In Depth: Tax Subsidies

Tax subsidies are the result of selective tax legislation1 that benefits particular groups of people or industries in the economy.  In effect, they share the costs of certain actions between the private sector and the government and impact investment decisions by increasing the expected returns associated with a particular pattern of economic activity.  Tax subsidies may be applied in a number of ways to any one or a combination of economic variables (land, labor, capital).
While some provisions (e.g., the general investment tax credits) may be available to an entire class of economic activity, such provisions may still be viewed as subsidies because other classes of economic activity are placed at a relative economic disadvantage.  In this case, for example, the government has made a decision to favor capital-based productive methods rather than alternatives (such as labor).  Similarly, subsidies to new investment favor supply expansions (such as new power plants) over improved efficiency in the use of existing capacity (such as many demand-side management approaches) and constitute a de facto governmental choice of the method by which to meet market demand. 

Tax subsidies are generally measured in reference to a normative or baseline tax system, and estimates assume no other changes in the tax code.  Each tax expenditure is calculated assuming that there is no interaction with other provisions.  As a result, the estimates can't be added directly together without errors.  As it is very difficult to estimate the potential interactions from simultaneous removal of multiple subsidies, though, most analyses do add the tax expenditure values together anyway. 

Since the government forgoes revenue that would have been collected had there been no special legislation and must make up those revenues through higher taxes on other economic activities, these policies have real costs.  These costs are classified as "tax expenditures." Within the United States, we are lucky in that two separate groups (the U.S. Treasury and the Joint Committee on Taxation) both independently estimate tax expenditures associate with current and proposed legislation.  Many US states and countries have no information at all about these special tax rulings.  As a general rule of thumb, where there is no light the largest mushrooms grow.  However, even within the US, the estimates for tax losses for the same provision by the two bodies can differ by hundreds of millions of dollars.  The estimation methods or assumptions are not made public, so improving the accuracy of these estimates is not clear-cut.

The stated goal of tax subsidies, according to the U.S. General Accounting Office, is to promote some policy objective such as "economic growth or a desirable expenditure pattern by taxpayers."2   However, there is a great deal of disagreement over whether particular tax benefits typically encourage "socially desirable" economic behavior.3   Further, even if the policies are effective, they are static and may become ineffective or counterproductive as circumstances (be they demographic, technological, or economic) change.  For example, percentage depletion allowances were significantly expanded when crucial minerals were needed for war efforts.  As these initial conditions changed, the policies did not necessarily evolve with them.

In summary, tax subsidies are neither inherently right or wrong.  They are inherently distortionary, however, in that they alter patterns of economic activity to promote particular areas (targeted by Congress) that would not necessarily have received investment or consumer demand in the absence of the government intervention.  The subsidies need to be considered as a real cost when evaluating alternative long-term energy options.  These costs include the direct cost of increased taxes in other areas to individual taxpayers, and the indirect costs to the economy as a whole through the distortionary effect of the subsidies on R&D, investment, and consumption patterns. 

There are a few issues to keep in mind regarding our net tax expenditure estimates.  First, special taxes on energy have been treated as negative subsidies if they are used for general revenue purposes.  If they are earmarked for specific energy-related uses, such as oil spill cleanup, they are considered user fees and are netted from the total government cost of dealing with the particular energy-related problem.  Second, energy-payments such as royalties reflect a return to the resource-owner for selling the oil or minerals in question, and are not a tax.  Finally, given the fact that the data regarding Treasury losses from tax provisions are somewhat crude and that interactions between the various tax preferences are not incorporated into these data, our quantification of the tax subsidy magnitude should be viewed as an estimate.

How Tax Subsidies Work

Tax subsidies increase expected returns by decreasing the costs associated with taxation.  This is accomplished in four main ways:  providing tax credits; altering the statutory tax rate; altering the taxable basis (i.e., the activities and expenses which are or are not included in the calculation of the tax base); and altering the taxable entity (such as by allowing losses from one corporation to off-set profits of another).  Each of these methods of subsidizing private activity via the tax code has additional variants as well, which are described in more detail below.

Tax Credits

A portion of certain expenditures may be deducted from net taxes owed.

Altering the Tax Rate 

Allowing one type of entity to pay a lower tax than others conveys a financial advantage to the firm with the lower tax rate.4   Three approaches are used to accomplish this end. 

