Multimedia | Budget, Taxes, and Public Investment

Do-it-yourself tax cuts: The crisis in U.S. tax enforcement

Download PDF

April 12, 2005 | EPI Briefing Paper #160

Listen to an April 12 news forum featuring the author and six other tax experts

Do-it-yourself tax cuts

The crisis in U.S. tax enforcement

by Max B. Sawicky

According to the latest estimates, as much as $353 billion in taxes–16% of the total owed–went unpaid in 2001. If we ranked states in order of aggregate individual income tax liability, this “tax gap” would far exceed total federal taxes paid by the residents of any single U.S. state and the total taxes paid by all 29 states combined with the lowest total tax liability.

More importantly, recovery of unpaid taxes would eliminate the bulk of projected federal budget deficits over the next 10 years. Burgeoning budget problems (Sawicky 2005; Price and Sawicky 2004) and the associated difficulties of enacting tax increases make closing the tax gap a salient political issue and an important policy priority. Indeed, one reason to resist tax increases is the knowledge that many others are escaping their own tax obligations.

This Briefing Paper provides an overview of what a former commissioner of the Internal Revenue Service (IRS) calls “the crisis in tax administration” (Aaron and Slemrod 2004), and it discusses some remedies to address this problem. Because this Briefing Paper aims to provide a non-technical overview of the topic, citations have been kept to a minimum.Data on the tax gap are from the IRS (2005). The benefit/cost ratios of assorted enforcement measures are from Plumley (1996; 2004). Information on the IRS workload, audit rates, and funding is from Rossotti (2004), Burman (2004), and the IRS (2005). Some of the most recent research on tax enforcement can be found in Aaron and Slemrod (2002).

The Tax Gap

For fiscal year 2005, over $2 trillion in federal taxes will be collected. The budget of the Internal Revenue Service is about $10 billion. To a great extent, U.S. tax collections benefit from widespread, voluntary compliance with the tax code. The extent of this compliance lifts an enormous burden from government.

Of course, to some extent the voluntarism is not an act of civic-mindedness but instead inspired by a desire to avert the threat of IRS enforcement measures. In this sense, the slim $10 billion spent by the IRS has some ripple effect on taxpayers who never directly run afoul of tax collecting authorities. The estimated effect for tax year (calendar year) 2001 amounted to $12 of regular tax payments for every dollar collected through an audit.

Even so, estimates of taxes owed but not paid are significant. As noted above, unpaid taxes in 2001 range between an estimated $312 billion and $353 billion, as opposed to $1,767 billion paid “voluntarily and timely.” (All estimates are in dollars and do not necessarily reflect proportions of taxpayer types.) Payments made late and because of enforcement measures closed the gap by $55 billion, for a net of $257 to $298 billion in missing money. These numbers do not include illegal activities in the underground economy, but they do reflect taxes due on otherwise legal economic transactions, such as “working off the books.”

Until recently, a leading concern in tax enforcement has been the quality of data available to the authorities. Information on the sources of non-compliance enables the IRS to target potential evasion and better allocate its extremely limited resources.

For nearly two decades, the IRS has been limited to data on taxpayers collected for tax years 1979, 1982, 1985, and 1988. This resulted from auditing a representative sample of taxpayers, in addition to other audits and research. In general the operating assumption has been that the pattern of non-compliance is unchanging since 1988, an assumption virtually certain to be wrong, but nevertheless the best benchmark possible in the absence of more current data.

At the time of this writing, new information about compliance is just becoming available to the IRS in the form of the IRS National Research Project (NRP). The range of the tax gap reported above is a preliminary NRP finding. Given the limited release of new data thus far, the similarity of the latest findings to those based on 16-year-old information might be surprising. The potential change in the noncompliance rate for 2001–the tax gap divided by total estimated taxes owed–changes by less than two percentage points: from 14.9% (based on 1988 data) to as much as 16.6% (based on the most recent data).

The NRP’s focus is on measuring underreporting in the individual income tax. Research on corporations, partnerships, and other types of taxpayers will come later, notwithstanding the high rates of non-compliance that can be found among these groups.

