed with tax-exempt bonds to take full accelerated depreciation deductions over the assigned class life of the property rather than restricting deductions to straight line over actual class life. All bills allow small-issue industrial development bonds (IDBs) used to finance EZ property to continue to be tax exempt after the scheduled December 31, 1989 sunset date for small-issue IDBs. The Danforth and Stangeland bills would reserve five percent of each State's volume cap for private-activity tax-exempt bonds for use in EZs. And the Ireland/Boschwitz, Rangel, and Armey bills eliminate the existing capital restrictions for projects financed with small-issue IDBs.

             TABLE 1. TAX INCENTIVES FOR ENTERPRISE ZONES:

Wage incentives are provided to both employers and employees. The two employer credits are interdependent with the existing Targeted Jobs Tax Credit (TJTC). Current law provides a 40 percent TJTC for wages paid for the first year's employment of certain classes of disadvantaged individuals, without regard to the location of the employment. The firm's wage base is NOT reduced by the amount of the credit.

QUALIFIED WAGES CREDIT: All EZ bills provide a 10 percent credit for increased wages paid in an EZ. The base for determining the increase is the wages paid by the employer in the zone in the twelve months PRECEDING designation of the zone. The annual maximum creditable wages per employee is $17,500. The credit is available for an employee's qualified wages over the life of the zone, a maximum of 24 years, and is phased out over the last four years of zone life (reduced by 0.25, 0.50, 0.75, and 1.0). Any employee whose wages receive this credit may not be included in the base for calculating the TJTC. The firm's wage basis is reduced by the amount of the credit.

ECONOMICALLY DISADVANTAGED CREDIT AMOUNT: All bills except S. 35 provide a credit for the wages of economically disadvantaged employees that begins at 50 percent for the first three years the employee's wages are paid, phases down 10 percent a year from the fourth through the seventh year, and remains at 10 percent until becoming zero in the twentieth year. Eligible individuals include those eligible for the TJTC and others whose family income does not exceed the AFDC grant level plus the value of the family's food stamps. Any employee whose wages receive this credit may not in the same year be included in the base for calculating the qualified wages credit. The firm's wage basis is reduced by the amount of the credit.

The annual EZ wage credits cannot reduce the firm's tax liability below its minimum tax. Unused credits can be carried forward 15 years and backward three years, subject to the minimum tax restriction.

EMPLOYEE CREDIT: All bills except S. 35 provide a nonrefundable credit for employees equal to 5 percent of their qualified wages.

EFFECTS ON FACTOR COSTS

The purpose of an incentive is to reduce the price of a factor of production relative to the price of competing factors, thereby creating an incentive to use more of this factor. This section estimates the effect of these tax incentives on the relative prices of capital and labor when used in EZs as compared to their use outside of EZs.

There are two views as to why EZs have up to now experienced little investment and employment. One view suggests that factor costs are higher in EZs than in other geographic areas, and that these tax incentives will therefore help to equalize factor costs spatially. A second view suggests that factor costs do not differ between EZs and other areas. Investment and employment do not occur in EZs due to a variety of social costs (e.g., crime and low education levels) that cause returns to investment in EZs to be less than available in other areas. In this view, factor costs are equalized already, and tax incentives lower EZ factor costs below costs in other areas to compensate for the higher social costs experienced in EZs.

For our purpose, in either case, it is the price of EZ factors RELATIVE to the price of other area factors that is relevant. If it is assumed that other area factor costs do not change (and there is no reason to believe they will change since EZ activity will represent an insignificant portion of the total capital and labor markets), then the change in relative factor price is entirely determined by the change in EZ factor costs. It makes no difference whether the change is computed from an EZ base higher than or equal to other areas' factor costs.

ESTIMATED REDUCTIONS IN THE RENTAL COST OF CAPITAL

This section presents estimates of the effect of each of the EZ proposals on the cost of capital. The derivation of these estimates is detailed in the appendix.

The rental cost of capital estimates in this report assume that the investment is representative of the U.S. capital stock: 5 percent represents R&D spending, of which 60 percent is intangible, eligible for the R&D credit, and expensed, and 40 percent is tangible, not eligible for the R&D credit, depreciated, and eligible for the investment tax credit; of the remaining 95 percent of the investment, 60 percent is structures and 40 percent is equipment. The values of all other parameters are specified in appendix table A and reflect a variety of sources such as Moody's Bond Record and The Economic Report of the President.

Table 2 shows the rental cost of capital for current law in column 1 for three different assumptions about the share of the investment financed with tax-exempt bonds (10, 25, and 50 percent). This cost ranges from $0.1615 per dollar of investment for an asset financed with 10 percent tax-exempt debt, to $0.1548 for an investment financed with 50 percent tax-exempt debt.

