November 8, 1996
With great eloquence, Judge Learned Hand expressed the dismay many Americans experience when confronted by the complex federal income tax system. "In my own case the words of such an act as the Income Tax ... merely dance before my eyes in a meaningless procession; cross-reference to cross-reference, exception upon exception – couched in abstract terms that offer no handle to seize hold of – leave in my mind only a confused sense of some vitally important, but successfully concealed, purport, which it is my duty to extract, but which is within my power, if at all, only after the most inordinate expenditure of time. I know that these monsters are the result of fabulous industry and ingenuity, plugging up this hole and casting out that net, against all possible evasion; yet at times I cannot help recalling a saying of William James about certain passages of Hegel; that they were no doubt written with a passion of rationality; but that one cannot help wondering whether to the reader they have any significance save that the words are strung together with syntactical correctness."
Perhaps because the governmental entity with which most taxpayers interact directly is the Internal Revenue Service, much of the blame for this enormously complicated system is placed on the IRS. Before this Commission can make meaningful recommendations with respect to the Service and tax complexity, it must determine whether that blame is deserved, or if the ingenious inventors of the tax monster are more likely to be found in congressional committee rooms and on the floor of the House and the Senate. If this Commission can increase public and congressional awareness of the features of the legislative process that give rise to complexity, it may be able encourage appropriate reform efforts. In addition, because the Service is limited in its ability to ameliorate the problems of complexity without a change in congressional behavior, this Commission should focus its energy on issues over which the IRS has more control and where it could significantly improve its service to taxpayers.
Perceptions of Complexity versus Reality
Before we can talk about the complexity of the federal income tax system, we need to define the term. In his study of legal complexity, Peter Schuck identified four variables that determine the complexity of a particular rule or legal system. Dense legal rules are numerous and cover in minute detail the transactions that are subject to them. Technical rules, which often are also dense, require significant levels of expertise and sophistication to understand and apply. A high degree of differentiation among sources of legal authority characterizes a complex regime; in the context of federal business income taxation, for example, taxpayers must be aware of congressional enactments, Treasury regulations, and numerous pronouncements from the IRS. Finally, uncertainty causes complexity, not only for those who are attempting to comply with rules, but also for those who must enforce them. In conditions of uncertainty, the taxpayers, the Service, and sometimes the judiciary must expend resources to determine what the law means, which version of the law was effective when the transaction occurred, and whether that law applies in a particular case.
The nearly three-quarters of individual taxpayers who do not itemize and can file one of the shorter tax forms do not encounter any of these types of complexity in a significant way. Certainly, the Service should continue to strive to improve these tax forms and their instructions in order to minimize whatever technical complexity or density that the average taxpayer faces. Moreover, individual taxpayers with income solely or primarily from wage income subject to withholding do not confront significant complexity; studies indicate that individuals subject to the greatest compliance costs are those who must report investment or business income.
It is in the realm of business taxation that all the factors of complexity come into play. In particular, studies indicate that certain tax provisions -– foreign-source income tax provisions, the alternative minimum tax, and inventory capitalization rules – are the main culprits causing complexity along all the dimensions we have identified. For example, over one-third of the respondents in a study of the compliance costs facing Fortune 500 companies cited the alternative minimum tax as contributing most significantly to their costs. Being subject to the AMT increased incremental compliance costs for such businesses by nearly 17 percent. These same businesses reported that nearly half their costs of compliance (which included tax planning, as well as recordkeeping, filling out forms, dealing with audits, and litigation) related to the foreign-source income provisions, although only about 20 percent of the revenue from business taxes comes from these provisions.
I do not mean to suggest that efforts to simplify in the area of business taxation are not worthwhile endeavors, nor am I arguing that average Americans would not benefit from such simplification. Of course, they would. They are the owners, employees and customers of businesses that now spend tens of billions of dollars to comply with the federal income tax. I am saying that to the extent the rhetoric of simplification leads average taxpayers to believe that they will benefit more directly from a reduction in complexity, that rhetoric is seriously misleading. Moreover, the fact that business taxpayers form the natural constituency for simplification efforts has a great deal to do with the political dynamics that surround deliberations of comprehensive reform proposals. Perhaps this reality suggests that an administrative agency such as the IRS is a more likely source of simplification than Congress, but, as I will now discuss, the Service’s ability to reduce complexity is limited as long as the tax code itself remains unchanged.
Congress’ Role in the Problem of Complexity
The legislation that established this Commission charged it with presenting to Congress recommendations for restructuring the Internal Revenue Service. As you think about proposals that relate to tax complexity, it is worthwhile to determine whether the Service bears the primary responsibility for complexity, or whether the IRS is trying to administer laws that are drafted in ways that necessarily result in tax complexity. If the latter is the case, as I believe, then the Service’s efforts to simplify will be relatively unavailing without changes in congressional behavior. Let me identify a few aspects of tax lawmaking that lead me to place much of the blame for complexity squarely on the shoulders of Congress.