Components of the tax gap

Most of the tax gap — $250 to $292 billion — is founded on the underreporting of net income (i.e., reporting too little income and/or too many deductions). A smaller portion of this tax gap is due to non-filing ($30 billion) and underpayment ($32 billion). Nearly all of the estimated non-filing and underpayment pertains to the individual income tax.

In the underreporting category, the taxes most often underreported are: individual income (between $150 and $187 billion), corporate income ($30 billion), payroll ($66 to $71 billion), and estate and excise ($4 billion).

Within the income tax category, about half of the gap is due to underreporting of business net income. Lesser shares are due to underreporting of non-business income, and less still to incorrect reporting of deductions, exemptions, adjustments, and credits ($25 billion). In the payroll tax category, the bulk of non-compliance is attributed to the self-employed ($51-56 billion).

Confidence in these estimates varies, given the spotty character of available data and the idiosyncratic difficulties in analyzing assorted components of the gap. Estimates of the largest pieces–underreporting of business income, corporate income, and self-employment income–used to be considered the less-reliable estimates; presumably with the new NRP data, confidence has improved.

Considered in terms of the percentage of dollars lost by tax category, the three leading categories are self-employment tax (nearly 60%), corporate income tax (over 40%), and the estate tax (over 20%).

Another breakdown is by “visibility” categories within the individual income tax. The vast bulk of underreporting is attributed to nonfarm proprietors, rents and royalties, farm income, and informal suppliers (“proprietors who operate in an informal business style,” which includes street merchants and the like), among other areas. These types of taxpayers are required to conduct “little or no information reporting” of their financial affairs (Plumley 2004).

One way to consider the extent of unreported income is to compare data from the Internal Revenue Service with other sources. The Bureau of Economic Analysis of the U.S. Department of Commerce prepares data on the national economy in the National Income and Product Accounts (NIPA). This includes estimates of the “AGI Gap,” which is the difference between adjusted gross income measures implied by returns filed with the IRS and the income level data gathered byother sources. Figure A shows the extent of discrepancies by type of taxable income. The total for 2002 exceeded $900 billion.

Figure A

Leading Enforcement Problems

A basic distinction in tax enforcement is that between evasion and avoidance. Evasion pertains
to illegal under- or non-payment of taxes. Avoidance refers to legal rearrangement of one’s financial affairs to reduce or eliminate tax liability. In the many areas where the law is vague, owing to the complexity of the issue, the line between evasion and avoidance can be difficult to trace. The general intent of the law is to deny tax benefits to transactions or arrangements motivated solely by tax considerations, but spelling out that aim in practice is daunting. This overview is mostly concerned with evasion.

Tip of the iceberg: the Earned Income Tax Credit

One of the most talked-about lapses in tax compliance is unjustified claims for the Earned Income Tax Credit (EITC) (Sawicky 2002). The EITC is a refundable tax credit available to workers with low incomes. A narrow concern with the EITC could not possibly be justified by worries over insufficient total revenue collections. EITC over-claims are but one piece of the overall gap, perhaps $9 billion. Returns from improved tax enforcement surely can be found in many other places, as well.

Non-compliance under the EITC does not necessarily signify dishonesty. The data do not reveal who is consciously cheating and who is not. They do show, moreover, that the complexity of the EITC often provokes claims for too little a credit, as well as for too much (Wasow 2002). In fact, a recent study found that about four in every 10 low-income taxpayers had never heard of the EITC (Maag 2005). Given such a common lack of knowledge, it should not be surprising that some persons file incorrect claims for the credit.

The IRS is instituting elaborate, extraordinary enforcement measures directed at EITC compliance, raising questions about fairness in the targeting of enforcement efforts. In 2003, the IRS requested a 69% increase in funds for EITC enforcement but only a 3% increase for enforcement in all other areas (Burman 2003).

Tax shelters

In contrast to low-income persons filing over-claims for the EITC, the typical abusers of corporate tax shelters are likely to understand their tax obligations very well. Estimates of losses from underreporting by corporations are $30 billion for 2001, of which tax shelters accounted for between $10 and $15 billion annually (Stratton 2005).