The Rangel and Stangeland bills generate almost identical costs of capital. The incentives they offer differ very little -- Stangeland offers a 30 percent rather than 37.5 percent R&D credit. The combination of the four tax incentives (investment tax credit, R&D credit, full accelerated depreciation deductions for the tax- exempt bond-financed portion of the asset, and the stock purchase deduction) causes the cost of capital to decline to $0.1333 per dollar of investment with 10 percent tax-exempt financing. The cost falls further to $0.1245 for 50 percent tax-exempt financing.

Boschwitz and Ireland are identical to Rangel except they offer nonrecognition of capital gains rather than the stock purchase deduction. The cost of capital rises two-tenths of a point compared to Rangel and Stangeland, to $0.1355 per dollar of investment for 10 percent tax-exempt financing and to $0.1268 for 50 percent tax-exempt financing.

The Armey and Danforth bills generate higher costs of capital than the other bills. The Armey bill offers an investment tax credit while Danforth does not, and an exclusion of EZ capital gains while Danforth offers a stock purchase deduction. Compared to current law, the cost of capital for Armey declines only to $0.1478 per dollar of investment with 10 percent tax-exempt financing and to $0.1388 with 50 percent tax-exempt financing. The cost of capital is slightly lower under Danforth.

              TABLE 2. ENTERPRIZE ZONE CAPITAL SUBSIDIES:

This section presents estimates of the effect of the EZ proposals on wage costs. The derivation of these estimates is detailed in the appendix.

The estimates of aftertax wage cost in table 3 make three different assumptions about the proportion of the wage bill devoted to disadvantaged workers. Under current law, a dollar of wages for which 10 percent is devoted to economically disadvantaged workers reduces the cost to $0.96 due to the TJTC. The after-credit cost declines to $0.80 if 50 percent of the wages are devoted to economically disadvantaged workers.

Introduction of the EZ credits reduces the wage cost further. Danforth provides only the qualified wages credit, lowering the wage cost to $0.8984 and $0.7384 for 10 percent and 50 percent economically disadvantaged wages. The other bills all have an economically disadvantaged credit, and reduce wage costs even further to $0.8888 and $0.6902 for 10 percent and 50 percent of wages devoted to disadvantaged workers.

          TABLE 3. ENTERPRIZE ZONE EMPLOYER LABOR SUBSIDIES:

These estimates represent the initial changes in capital and labor costs. ² The equilibrium changes in these costs and the effect on economic activity in EZs depends upon the sensitivity of the demand for capital and labor in EZs to these initial price changes, and the ability of firms to substitute between capital and labor in the production process. The estimates in this section assume a Cobb-Douglas production process and product market. The elasticity of demand for capital, labor, and product is unitary, which means that a 1 percent price reduction results in a 1 percent increase in demand.

The second panel in table 4 contains estimates of the percentage increase in the demand for capital in EZs. These increases are attributable to the tax-induced reductions in the rental cost of capital presented in table 2. The demand for capital increases by a maximum 19.57 percent under the Rangel proposal for an asset financed with 50 percent tax-exempt bonds. Demand increases by a minimum 8.48 percent under the Armey proposal for an asset financed with 10 percent tax-exempt bonds.

The third panel in table 4 contains estimates of the percentage increase in the demand for labor in EZs. These increases are attributable to the tax-induced reductions in wage rates presented in table 3. The demand for labor increases by 13.73 percent under all but the Danforth bill if 50 percent of the wage bill is paid to economically disadvantaged workers. Demand increases by 6.42 percent under the Danforth bill if 10 percent of the wage bill is paid to economically disadvantaged workers.

    TABLE 4. POTENTIAL EFFECT ON EZ ECONOMIC ACTIVITY ATTRIBUTABLE

It must be remembered that the effects for any given EZ might be considerably larger or smaller than estimated here. An EZ attracting firms that financed with less fully taxable debt and more tax-exempt debt or equity, hired a larger percentage of economically disadvantaged employees, and enjoyed a much higher share of factor income going to capital (these estimates assume an average capital/labor ratio of 5) would experience larger increases in economic activity. Capital and labor costs would decline more, and capital's contribution to the price reduction would be larger because its share of factor income would be larger.

JUSTIFICATIONS FOR EZ TAX SUBSIDIES

As noted in the introduction, tax incentives for enterprise zones are frequently justified as a vehicle for increasing Federal tax revenue and for generating social benefits. This section addresses these two issues.

