Statement of C. Eugene Steuerle*, Senior Fellow, Urban Institute

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, 2001

Mr. Chairman and Members of the Subcommittee:

The 2001 tax act was only one in a long series of tax laws complicating an already byzantine tax system. Ever the bridesmaid, simplification seems never to get the attention it deserves, no matter which political party is in power—mainly because broader agendas are always being pursued.

It would be a mistake, I believe, to fault elected officials for pursuing those broader agendas. That is their job. Government doesn’t exist to simplify itself. It is entirely appropriate for tax policy to be the handmaiden to broader budgetary and economic policy, whether the issue is rate reduction in 2001 or deficit reduction in 1993. Moreover, simplification is merely one principle among several, sometimes conflicting, principles. For example, taxing all income on an equal basis generally makes the tax more efficient and fair. But carried to an extreme, it can add to complexity.

Still, in pursuing broader objectives and balancing principles, almost everyone would agree that simplicity has been given far too little weight in the legislative process. Many items in the tax law add significant complexity with little gain in achieving any other legislative goal. Almost no one would introduce many of the provisions now in current law, if designing a code from scratch. But once there, these complexities are hard to remove.

Complexity creates waste, not merely cost. Here one must distinguish between costs that might provide benefits and those that do not. A transfer of $1 from me to you may cost me $1, but there is an offset in the $1 that you pick up. Waste—including extra time and effort—involves resources that are simply lost to everyone. Professor Joel Slemrod of the University of Michigan and the National Tax Association has concluded that for each $100 of tax collected, we spend about $10 in time, effort, and administrative costs. Another cost, although more subtle, is taxpayers’ resentment from filling out an unreasonable number of forms. Needless tax complexity increases their cynicism toward government and frustrates a healthy relationship between a citizenry and its government.

My testimony will concentrate on four items. First, I will give two examples from recent legislation of how complexity arises. Second, I will suggest ways that I believe that the process can be reformed to give greater weight to simplification. Third, I will argue that simplification could unify and give direction to tax policy efforts in the near future. Furthermore, I will show how the inevitable need to deal with the Alternative Minimum Tax (AMT) could trigger simplification reform. And, finally, I will list some of the items that should be addressed when reform comes. The last list is not comprehensive, and many of the issues are covered elsewhere, including the recent Joint Committee report on simplification.

Example 1: Excessive Complexity in the Refundable Child Credit. During the legislative process leading to the 2001 tax cut, a number of members of Congress and private groups sought relief for those with incomes too low to pay income tax. The result was a provision that allowed the new child credit to be partially refundable, along with the retention of an alternative method of calculating a refundable child credit for taxpayers with more than two children. But combining this new credit with the refundable earned income tax credit (EITC), while retaining an alternative child credit, adds whole layers of complexity to a tax system for low- and moderate-income Americans that is already among the most complex possible.

If Congress wants to channel refundable dollars to this portion of the population, three options could achieve roughly the same distributional and revenue effects:

• Simplest of all, adjust the EITC—in particular, by slowing down the rate at which the credit phases out, which for many taxpayers effectively adds a 21 percent tax rate on additional earnings;

• Next most simple, add on the new refundable child credit but remove the older, scarcely used and exceedingly complex, form of the refundable child credit that applies to households with more than two children; and

• Not so simple, add a new refundable child credit, but give taxpayers the option of the alternative child credit if they have more than two children, and add these two child credit calculations to the EITC calculation already required.

Almost all analysts and students of tax policy, conservative or liberal, Republican or Democrat, agree that the first option would work best and the second would be the next most preferable. Congress, nonetheless, chose the third, most complex, option. Bad intentions weren’t at play, but simplicity didn’t receive its due in the bargaining process. Here were the logical steps that led to the final result:

• First, the President and leaders of Congress wanted to prevent the tax bill from being overwhelmed with additional provisions. They sought to limit amendments only to the main items put forward by the President (e.g., rate relief, the child credit, marriage penalty relief). They interpreted this process rule to mean that major amendments to the EITC, other than marriage penalty relief, were not allowed.