What is a tax shelter? In the current income tax system, tax rates can vary according to the type of income, the type of asset generating the income, the form of legal organization of the business firm, the manner in which business investment is financed, and a host of other factors. For individuals and business firms with complex financial affairs, there is usually flexibility in how to organize or account for these activities. A basic means of tax minimization is to receive income under circumstances in which tax rates are low, or absent, and to structure deductions where tax rates (hence the values of the deductions) are high.

In some circumstances “tax planning” of this type is permitted by law, while in others it stretches what the law allows, and in others still it can be in blatant violation of the rules.

Litigation is a key IRS tool in combating abuse, but since it is expensive, resolution of non-compliance allegations is often based on compromise rather than strict judgment of guilt or innocence.

Increasing workloads, decreasing staff, the statute of limitations, and the “hazards” (uncertainties and costs) of litigation put pressure on the IRS to settle out of court. Taxpayers are likely to anticipate higher net benefits from tax litigation than the IRS, putting downward pressure on the amount of any out-of-court settlements (Plumley 2004).

The government cannot afford to investigate and pursue every case to a judicial conclusion. In effect, there is an incentive for well-heeled corporations and individuals to make aggressive claims in their returns, banking on the IRS seeking other enforcement targets, or on using their legal resources to win major concessions from the under-staffed, under-funded authorities.

International transactions

Transactions involving foreign countries magnify two dimensions of the tax enforcement burden.

One is the added complexity of tax law applying to multinational corporations and cross-border transactions. Complexity and ambiguity in tax law facilitate tax shelter abuse.

A second is the added information burden on authorities, since the foreign side of transactions is more difficult for the IRS to monitor and investigate. Much tax collection from business firms relies on accurate and honest self-reporting of income and expenses. Tax havens with bank secrecy laws, whose financial institutions are patronized by U.S. taxpayers, raise obstacles to tax enforcement. In fact, compared to some Western European countries, the U.S. itself is something of a tax haven (Sullivan 2004).

Vito Tanzi, a tax expert and former official of the International Monetary Fund, has estimated that $7 trillion in assets generate income not reported to any nation’s tax authorities. In 2001, Manhattan District Attorney Robert M. Morgenthau testified before the U.S. Senate that, among off-shore tax havens, Grand Cayman alone has $800 billion in U.S. dollar deposits, and that amount was increasing by $120 billion annually.

The real story on offshore wealth that has been spirited out to tax havens is the lack of solid information on the amount, the source, the ownership, and the legality of the activities that generated said wealth. The national security dimension of this problem in the wake of the terrorist attacks in 2001 has become obvious.

Former IRS Commissioner Charles O. Rossotti testified that the extent of lost revenue could be as much as $40 billion. IRS consultant Jack Blum suggested a higher lost revenue of $70 billion in 2002 alone (Sullivan 2004).

There is no shortage of business enterprises–including respectable ones with famous names–providing “wealth management services” to the rich, services that entail avoiding “unnecessary taxes.” As for other enforcement matters, the amount of foregone revenue is only half of the concern. The integrity and fairness of the tax system affects the strength of voluntary compliance, without which the finances of the U.S. government would come crashing down.

Trusts, partnerships, and S-corporations

Trusts, partnerships, and S-corporations, also known as “pass-through entities,” are organizations that stand between an individual or individuals and the income and losses being generated by a business firm or investment portfolio. They serve various legitimate purposes, but they are also used by some taxpayers to hide income or generate artificial losses to offset taxable income. These pass-through entities may serve no business purpose at all, other than to evade or legally reduce taxes.

“Small” businesses and the self-employed

The small business and the self-employed tax grouping is highly diverse, though its members are dealt with under a single branch of the IRS. It encompasses a wide range of incomes and stations, from street peddlers to corporate law practices. Partnerships and S-corporations might have relatively few employees, but 60% of income to such entities is received by taxpayers with $500,000 or more annual income (Slemrod 2002). By contrast, the incomes of sole proprietors resemble more the population at large. As noted above, the non-compliance rate for this group is extremely high–approaching 50%.

The tax enforcement problem is particularly acute for assorted sub-groups in this category, although in different ways. At the low-income end, there are taxpayers who are not compelled to maintain detailed records of their business transactions, as well as those who are able to do business in cash. For many–especially in the cas
e of farm income–“sole proprietor” incomes are negative and offset other taxable income. (Many such cases are part-time farmers.)