INCREASING FEDERAL TAX REVENUE

The ability of EZ tax incentives to stimulate investment, employment, and Federal tax revenue is an important issue, but it is also a macroeconomic issue. If the attainment of macroeconomic goals requires a larger budget deficit to finance investment, it makes little difference whether this is accomplished by not collecting tax revenue to finance EZs, not collecting tax revenue to finance accelerated depreciation allowances, or increasing Federal outlays for public investment. The choice of which instrument is to be used to attain the goal should be determined by the comparative advantages of the various stimulus programs in achieving other policy goals (which is really what the social benefits discussion in the next section is about).

Perhaps the irrelevance of the revenue question to the EZ debate can be more clearly understood by tracing the effects of EZ tax incentives. The success of the EZ (or any other subsidy) program in stimulating investment, employment, and Federal revenue depends to a significant extent upon the responsiveness of savers to an increase in the interest rate. ³ If savings are responsive to a change in the interest rate, EZ tax incentives can stimulate investment and increase tax revenue, but so can numerous other investment stimulus programs. There is nothing unique about the stimulation properties of EZs.

One must ask what the foregone Federal revenues (that finance the EZ tax incentives) will be used for if they do not subsidize EZs. Either taxes will be lower and aftertax disposable incomes higher, or the Federal Government will use the money to subsidize some other activity. In either case, all the beneficial investment and employment ascribed to EZs will occur in some other industry, activity, or location. These foregone benefits represent the costs from Federal subsidy of EZs, and must be subtracted from EZ benefits to arrive at a measure of net benefits. It does not appear likely that a net benefit will result from such a calculation.

If savings are not responsive to a change in the interest rate, the answer is more obvious. EZ tax incentives could be expected to simply reallocate investment and employment among locations. The revenues associated with the EZ investment can be estimated, but it is still necessary to estimate the loss of revenue from investment and employment displaced by the EZs. The end result is that little or no net revenue gain is likely to occur.

This discussion suggests that, in terms of net revenue gain or loss, it does not matter whether EZs seek new rather than relocated investment. Either one imposes revenue losses to offset the revenue gains. From an aggregate perspective, EZ tax incentive stimulus to net investment and Federal tax revenue must be compared to the effects provided by other subsidy programs.

SOCIAL BENEFITS

The generation of social benefits is a conceptually sounder basis upon which to judge the desirability of EZs. To reiterate, the argument is that guiding investment and employment to geographic areas that have had low levels of economic activity provides benefits to society in excess of what would be obtained if the investment and employment were directed to areas that experienced healthy economic growth. One might say that society values the distribution of the investment and employment growth. Rather than helping the residents of low-growth areas by establishing social transfer programs, the idea behind EZs is that residents' incomes will be lifted indirectly by participation in a growing market economy. Some also argue that such decentralization of economic activity has environmental, law enforcement, and other social benefits.

Such an approach contradicts a long-held belief that social welfare is higher when the needy are helped by making transfer payments (money) rather than by interfering in market investment and employment decisions to raise their incomes. But transfers increasingly have come to be viewed as harmful to recipients' work incentives and long-term economic prospects. In a sense, this argument suggests the "no leaf raking" school of thought is myopic because it seeks distributional justice and short-term efficiency at the expense of long-term efficiency and economic growth. This perspective is used to justify market interference targeted to help the poor.

Once the importance of these social benefits is accepted, the success of the EZ program is to be judged by the increase in incomes and employment of the residents of these areas. These increases occur in several ways. The purchase of capital equipment and structures that are made by residents of EZs would represent direct (but one time) increases. The employment of zone residents in EZ business operations would also represent a direct (and continuing) increase. Indirect increases would result from economic activities stimulated by the formation of the EZ business, e.g., employee purchases of lunches and other consumption items, or the establishment of firms to supply the EZ business. Residents' incomes would also increase due to tax base expansion, which would enable an increase in public services or reduction in local taxes. ⁴

Measuring such changes is very difficult. But the benefit per dollar of Federal revenue loss can be maximized by giving some thought to the appropriate structure of the tax incentives. These six congressional proposals raise several structural issues.

DIRECT SOCIAL BENEFITS FROM CAPITAL SUBSIDIES

Capital subsidies are unlikely to generate substantial direct employment and income increases for EZ residents. By definition, the EZs are areas suffering from a lack of economic activity. The capital equipment purchased and the structures built or rehabilitated are unlikely to use the labor of EZ residents. So there is little direct benefit to residents from capital subsidies. This does not mean that such subsidies should not be provided, for even indirect benefits are not attainable if no investment takes place. It does mean that the value of the capital subsidies must not be automatically counted as benefits to EZ residents, for these benefits are likely to accrue to people currently living outside designated EZs.