• Second, substantial dollars were being offered in the form of a child credit. Many thought it would be easier to explain that low-income households got some portion of the new child credit rather than that taxable households got the child credit but that others got a slower phase out of the EITC. In fact, the EITC is close to a child credit in design, although its phase-in and phase-out schedules would have to be adjusted to achieve roughly the same net result.

• Third, the spirit of the tax bill was one of "creating no losers." Every tax break was to be a reduction in rates or an additional credit or deduction patched onto the existing system—no one would face additional tax. Hence, Congress decided also to keep an old child credit for those with more than two dependents to cover the few cases where that calculation might yield a higher credit than the new refundable child credit.

Note that each of the first two goals—to circumscribe what would be considered in the bill and to grant some share of new credits to lower-income individuals—is perfectly reasonable when considered by itself. The problem is that simplicity was not given much weight in the process; no one had strong authority to come forward with easier ways to pursue the goals. The third objective—creating no losers anywhere—almost guarantees that systems will grow more complex since new options are not allowed to supersede older ones.

Example 2: The Alternative Minimum Tax. The alternative minimum tax (AMT) problem keeps growing in size, not because anyone really likes it, but rather, because no one wants to bear the cost of addressing it. If you will allow me to generalize, Republicans would be glad to get rid of the AMT or have a skeletal representation. But, historically, given a choice between lower statutory rates and fixing the AMT, they will choose lower statutory rates. Democrats, of course, would be glad to have some AMT fix also. They simply don’t want to give away any more money to those in the upper-income brackets or to pay for it by giving up other tax breaks that they favor too. Given a choice between a bill with an AMT fix and a less progressive distribution of taxes and one without an AMT fix and more progressivity, so far they have chosen the latter.

In a sense, both political parties get what they want: the Republicans get some of the statutory rate cuts they want and the Democrats maintain some of the progressivity they seek. The AMT provides the funding to do both. This is how it’s been for a long time now; the recent tax bill is only the latest act in the drama. The current tax bill gave tax cuts with one hand (mainly statutory rate reduction) and then took some of them back with the other (the AMT). But this wasn’t the first time, and everyone plays the game.

What’s going to end the game? With or without a broader agenda on which to hang the AMT reform, it’s going to require movement beyond current positions. For some, it will mean accepting a somewhat less progressive system. For others, it will require accepting higher statutory rates. Once again, progressivity and lower rates are both legitimate goals or principles. Simplification is simply going to have to be given more weight when choices among competing principles are made.

Out of the thousands that could be cited, the two examples just presented imply that while some complexity in tax law is inevitable, at its heart, excessive complexity is a failure of process. This process failure could be mitigated by the adoption of certain Executive Branch and Congressional procedures that would grant simplicity a higher priority in the policy process. More fiduciary-like responsibility needs to be assessed and formalized in specific ways. Below, I list two types of process reforms: (1) those that would involve periodic reporting on existing law; and (2) those that would apply to new legislation. Of course, in the end, what makes any process work is the good will of the parties involved to see that its spirit is maintained.

• My first suggestion is that the biennial requirement for a study of the overall state of the federal tax system (if funded by the Appropriations Committee) should be upgraded in status. It should be published every year much as the Congressional Budget Office used to publish potential expenditure cuts and tax increases to deal with the deficit problem. The list should receive continual updates, and options over time should be spelled out in greater detail and variety. By raising the status of such a list, tax simplification is liable to get the greater attention it deserves, year after year.

• The Government Performance and Results Act of 1993 requires a performance plan review that has been extended on an embryonic basis to Treasury’s tax expenditure budget. Treasury has made some very tentative steps here, though officials complain about the lack of data. While it would be foolish to think that Treasury could study each of these programs adequately each year—Congress continually mandates studies without providing the resources to back up the mandate—a cycle could be established so that each would be reviewed periodically.

At the same time, I believe there is a fundamental failure in the IRS administrative structure that leads to Treasury’s complaints about inadequate information and, indirectly, to some of IRS’ internal management problems. That defect is IRS’ failure to partially organize itself by program. Currently, IRS organizes itself by tax return category or type of taxpayer, not by the programs under its administration. It prepares few analyses of these programs and takes no responsibility for their success or failure because of its excessive focus upon itself only as a tax collector. Only indirectly do we find out about these programs, as when IRS measures error rates by line item on returns. It is not surprising, then, that IRS almost always ends up behind the 8-ball when Congress suddenly decides to examine the effectiveness of, say, the Earned Income Tax Credit, the tax exclusion for employer-provided health insurance, or the compliance costs imposed upon charities.