At the high end, in the case of partnerships and S-corporations, there are taxpayers with the resources to structure their activities in complex ways that walk the line between legal avoidance and illegal evasion.

In both cases, to an important extent paying taxes is done on the honor system, albeit under the shadow of possible IRS intervention.

The economies of enforcement are less favorable in this context as well. A given case may involve a minimal amount of recoverable taxes, relative to the costs of auditing and prosecuting tax evasion. Conversely, a given target of investigation may mobilize formidable legal resources in its defense.

Enforcement and the IRS

Expanded measures for enforcement of the tax code would be expected to pay for themselves in improved collections. Estimates of the extent of the payoff vary, depending on how enforcement is intensified. IRS Commissioner Mark Everson believes the benefit/cost ratio is more than four-to-one (Kenney 2005a), which means a dollar more for enforcement brings at least $4 back to the Treasury and a net $3 reduction in the budget deficit (or increase in the budget surplus). We noted at the outset of this Briefing Paper that the $10 billion IRS budget (which also goes to functions other than enforcement, such as taxpayer assistance, and processing over 130 million individual income tax returns) accounted for $56 billion in additional collections in 2001.

Of the taxes that remain unpaid, some underpayments are actually already known to the IRS, but the resources to recover this debt simply are not currently available. Former Deputy Assistant Secretary for Tax Policy Leonard Burman, citing former IRS Commissioner Charles Rossotti, estimated the known arrears to be $30 billion. In his view, this could have been recovered with an additional $2.2 billion in resources for enforcement.


There is little dispute that the workload of the IRS has been increasing more rapidly than its organizational capacity. The number of tax returns has increased steadily, partially due to a growing population and economy. There is also evidence, however, that returns have increased faster than population growth, in the form of more returns for unmarried persons and for children. Another dimension adding to the increased IRS workload involves high-income persons and the growth of pass-through entities (as noted above). Still another factor straining IRS capacity is the multiplication of highly complex and specialized financial instruments. The estimated overall increase in workload between 1992 and 2002 is 16%.

The complexity of the tax code has increased by recent legislation. One example was the tax treatment of capital gains in 1997 legislation. A second example is the growth of benefits for education. A third is the increasing variety of tax-advantaged accounts for retirement. On the corporate income tax side, there are new provisions passed in 2004 aimed at promoting exports without violating World Trade Organization rules on discriminatory subsidies to industry.

The burden of increased enforcement tasks was further magnified by the diversion of IRS personnel away from enforcement and into other tasks. These included assisting taxpayers in return preparation; administering the expansions of refundable credits available under the individual income tax in 1993, 1997, and 2003; and regulating “527” political committees.

At the same time, IRS resources to finance personnel have decreased since 1992. The estimated reduction of “full-time equivalent” staff from 1992 to 2001 is 20,000 positions. Rossotti (2004) sums up as follows: “If the IRS staff grew by 2% per year through 2010, [then] the total staff would still be smaller than it was 20 years earlier (1990), while the economy is projected to be 86% larger and the tax system far more complex.”

The IRS Oversight Board called for a 13% percent increase in the IRS budget for fiscal year 2006, 9% more than the Bush Administration budget proposal (Kenney 2005b).


In September 1997 and April 1998, the Senate Finance Committee under the chairmanship of William Roth (Rep.-Del.) held hearings on the practices of the Internal Revenue Service. Witnesses painted a black picture of an agency run amuck, trampling on the rights of honest taxpayers. The hearings provided a political launching pad for the Internal Revenue Service Restructuring and Reform Act, which among other provisions inaugurated a “Taxpayers Bill of Rights.”

Subsequent investigation by the General Accounting Office–known now as the Government Accountability Office–resulted in a report not released to the public. This and other independent investigations found that testimony offered by witnesses at the hearings could not be substantiated (Donmoyer 2000).