DIRECT SOCIAL BENEFITS FROM LABOR SUBSIDIES

Labor subsidies provide an incentive for the EZ firm to hire economically disadvantaged EZ residents whose productivity may be lower than the productivity of nonresidents. Such subsidies clearly provide direct benefits to EZ residents. But the incentive to hire EZ residents is most effective if the subsidy for the noneconomically disadvantaged workers is restricted. This is not the case with the current wage credit proposals.

These proposals provide subsidies to all labor, whether or not economically disadvantaged. All except for Danforth provide an additional subsidy for economically disadvantaged workers. Comparing the percentage reductions in wage costs in table 4, a firm using ten percent economically disadvantaged labor enjoys a wage subsidy of 6.42 percent under the Danforth proposal (with no economically disadvantaged subsidy) and 7.42 percent under the other proposals. In other words, a firm hiring no economically disadvantaged workers receives 86.5 percent of the subsidy received by a firm hiring ten percent disadvantaged workers. Even when one firm hires 50 percent disadvantaged workers, a firm hiring no such workers receives a 7.70 percent subsidy, which is 56.1 percent of the subsidy available to the first firm.

Again, this discussion is not intended to suggest that some wage subsidy to all workers may not be necessary to induce some firms to locate in an EZ. Advantaged workers may find the location undesirable and require some premium to work for the firm. But after some point, every dollar spent on such subsidies lowers the social benefits per dollar of Federal revenue loss. In the extreme case, a firm could hire no disadvantaged workers, leaving EZ residents to benefit from the Federal program indirectly.

INDIRECT BENEFITS OF FACTOR SUBSIDIES

EZ residents' indirect benefits arise due to investments stimulated by the Federally-subsidized investment and the improvement in the EZ's tax base. Measuring these effects is even more difficult than for the direct benefits, particularly with respect to the effect on the tax base. If the EZ is successful, the demand for land will increase. Land prices will be bid up, making it more expensive to live in the area and offsetting some of the improved public services or reduced tax rates financed by the new investment. The indirect benefits to EZ residents may be further curtailed because the State or local government is expected to also provide financial incentives to the firms, thereby limiting the beneficial effects on the tax base. In addition, as land becomes more expensive the process of gentrification may set in, driving some of the economically disadvantaged from the EZ and into areas in which the tax base has not been enhanced.

THE NECESSARY MAGNITUDE OF SOCIAL BENEFITS

No estimates of the social benefits from EZs are available. We can, however, get some idea of the value of these social benefits that would compensate society for the lower returns it earns on these investments compared to what could be earned in the best alternative investment. This provides one benchmark against which to measure the social benefits sought from EZs.

The EZ capital subsidies lower the private pre-tax rate of return required to make an investment profitable. From the first panel of table 5 we know that the required return without EZ capital subsidies is 8.44 percent, 8.19 percent, and 7.78 percent depending on the share of tax-exempt financing used (10, 25, and 50 percent respectively). With EZ capital subsidies, the required return declines substantially.

    TABLE 5. THE EFFECT OF ENTERPRISE ZONE CAPITAL SUBSIDIES ON THE

[APPENDIX CONTAINS EQUATIONS WHICH ARE NOT SUITABLE FOR ONLINE REPRODUCTION]

² The initial change in wage cost is probably overestimated. The employer credit reduces the aftertax wage cost to the employer of hiring any unit of labor, therefore shifting the net-of-subsidy supply curve down by the amount of the employer tax credit. But the employee credit (which is excluded from the estimates) has the effect of raising the aftertax wage rate for each unit of labor the employee offers, therefore shifting the net-of-subsidy supply curve up by the amount of the employee tax credit.

³ The sensitivity of savings to the interest rate is a matter of considerable discussion. Most evidence seems to suggest savings are not very sensitive to interest rate changes, although some estimates do suggest a positive relationship that is significant. Some observers have suggested that, given the uncertainty surrounding this relationship, the most certain method of raising national savings is not through tax policy, but through raising government savings (reducing the deficit). See: vonFurstenburg, George M., and Burton C. Malkiel. The Government and Capital Formation, A Survey of Recent Issues. Journal of Economic Literature. September 1977; and E. Philip Howrey and Saul H. Hymans. The Measurement and Determination of Loanable-Funds Saving. In Joseph A. Pechman, ed. What Should Be Taxed: Income or Expenditure? Washington, D.C.: Brookings Institution, 1980.

⁴ This of course presumes that State and local tax and spending incentives granted to attract EZ firms do not cost the fisc more than is generated by expansion of the tax base.