IRS sometimes excuses itself by saying that it is in charge of administration, whereas Treasury and the White House set policy. I have some sympathy with this argument, but it is weak. No one can properly administer a program without understanding how target efficient it is and analyzing the costs of administration for both the government and its customers. IRS does not have to make any judgment on the policy itself—just on who gets the benefits, the costs of administration, and error rates (both underclaims and overclaims). In effect, it has responsibility for better development and dissemination of the information it acquires in administering the programs.

IRS is also scared to put out reports on administrative effectiveness. In reporting on the EITC during the 1990s, for instance, it knows that both former President Bush and President Clinton favored an increase in the grants made under this program. It’s not going to make the political mistake of rushing out a report on problems associated with the 2001 tax rebate. And so on. Unless a regular reporting schedule is mandated, IRS will fear that the timing of any report release will appear to be politically motivated by one side or the other.

Here, in turn, are some methods for giving simplicity greater weight in the legislative process:

• Testimony on proposed bills should always include at least some witnesses who focus solely on the simplification and administration issues. Although affected persons should be invited, some witnesses should be more impartial and not represent stakeholders.

• When the markup of a bill occurs, one individual at the witness table should have the sole assignment of providing information on the administrative aspects of the bill. This individual might be from the IRS, the Treasury’s Office of Tax Policy, or the Joint Committee on Taxation.

• Before going to conference, the IRS should produce mock tax forms showing exactly what has been wrought from bills produced in both houses. Changes in number of users of forms and line items should also be provided, when possible.

• In conference committee, one person at the witness table should be held responsible for providing information only on the simplification aspects of the bills from both chambers of Congress.

• The Joint Committee is required to provide a "Tax Law Complexity Analysis" after reports on bills are filed. This assessment somehow needs to be given higher status in the legislative process itself. One option might be to devote one day of hearings to this type of analysis near to completion of a tax bill.

In sum, if simplification is important, then processes must be set up to insure that it is given attention and that needed resources are devoted to tracing potential and actual failures. I am hopeful that this subcommittee will devote some attention to these process efforts and not merely concentrate on items worthy of reform.

Of course, no process reform guarantees that simplification will occur. Nor, as I noted in my opening remarks, should simplification be the only factor under consideration. Nonetheless, a combination of some, if not all, of these procedures could help deter new sources of unnecessary complexity and spur the types of simplifications this subcommittee seeks.

Despite the trend toward increased complexity, significant tax simplification has a good chance of being passed sometime in the near future. The first Secretary of the Treasury, Alexander Hamilton, had it right. "The truth is," he asserted, "in human affairs there is no good, pure and unmixed; every advantage has two sides." So, let me argue, does every disadvantage. The seed that could sprout into major simplification is in one of the worst failures of the recent legislation: the extraordinary growth scheduled in the number of taxpayers subject to the Alternative Minimum Tax (AMT).

Under the AMT, a taxpayer calculates a separate tax on a different and narrower tax base than the regular income tax. He or she then pays the higher of the two. The AMT grows much faster than the regular tax because its exemption levels grow more slowly. Meanwhile, the new tax cut will make it more likely still that the AMT will be higher than regular tax. Thus, millions of taxpayers will get a far smaller tax cut than they anticipate. The AMT basically cancels out many of the benefits of the new lower rates in the regular income tax.

According to the Joint Committee on Taxation, the number of taxpayers subject to the AMT will grow from 1.4 million today to 5.3 million in 2004 to 19.6 million in 2006 to 35.5 million in 2010. Moreover, the revenues to be paid under that tax are also scheduled to grow into tens of billions of dollars. What puts more and more taxpayers under the AMT are not the "tax shelters" it was designed to expose but such simple items as dependent exemptions and state and local tax deductions, which the AMT doesn’t allow.

For the immediate future, rude surprises are inevitable for taxpayers expecting palpable relief. And help is not on the way. With the decline in revenues as the 2001 tax legislation is phased in, and the increase in spending on national defense, drug benefits for the elderly, and new and expanded educational programs, not a lot is left over to pay for simplification.