In any case, the 1998 act tilted the playing field in tax disputes in favor of taxpayers suspected of non-compliance. The law also specifies misconduct that is grounds for IRS employee dismissal, known among IRS personnel as “the 10 deadly sins.” The law further established a Treasury Inspector General for Tax Administration (TIGTA) with a mandate to investigate abuses by IRS staff. Oversight of the IRS by TIGTA, the Governmental Accountability Office, the Office of Management and Budget, and six Congressional committees consumes IRS resources (Rossotti 2004).

The question is the extent to which these new rules and administrative institutions block inappropriate practices by the IRS, as opposed to discouraging vigorous enforcement of the law in the legitimate pursuit of tax evasion.

An evaluation of the so-called “Taxpayer Bill of Rights” is beyond the scope of this analysis. What is clear is that the implied enforcement burden on the government has become heavier as a result.


In contrast to the trend of regulating IRS personnel is the new initiative to contract out for collection of delinquent taxes. Legislation was passed in 2004 to permit the IRS to use of private sector debt-collection agencies, which would be paid commissions based on the amount of dollars recovered. Concerns about possible abuse by IRS agents would presumably apply at least to the same extent to the private sector for-profit firms that are paid on the basis of how much money they retrieved. This experiment in tax collection has yet to be implemented.


Document matching

The IRS is able to independently crosscheck taxpayer reporting of certain types of income–wage and salary, interest, dividends–under its “Information Returns Program” (IRP), which requires employers and financial institutions to report this information directly to the IRS. The use of computers is an obvious boon to this “document matching” function. If regulations for more comprehensive reporting of tax information by third parties can be established, such crosschecking could be even more valuable in collecting taxes.

Expanded document matching might also apply to deductions, as it does to some extent already. For instance, mortgage interest and state and local income and property tax liabilities are reported to the IRS by mortgagors and state governments, respectively. In the United Kingdom, non-profit institutions report charitable donations made in cash, facilitating their “return-free” system.

Complete coverage of such information may not be necessary for effective enforcement. Plumley (1996) suggests that because taxpayers cannot be certain whether their information is collected by the IRS from third parties, the risk-averse among them will act as if it has been and their com
pliance with tax obligations will be enhanced. Conversely, insofar as coverage approaches 100%, taxpayers will act under the certainty that the IRS computers will catch any discrepancies between information reported by them and by third parties.

The cost of crosschecking tax information is nearly negligible–about three cents, and the estimated payoff averages $21.00 (resulting in a revenue/cost ratio of 668-to-1.)

Tax Delinquency Investigations (TDI)

When the IRS fails to receive a return from someone it has reason to believe is required to file, it sends a notice by mail. The reason could be that the taxpayer filed the previous year, or that a document match suggests a return ought to be filed. If there is no response to repeated TDI notices, further investigation can be triggered.

The benefit-to-cost ratio estimated for this device is very high (3,766-to-1). A TDI notice costs 31 cents, and the average return per dollar spent on notices is over $12,000.

Audits: enforcement and deterrent

There are audits, and then there are audits. The traditional process of a taxpayer reporting to an IRS field office, to subject himself and his personal or business records to detailed examination, is only one device. A second is known as the “correspondence audit,” which entails a letter from the IRS to the taxpayer narrowly focused on specific questions about the person’s tax return.

The “audit start rate” in 1991 was estimated at 0.65%. Raising it one percentage point to 1.65% would yield estimated net revenue of $56 billion on an annual basis. The average cost of an audit was $1,300. Its average yield was $71,000. As noted previously, by far the greater part of this is the indirect impact on taxpayer behavior, on the part of those who are never themselves audited.

Over time the audit rate has fallen significantly. In 1978 the overall rate for individuals was 2.15%; by 2001 it was 0.58%. From 1995 to 2001, the rate for business income fell from 4.0% to 2.0%. From 1993 to 2001, for corporations the rate fell from 3% to below 1.0%.

Since 1996, audits first took a deep dive, but then recovered to some extent. Individual audits fell from nearly two million in 1996 to about 600,000 by 2000, before climbing back to a million audits by 2004. High-income audits fell from 210,000 in 1996 to 92,000 in 2001, and then more than doubled to 195,000 by 2004. Criminal prosecutions have followed a similar pattern. By 1996 standards, the number of audits is still low, especially in light of growth in the scope of the task.