First, modern tax legislation is the result of incremental policymaking, as lawmakers constantly tinker with existing provisions to deal with new developments without altering the underlying structure of the system or of many provisions. A good example of such incrementalism is the recent change in the research tax credit. For years, businesses that were engaging in significant amounts of research and development but that also had substantially increased their sales over the base period had argued that the tax subsidy was not drafted to allow them to take advantage of it. Some members of Congress, including Senator Baucus, former Senator Danforth, and Chairman Archer, worked with experts in the field to reconfigure the tax credit so that it would better achieve its objectives. Yet when the 104th Congress extended the credit, rather than fundamentally changing the provision, it acted incrementally, engrafting on the existing provision a new elective alternative incremental credit computation. Certainly, this incremental change, and the regulations that will be drafted to implement it, will increase complexity along many of the dimensions – density, technicality, and uncertainty.
Similarly the enactment of the alternative minimum tax can be viewed as a result of incrementalism. As you know, the concern that drove the enactment of the AMT was that certain businesses were zeroing out their tax liability by taking advantage of deductions, credits and other tax preferences. Rather than addressing the fundamental question of whether such tax expenditures were warranted or whether taxable income should be aligned more closely with economic income, Congress added a second tax system to the Internal Revenue Code as a backstop to ensure that all businesses paid at least some tax. The effects of the complexity of the AMT are not limited to only those taxpayers who pay the alternative tax. A General Accounting Office study revealed that although only 2,000 large corporations were responsible for 85 percent of the AMT payments in 1992, over 400,000 companies had to file the dense and technical AMT forms, thereby incurring significant compliance costs.
Professor Witte strongly argues that incrementalism is perhaps the tax system’s greatest liability. "The detrimental aspects of cumulative change ... are that policies can be expected to expand in scope, cost, and structure. The results of this expansion are likely to be increasing complexity, limited options, and an unintentional but unavoidable undermining of the original and primary purpose of the policy. For tax policy this means the development of a tax system of enormous complexity, which may have reached the limits of legitimacy, the capacity to meet revenue demands, and the capability of reform."
For the Service, incrementalism means increasingly complex provisions must be administered; increasingly complicated forms and instructions must be developed; and often opaque regulations must be promulgated to harmonize new provisions with existing tax law and to deal with transition issues. This latter task can often be quite difficult. One characteristic of incrementalism is that Congress focuses narrowly on the problem at hand, rather than adopting a larger perspective that would force it to consider the various interactions among tax code provisions.
A second reason for complexity is that Congress must often act quickly and without full information about the consequences of certain tax provisions. It may not be able to foresee all the circumstances in which a particular provision will be implicated or to predict how taxpayers will rearrange their affairs in the wake of the enactment. In such circumstances, Congress might well decide to enact a relatively open-textured provision, leaving it to Treasury and the Service to add specificity after generating more information and developing experience with the law.
This point is a little different from the more typical indictment of legislation as unacceptably vague. Indeed, provisions that do not seem to be vague on their face may turn out to be open-textured. As Frederick Schauer observes: "In contrast to currently identifiable vagueness, open texture ... is the possibility that even the least vague, the most precise, term will turn out to be vague as a consequence of our imperfect knowledge of the world and our limited ability to foresee the future. Open texture is the ineliminable possibility of vagueness, the ineradicable contingency that even the most seemingly precise term might, when it confronts an instance unanticipated when the term was defined, become vague with respect to that instance."
In cases of open-textured legislation, the Service may be blamed by taxpayers for complexity because it has been delegated the task of applying the rule, taking account of the facts of the particular cases before it. To taxpayers, it may seem as though the regulators are unnecessarily adding to the density or technicality of a relatively simple rule, but, in fact, the IRS is appropriately responding to the unforeseen contingency or the unpredicted development. Of course, this point can also be made with respect to vague or ambiguous legislative language that may result from poor drafting or the inability to reach agreement among lawmakers on more precise language. In all these cases, the complexity that comes from uncertainty will also be increased, because taxpayers will be unsure how the Service will fill the legislative gap, and they will expend resources anticipating contingencies, planning for various regulatory responses, and commenting on regulations that propose solutions. At the least, any meaningful assessment of the Service’s responsibility for the problems of complexity must include an analysis of how often the underlying code provision is vague or open-textured.
The final example of the congressional role in complexity relates to the interaction between the federal budget process and tax legislation. Most obviously, the number of tax bills and changes to the tax code has increased substantially in the two decades since the enactment of the 1974 Budget Act. Surges in legislative activity are often directly responsible for surges in regulatory activity, and the latter may be greater in magnitude as the regulators have to deal with unanticipated interactions between new provisions and existing laws.
In addition, budget rules that require revenue neutrality require Congress to offset provisions that result in revenue loss with tax increases. Because members prefer not to vote for tax rate increases, fearing defeat at the polls, they may resort to complicated provisions that may hide tax increases from voters, such as Section 68 (the overall limitation on itemized deductions) and the phaseout of the personal exemption. When members target an existing tax expenditure as a means of raising revenue, affected interest groups lobby to convince them to narrow the provision, rather than to eliminate it entirely. If members respond to this pressure, as they often do, the resulting tax provision is usually significantly more complicated.