  • Activities Exempt from Taxation.  Certain activities or products may be exempt from tax (e.g., the alcohol fuel partial exemption from motor fuels taxes (replaced with a tax credit at the federal level, but still existent in a number of states), or tax-exempt interest on certain types of bonds),5   although like or substitute activities or products are fully taxed.
  • Entities Exempt from Taxation.  Entire entities, such as some publicly-owned utilities, may be exempted from taxation, although they may compete with other providers of the same product or service that are taxed.
  • Lower Tax Rate.  A particular type of firm or activity pays a lower percentage tax (such as capital gains taxes).

Altering the Taxable Basis

Although the actual percentage tax rate may remain constant, government intervention may redefine which activities must be included in the taxable basis to reduce the resultant net profit figure to which that percentage tax is applied.  These policies encourage taxpayers to shift spending to the activities that will help them reduce their final tax bill.  By altering either the timing or the size of tax deductions, the government creates incentives to engage in particular behavior.  The current deduction for intangible drilling costs associated with oil, gas, and mineral exploration is an example of this type of provision.

  • Timing.  Policies may allow a company to deduct investment or construction costs at a rate far faster than the rate at which the assets are actually consumed (depreciated).  Such intervention goes against traditional accounting methods of capital recovery.6   Allowing current deductions conveys a subsidy by reducing current taxes (thereby increasing current after-tax profits) rather than future taxes.  Since one dollar today could be invested and earn interest, it is worth more than a dollar in the future.
  • Amount.  By excluding certain portions of income from taxation, the government conveys a benefit on methods that produce that type of income.  An example of this is the tax-free dividend re-investment allowance for certain electric utilities (no longer in effect).  In some circumstances, the government may allow the deduction of items that were not actually incurred as a cost.  Percentage depletion deductions, which are based on the gross value of ore mined rather than the investment to mine it, is an example of this phenomena.

Altering the Taxable Entity

By allowing taxpayers to consolidate their tax returns in ways generally restricted by the IRS, the government may facilitate the offset of taxable income in one area with losses from another area.  In some cases, taxpayers may be able to use consolidation to gain more in reduced taxes than they had actually put into the money-losing enterprise.  Redefining the taxable entity gives rise to tax subsidies in at least two ways.

  • Exceptions to General Rules of Taxation.    When Congress makes exceptions to these general guidelines of consolidation, a tax subsidy ensues.  The oil and gas exemption from the passive loss rules (which limit the use of losses from an activity to offset profits in another) is an example of such a variance.
  • Shifting Profits Among Entities in a Vertically-Integrated Corporation.  When the taxable entity is difficult to define and transactions between divisions are not "arms length" transactions, corporations may shift "profits" among divisions through the use of transfer prices in order to reduce total tax burden.  Admittedly, this category of tax subsidies is difficult to measure.  However, there is some historical data suggesting that profit "management" and off-shore shipping subsidiary arrangements in the oil industry (no longer in existence) were used in conjunction with the Foreign Tax Credit provisions to practically eliminate taxes in that industry for a long period of time.7    These concerns were also a driving force behind revisions to rules governing income earned in foreign subsidiaries (Ambler, 3), and behind the recurring concerns on the adequacy of taxes paid by foreign-owned corporations operating within the United States.

The Size of the Benefit

The financial loss to the U.S. Treasury from a particular tax subsidy depends on three factors:  the size of the eligible industry or activity, the magnitude of allowable benefit, and the strictness with which eligibility is interpreted by the Tax Court.  The important point to remember is that a 10 percent tax credit on oil and gas production can yield revenue losses far greater than a 50 percent solar energy credit, simply because of the relative sizes of the two industries.

The creation of an Alternative Minimum Tax (AMT) in the 1980s reduced the benefits of tax preferences for many in the energy sector.  The AMT was developed to ensure that every profit-making venture paid some taxes.  Thus, any eligibility for tax benefits below a company's AMT would not be able to be used.  Conversely, any relaxation in AMT requirements (such as is provided to independent oil and gas producers in the Energy Policy Act of 1992) would result in higher tax expenditures.  The Treasury and Joint Committee on Taxation estimates of tax expenditures used here already incorporate their best-guess on the impact of AMT limits on the size of the subsidies.