As a result of an increase in computer-generated correspondence audits (that is, instead of face-to-face audits), the increase in the overall audit rate for individual returned to 0.65% in 2003, up 14% from 2002 (TRAC 2005, cited in Kenney 2005b). Correspondence audits entail a notice sent to the taxpayer suggesting an error in the return and an invitation to rectify it without further penalty. The relevant question is the relative cost-effectiveness of alternative audit procedures, as well as other remedies.

Criminal prosecution

The most costly means of enforcement per case is actual prosecution of tax evaders. Prosecution yields net revenue gains, in terms of more honest reporting of tax information. Here again, the ripple effect is salient in two respects. Well-publicized indictments magnify the threat to others, and they may also raise confidence and voluntary compliance in general, by assuring the public that the law is being applied vigorously to high-income persons.

In 1991, the average cost of a criminal conviction for tax evasion was $103,000, and the average returns in revenue were $1.7 million per conviction.

Property seizures, a prominent feature of IRS horror stories, are actually quite rare. The Treasury Inspector General for Tax Administration–charged with oversight of the IRS–reported that from May to September in 1999, there were 35 seizures in the United States, and all were conducted properly. The IRS was processing about 80 million returns annually in that period (Williams 2000).

Tax experts favoring stronger enforcement differ on the adequacy of the law in respect of abusive tax shelters. On one side is the view that legislation that clarifies what constitutes abuse would aid enforcement. Clarification in this context means specific definitions of where legitimate business purpose ends and tax-avoidance behavior begins. On the other hand, some feel that the lack of specificity in the law affords the authorities some flexibility in discouraging abuse. From this point of view, ingenious, highly-compensated attorneys, accountants, and economists can devise ways to circumvent any new law. More explicit language in law could give rise to the argument that anything not stipulated as out of bounds is automatically permissible.

There is no disagreement that more resources for investigation and prosecution, whatever the state of the law, would advance enforcement goals.

Tax preparation assistance

Filing an income tax return remains a complex task for many taxpayers. Computer software and tax preparation services make the job easier, but the IRS could expand its own assistance in this vein.

There already exists an industry of private-sector tax preparation services. Evidence of abuse by some of these service providers has led to suggestions of a need for regulation and perhaps licensing. Barbers and hair stylists often must be certified by boards, but there is no counterpart for tax preparers. Preparers serve both the “high-end” and the low, as far as taxpayer incomes are concerned. The IRS is concerned about high-end providers secretly marketing abusive tax shelter schemes, as well as about those serving low-income clients encouraging over-claims for the EITC.

There presently exists a network of non-profit tax assistance centers for low-income persons, largely staffed by volunteers. Expanding this network would have two, offsetting impacts on revenue.

By increasing awareness of refundable credits available to low-income taxpayers, these centers would increase IRS cash refunds. At the same time, however, providing more accurate assistance in tax preparation could reduce over-claims for the EITC and other tax-based benefits for families with children. Over-claims are thought to account for roughly a third of the cost of the EITC, so it seems plausible that expansion of tax assistance to low-income persons would save more money than it would cost.

IRS assistance currently consists of answering questions by phone and providing direct assistance in preparation. An average cost of such activities per return was $14 in 1991, and the estimated revenue gain was $5,440.

The shift of IRS resources from enforcement to taxpayer assistance has inspired some criticism, since both activities arguably deserve to be expanded. Defenders of the shift may identify with the assertion of Rossotti, to wit: “I have never understood why anyone would think it is good business to fail to answer a phone call from someone who owed you money.” (2004)

Withholding of tax at source

Employers are required to withhold individual income tax payments from their employees’ wages and salaries. Discrepancies between total taxes paid and owed are settled when the worker files a return. There is no reason why such a practice could not be applied to interest and dividend income. Financial institutions could be required to withhold estimated tax.

For certain forms of consumption taxation, the taxpayer needs to calculate net saving for the tax year. Net saving is the difference between purchases (saving) and sales (dissaving) of financial assets. Given the multiplicity of assets and accounts held by some individuals, Aaron and Galper (1984) suggested the establishment of central asset management accounts. If al
l financial transactions are routed through a single account, the taxpayer is relieved of the need to calculate net saving, and the IRS has an independent source to verify capital income reported by the taxpayer. It might be possible to establish such accounts under the current tax system, to ease the compliance burden on the taxpayer and the enforcement burden on the government. Such accounts would also facilitate tax withholding.