But lets play this scenario out a bit. As millions of taxpayers get added to the AMT roles every year, the level of protest is going to rise quite rapidly. Nothing arouses public ire more than perceived unfair treatment, and Americans don’t take kindly to the notion that their dependents and forced payments of taxes to state and local governments are tax shelters. Take my word for it: something will be done to fix the AMT despite all the difficulty.

At issue, though, is what type of tax bill will contain it. A large AMT fix by itself would mainly lower taxes for those with incomes above $70,000, still well above the average income. Previous Presidents, including Bill Clinton and the senior George Bush, as well as both Democratic and Republican Congresses, have shied away from any bill that would cater only to higher income groups. Even the 2001 legislation was pitched as applying to taxpayers in all income classes.

Any politically feasible AMT fix probably also has to do something for taxpayers in middle and lower income classes. But AMT reform isn’t a natural fit with, say, expanding welfare benefits or offering special deductions for the middle class. The most logical—perhaps the only logical— way to help the less well off too would be across-the-board simplification.

Congress and the President are going to have to simplify taxes one way or the other. Moreover, there’s another issue at stake: gaining control over the agenda. There are going to be a lot of tax proposals put forward in the near future. Simplification offers the President, as well as Congressional leaders, some chance of channeling this momentum, limiting the amount of special interest tax legislation, and keeping the focus on the attainment of a more efficient tax system.

Smart politicians will see personal opportunity in taking a lead and setting the agenda. If Hamilton could see a "national blessing" in a national debt, then surely some modern President, Secretary of the Treasury, or Congressional leader will recognize that rising tax complexity itself presents an opportunity to advance tax-simplification legislation before taxpayers rebel.

In addition to the alternative minimum tax noted above, among the many sources of needless complexity today are the following:

• Phase-out after phase-out of such allowances as earned income tax credits, eligibility for IRAs, eligibility for other saving incentives, eligibility for educational tax breaks, as well as the itemized deductions and personal exemptions temporarily dealt with in the 2001 legislation. Each of these phases-outs operates like an additional mini-tax system all to itself.

• Pension and saving incentives that add administrative costs and possibly even reduce net saving by providing different rules for withdrawals, penalties, Social Security tax treatment, allowable amounts of exclusion or deduction, and so on.

• A tax treatment of dependent children that needlessly makes millions of Americans file unnecessary tax returns;

• A capital gains tax law calibrated by 7 different tax rates and requiring taxpayers to fill out pages of forms even when they have only a few dollars of gains;

• A multiple choice system of taxation of mutual fund gains, as opposed to a single system whereby mutual funds could accurately report total gains from all transactions (not just gross sales) to their account holders and to IRS;

• Multiple educational tax breaks that are poorly coordinated with each other and with direct educational expenditures, thus requiring duplicate administration and complexity for students, parents, educators, and the IRS;

• Complicated rules for charitable deductions and charities, including multiple limits on giving as a percentage of income and a perverse excise tax on foundations that actually discourages charitable giving;

• Child credits and dependent exemptions that could easily be folded into one, and, even more appropriately, folded into the earned income tax credit (EITC), and

• Unnecessarily strict estimated tax rules that pick up very little extra revenue for all the complexity they introduce.

Simplification is achievable if given enough attention and effort. Process reforms can accord simplicity more weight in the legislative process, in Treasury analysis, and in IRS research. The good news in all this bad news is that the tax system has now become so complicated that almost any new legislation can make taxes simpler on balance. And the mandate for AMT relief could catalyze a much broader attack on the complexity of the tax system for all taxpayers, from poor to rich. As a practical matter, simplification offers the President and Congressional leaders a focus that could channel what could otherwise become a more chaotic tax policy process into an effort producing significant efficiency gains for the American economy.

*Senior Fellow, The Urban Institute, columnist for The Financial Times and Tax Notes Magazine, and First Vice-President, National Tax Association. Portions of this testimony were first discussed in Tax Notes and the Financial Times. Any opinions expressed herein are solely the author’s and should not be attributed to any of the organizations with which he is associated.