The size of the tax expenditure may be measured in two ways: net present value and annual flow.  The net present value (NPV) method evaluates the total value of tax losses from a provision going forward.  This approach is especially valuable when examining the relative costs of alternative policy options to achieve changes in market behavior.  For example, examining the NPV of losses from the oil and gas exceptions to passive loss restrictions would help policy makers determine whether there were more efficient mechanisms to achieve the goal of improved domestic oil security. NPV estimates require assumptions about discount rates, future (potentially long-term) market conditions, the marginal tax rates of taxpayers in each year, and interactions of the tax benefit in question with other tax options.  The NPV method has the added advantage that tax expenditure estimates are never negative (i.e., increasing returns to Treasury) as new activity using them declines.8

The annual flow method examines the reductions in tax collections from a tax provision in a single year rather than the for the entire life of the provision.  The annual flow method is normally used in Earth Track analysis for a number of reasons.  First, data on the magnitude of these losses were available both from the U.S. Treasury and the Joint Committee on Taxation.  Second, the flow-through approach provides the "snap-shot" of total support for energy in a particular year that we were trying to obtain.  Both methods are useful, and the NPV method should be done during consideration of any new provisions.

Treasury and JCT tax expenditure estimates, therefore, represent multiple years of investment behavior.  Many tax subsidies allow items which are normally deducted from taxes over a 20-30 year period to be deducted in 10 years or less.  In this example, for each of the first ten years after an investment is made, the Treasury will collect less tax revenue that it would have without the subsidy.  Other provisions, such as investment tax credits, can only offset a certain amount of income.  As a result, the credits may be "carried forward" and deducted against income in a future year. 

Expanding this example to reflect aggregate investment in the economy means that for any single year of tax expenditure estimates (e.g., 1989), the deductions taken in 1989 from all earlier investments which have not yet been fully depreciated are included.  To incorporate this multi-year aspect of investment tax credits and accelerated depreciation provisions, it is necessary to examine longer term patterns of investment in the energy sector. 

Who Gets the Money

Whether a tax subsidy is available to the producer or consumer of energy, benefits are shared between four parties: the producer (in the form of higher profits), the consumer (in the form of lower prices), the resource owner (in the form of higher royalties or rents), and the worker (in the form of higher wages).  The pattern of allocation often changes over time with market conditions, as the relative scarcity of resources changes or the relative market power of these groups shifts. 

Increasing expected profits to either producers or resource-owners will increase the supply of a material brought to market.  Increasing wage rates will attract more, and perhaps better skilled, workers.  Reducing price to the consumer will increase the demand for the energy source.  In all four cases, money is transferred from the general taxpayers to the specific entity (often a much smaller group of taxpayers, always a different group from the one making consumption decisions) who owns the energy minerals, develops them, or uses them as fuel.  For this reason, we are not concerned with which party ends up with the subsidies, only that the entire amount goes into increasing the attractiveness of a particular energy type.

There are important distinctions among the size of incentives received by various types of taxpayers.  For example, whereas corporate expenditures for improved energy efficiency are tax deductible, individuals are governed by a different set of tax rules.  As a result, expenditures by individuals for exactly the same purpose must be made with after-tax income.9

One other important point about tax subsidies is that they are estimates, and measure the revenue loss or private benefit of a provision given current estimated levels of growth and the existence of the tax subsidies.10   Although the actual deductions taken from a year could, in theory, be calculated from submitted income tax returns, in practice, such analyses are never publicly available.  Therefore, tax expenditure estimates should not be interpreted as estimates of the increase in Federal receipts (or reductions in budget deficit) that would accompany the repeal of the special provisions.11   Such repeal could change economic growth and aggregate income, reducing aggregate demand and therefore overall tax collections.  Second, since tax subsidies are enacted to encourage certain economic activity, removal of these subsidies would most likely yield significant changes in the level of activity occurring in the subsidized areas.  Finally, tax subsidies may, to some degree, substitute for one another.  Therefore, eliminating one or many of the subsidies would yield new tax avoidance behavior by taxpayers, and would affect the expected savings by removing any of the remaining provisions.12

Source:  Doug Koplow, Federal Energy Subsidies:  Energy, Environmental, and Fiscal Impacts -- Appendix B, (Washington, DC:  Alliance to Save Energy), 1993.  © Doug Koplow and Alliance to Save Energy, 1993.