International Cooperation

Intergovernmental cooperation is required to enforce tax evasion that exploits cross-border transactions and nations functioning as tax havens. Such cooperation, at a minimum, entails the sharing of information among tax authorities in order to trace unreported income and verify information reported by taxpayers.

Law enforcement cooperation is also necessary. One egregious example is the case of some wealthy persons who sought to emigrate from the U.S. and establish citizenship elsewhere for the purpose of protecting their capital gains income accrued in the U.S. from taxation.

More ambitious would be efforts to harmonize tax systems among nations. Uniform definitions of net income would facilitate integration of tax enforcement efforts. If such coordination could be accomplished, greater uniformity of tax rates would discourage tax competition among nations and avoid the waste of economic resources caused by relocation of economic and financial activities according to tax considerations, rather than business purpose.

The Organization for Economic Cooperation and Development (OECD) has taken up the problems of cross-border tax enforcement. Collaboration with the U.S. during the Clinton Administration was curbed in 2001 shortly after the election of George Bush. A conservative business coalition campaigned against the OECD’s initiative under the banner of tax competition. The accusation was that cooperation in tax enforcement was aimed at penalizing low-tax nations who refused to raise their tax rates to those in Western Europe, and the U.S. Rep. Nathan Deal (Rep.-Ga.) described the arrangement as “fiscal European imperialism” (Tax Notes 2001).

In fact, the possibility of tax harmonization among the highly diverse tax systems of the OECD nations is close to nil. The most important priority for the OECD was information-sharing that would combat tax evasion. Criticism in this dimension rested on the debatable grounds of taxpayer privacy. The Bush Administration affirmed its opposition to tax evasion, but it also criticized the OECD posture as overly aggressive. Currently, the financial privacy laws of the U.S., Switzerland, Luxembourg, Belgium, and Austria present obstacles to information sharing (Scott 2004). The terrorist attacks of September 11, 2001 led to only limited improvements (Sullivan 2004).

Tax simplification

Complex tax law is required at some level because the financial affairs of wealthy individuals and business firms can be complicated. At the same time, there are approaches to tax simplification that could reduce the compliance costs for taxpayers and ease the enforcement burden on the government.

For millions of taxpayers with modest or low incomes, taxes could be simplified to the point where filing a return became unnecessary. All tax liabilities could be met through an exact withholding system (Gale 2001).

Within the framework of the individual income tax, sets of provisions with similar purposes might be consolidated. One example is tax benefits for families with children (Cherry and Sawicky 2000). Another is the multiple opportunities for retirement savings accounts (Steuerle 1998). In a similar vein, the disparate treatment of different types of income–capital gains, dividends, and so-called “ordinary income” (i.e., wages and salaries, interest income)–could be eliminated in favor of all income being subject to the same set of tax rates. Multiple tax rates have little bearing on the complexity of a tax system; it is the definition of taxable income that creates the greatest compliance burden.

An especially burdensome facet of individual and corporate income taxes, as far as compliance is concerned, is the Alternative Minimum Tax (AMT). Given the expected growth in AMT coverage, the inescapable need for some kind of adjustment also will present an opportunity for reform. One obvious measure would be to replace the AMT with some kind of tighter cap on deductions.

More controversial are suggestions to transition from the current system, which has elements of both income and consumption taxation, to a full-blooded consumption tax. The chief argument here from the standpoint of simplification is that consumption is inherently simpler to measure than income. A shift towards consumption taxation could dramatically simplify rules pertaining to capital gains and depreciation. On the other hand, the transition between systems would create extra complexity, including opportunities for tax avoidance and evasion. Depending on the form of such taxation, rules pertaining to international transactions and financial institutions, among other areas, could remain complicated.