Budget constraints also encourage Congress to enact temporary provisions that occasionally lapse causing enormous compliance headaches for taxpayers and the Service. Again, the research credit is a good example, as are many of the so-called "extender" provisions that periodically appear on the congressional agenda. Finally, strategic drafting of tax provisions to circumvent budget constraints by taking advantage of the limited budget window adds to the Code’s complexity. If legislation can be drafted to accelerate the receipt of revenues and to defer revenue losses to the out-years, it is more easily enacted because PAYGO requires Congress to offset revenue loss only within the budget window. Such craftily-drafted provisions are typically more dense and technical, and they add significantly to taxpayer uncertainty. For example, can small businesses be certain that the increase in amount available for section 179 expensing, a change in the law that is phased in fully only in 2003, will not be repealed to raise money to enact a different tax expenditure?
Let me be very clear: I am not arguing that the costs of the budget process in terms of tax complexity are not worth the benefits of the PAYGO provisions and other procedural constraints on the tax-writing committees. It is clear that these provisions have been extremely helpful in controlling the growth of the deficit that occurs because of revenue loss from the enactment of tax expenditures and decreases in tax rates. Moreover, the budget acts have been instrumental in changing the way we think about the tax code; we now realize that many tax provisions are the functional equivalents of federal appropriations and should be analyzed in a similar fashion. The process is not perfect, however, and any attempt to understand tax complexity would be woefully incomplete without an awareness of the effect of the budget process on tax legislation. Again, this feature of tax lawmaking is beyond the control of the IRS.
Complexity as a Result of the Tax Code’s Many Objectives
Finally, tax complexity results in large part from the tax code’s role as one of the primary vehicles for this country’s social and economic policy. Indeed, to the extent that the typical individual taxpayer who does not itemize confronts any complexity, it occurs because of the social goals that the system is designed in part to serve. For example, even on the shorter tax forms, the taxpayer has to contend with the earned income credit, the credit for child care expenses, and now the adoption assistance credit. Perhaps David Bradford best put this part of the complexity problem into context when he wrote: "[T]he annual tax filing does offer distinct advantages as a mechanism for reaching a large fraction of the population in administering an array of public programs. Without this device, we might instead be confronted with a series of separate bureaucracies and procedures to implement programs such as subsidies for child and dependent care or matching grants for political contributions. Then there would surely be a clamor for the introduction of a streamlined, unified administration of federal subsidies to individuals. Perhaps we would view April 15 differently if we called it the ‘annual unified individual tax and subsidy’ filing date."
Again, we have to ask whether it is fair to tell Treasury and the Service that we want them to administer a range of substantive programs – from encouraging the development of the domestic energy business to giving incentives for employers to hire and train certain workers, from encouraging people to invest their capital in owner-occupied housing to ensuring that young Americans can attend certain programs of higher education – and then to hold the regulators responsible for the complex regulations that are required to administer such programs. The more that lawmakers use the tax code to make distinctions among taxpayers on the basis of certain behavior, the more complex the tax system. Indeed, one of the simplest definitions of complexity is that it refers to the number and difficulty of distinctions the rules make.
Even if we decided that the tax system should no longer be used as a tool for economic and social policies unrelated to the task of raising revenue to fund the national government, a degree of complexity is difficult to avoid. Part of the complexity of the current tax system stems from the attempt to define income precisely and so that we account for the increasing sophistication of modern economic arrangements. Drafters of the tax code, and the regulators who administer it, must work to ensure that economically equivalent activities result in equivalent tax liabilities. At the least, equity demands that taxpayers in substantially the same economic situation pay substantially the same amount to the government.
I am not arguing that complexity should be meekly accepted because it is impossible to avoid. Indeed, Richard Epstein, my colleague who recently published a book on legal complexity, persuasively concludes in his discussion of tax laws that "[c]omplex rules for a complex world are an invitation to disaster." Understanding the equity issues at stake and the plethora of financial instruments and arrangements that can be used to reduce tax liability, we need to work toward reducing complexity without sacrificing fairness. As we pursue simplification, members of Congress cannot decry the evils of complexity as they enact dozens of provisions that have more to do with social policy than with raising revenue. The tax proposals discussed during the presidential campaign demonstrate that both parties intend to continue to use the federal income tax system to pursue diverse objectives. The IRS can try to implement these policies in as simple a way as possible, but again its responsibility for complexity should not be overstated.
In conclusion, I thank the Committee for allowing me to discuss these important issues. Perhaps James Madison best articulated what is at stake in this inquiry into tax complexity. In Federalist No. 62, he wrote about complexity and uncertainty: "The internal effects of a mutable policy are still more calamitous. It poisons the blessings of liberty itself. It will be of little avail to the people that their laws are made by men [and women] of their own choice if the laws be so voluminous that they cannot be read, or so incoherent that they cannot be understood; ... [if they] undergo such incessant changes that no [person], who knows what the law is today, can guess what it will be tomorrow. Law is defined to be a rule of action; but how can that be a rule, which is little known, and less fixed?"