  • 1Once legislation has been passed, rulings by the Internal Revenue Service or the Tax Court may later affect the size of the subsidy by narrowing or broadening the applicability of the provision.  These shifts can be fairly dramatic.  A private letter ruling allowing a substantial portion of ethanol production facilities to be classified as a solid waste facility enabled billions of dollars of new access to tax-exempt debt for the ethanol sector.
  • 2GAO, Tax Expenditures: A Primer, 1979, PAD-80-26, p. 6. The terms "tax subsidy" and "tax expenditure" are used interchangeably in this report.
  • 3Tax subsidies, such as rapid amortization of pollution control equipment, may in fact have social benefits by accelerating the shift to less polluting means of production (though they may hide the benefits of less-pollutingsolutions). However, even here there is controversy over whether the tax benefits encourage behavior that would not have also occurred in the absence of the tax provision. Many tax subsidies, of course, also confer private benefits to energy consumers through lower energy prices, but there is not necessarily a social component to this benefit.
  • 4The incidence of the tax benefit may vary by individual circumstance. For example, tax benefits for certain types of property may increase the market valuation for that property - in effect capitalizing a portion of the tax benefit. The benefit would be shared by the original property owner and by the new purchaser. While the net cost to the federal government (and therefore to the general taxpayer) is the same regardless of how the tax benefit is shared, the incidence of the tax benefit will affect how the subsidy is reflected in market pricing.
  • 5One study estimated that the federal tax exemption on interest from municipal securities reduced the interest costs of issuing those securities by 25 to 30 percent. (Brannon, 19).
  • 6Although the goal is to match the depreciation period of capital with the useful life of that capital, in practice there is some gray area regarding how long the useful life really is. In general, it is much easier to tell when the depreciation period is clearly wrong, such as depreciating power plants over only 10 years, than to tell when it is exactly right.
  • 7See Glenn Jenkins, "United States Taxation and the Incentive to Develop Foreign Primary Energy Sources," in Gerard Brannon, ed., Studies in Energy Tax Policy. Cambridge, MA: Ballinger Publishing Co., 1975, pp. 203-245.
  • 8Tax expenditure estimates for some tax provisions are negative, implying that the Treasury is receiving more money with the subsidy than it would have without it. This enigma may be understood in reference to the timing of payments in the following example. A $10 purchase which lasts 5 years would generate a $2 depreciation charge each year, which is tax deductible. If an accelerated depreciation provision allowed the investment to be depreciated in 2 years, rather than 5, the tax deduction in years 1 and 2 would be $5, but would be $0 for years 3-5. Thus, accelerated depreciation would yield a tax deduction $3 higher in years 1 and 2, but $2 lower in years 3-5. However, the net benefit to the firm is still positive, since it may collect interest on its tax savings from earlier years, if it chose to. In addition, taxes owed in later years may be paid in inflated dollars.
  • 9In 1989 the IRS ruled that cash reimbursements to homeowners for the purchase of energy-efficiency improvements provided through utility demand-side management programs were to be treated as taxable income to the homeowner. (JCT, 3/1/90, 4). The Energy Policy Act of 1992 changed the rules to allow such payments to be excluded (in full by homeowners, in part by commercial entities) from taxable income. Residential purchases of efficiency improvements not paid for via DSM programs are still generally made with after-tax income.
  • 10The Treasury has also developed an "outlay equivalent" measure of tax subsidies. This measure estimates the value of the service to the recipient as if it had been provided by a direct agency outlay. Thus, outlay equivalents are pre-tax values; revenue loss estimates are after-tax. Arguing that the outlay equivalent measures took up too much space and could be calculated independently by individuals, Treasury stopped publishing outlay equivalent values entirely. It would have been more useful had they simply moved this report to a web-only publication, as many other agencies have done.
  • 11Office of Management and Budget. "Special Analysis G: Tax Expenditures," Budget of the United States FY 1985, pp. G15-G17.
  • 12For example, after the elimination of the capital gains tax benefits for timber, Treasury losses from deductions under the expensing allowance for timber interim management costs jumped from almost zero in 1986 to $130 million in 1987 and $278 million in 1988. (Temple, Barker & Sloane, Inc., II-12).