The enormous tax gap exerts a progressively debilitating impact on the fairness and productivity of the tax system. A poorly-functioning tax system impedes economic efficiency and growth. Perceived injustices of the system sap confidence and weaken the public resolve to play by the rules. And now especially, in light of serious deficit problems, upgrading tax enforcement is not only important for the health of the U.S. economy, but it is an urgent priority in fiscal policy. A variety of remedies have been found to provide very high revenue payoffs over and above very modest costs.

There are few free lunches in economics. Better tax enforcement may not be free, but it is one of the best bargains available in economic policy.


Aaron, Henry J. and Joel Slemrod (eds). 2004. The Crisis in Tax Administration. Washington, D.C.: The Brookings Institution.

Burman, Leonard E. 2003. “On Waste, Fraud, and Abuse in Federal Mandatory Programs.” Statement before the Committee on the Budget, United States House of Representatives. July 9.

Donmoyer, Ryan J. 2000. “Secret GAO Report Is Latest to Discredit Roth’s IRS Hearings.” Tax Notes, Volume 87, Number 4, p. 463-65.

Gale, William G. 2001. “Testimony Before the Subcommittee on Oversight and Subcommittee Select Revenue Measures of the House Committee on Ways and Means.” Hearing Series on Tax Code Simplification. July 17.

Internal Revenue Service. “The Tax Gap.”

Johnston, David Cay. 2003. Perfectly Legal: The Covert Campaign to Rig Our Tax System to Benefit the Super Rich–and Cheat Everybody Else. New York, N.Y.: Portfolio.

Kenney, Allen. 2005a. “Everson Evaluates State of IRS, Pledges Strong Agenda for 2005.” Tax Notes, Vol. 106, No. 1, pp. 40-44.

Kenney, Allen. 2005b. “Oversight Board Wants Bigger IRS Budget in 2006 Than Bush, Everson.” Tax Notes, Vol. 106, No. 12, p. 1348-50.

Maag, Elaine. 2005. “Disparities in Knowledge of the EITC.” Tax Notes, Vol. 106, No. 11, p. 1323.

Plumley, Alan. 1996. The Determinants of Individual Income Tax Compliance: Estimating the Impacts of Tax Policy, Enforcement, and IRS Responsiveness. Department of the Treasury, Internal Revenue Service.

Plumley, Alan. 2004. “Overview of the Federal Tax Gap.” Department of the Treasury, Internal Revenue Service, NHQ Office of Research.

Rossotti, Charles O. 2004. Letter to Senator
s Charles Grassley and Max Baucus. March 22.

Sawicky, Max B. 2002. Where the Money Isn’t: Misplaced focus of tax enforcement could be remedied by simplifying credits for children. Issue Brief #183. Washington, D.C.: Economic Policy Institute.

Scott, Cordia. 2004. “OECD Targets More Financial Centers in Tax Haven Crackdown.” Tax Notes, Vol. 103, No. 11, pp. 1347-49.

Joel Slemrod. 2004. “Does the Tax System Penalize, or Favor, Small Business?” In The Crisis in Tax Administration, Aaron and Slemrod, eds. Washington, D.C.: The Brookings Institution.

Steuerle, C. Eugene. 1998. “Pension and Saving Incentives by the Bushel-Load.” Tax Notes.

Stratton, Sheryl. 2005. “Officials Gauge Government Success in War on Shelters.” Tax Notes, Vol. 106, No. 8, pp. 883-84.

Sullivan, Martin A. 2004. “U.S. Citizens Hide Hundreds of Billions in Cayman Accounts.” Tax Notes, Vol. 103, No. 8, pp. 956-64.

Tax Notes. 2001. “Deal Letter to Treasury Opposing OECD Info Exchange Initiative.” p. 1269, December 3.

Transactional Records Access Clearinghouse (TRAC). 2005. “TRAC/IRS: National Profile and Enforcement Trends over Time.”

Wasow, Bernard. 2002. “Earned Income Credit: The Compliance Challenge.” Issue Brief. Washington, D.C.: The Century Foundation.

Williams, David C. 2000. “Progress and Problems in Implementing the Internal Revenue Service Restructuring and Reform Act of 1998.” Statement for the record by the Treasury Inspector General for Tax Administration, Joint Hearing before Committees of the United States Senate and House of Representatives, May 3.

Listen to an April 12, 2005 news forum featuring the author and six other tax experts