Today's topic before the Subcommittee is tax complexity. A comprehensive discussion of this issue would, by necessity, address the hundreds of billions of dollars worth of provisions that have been put into the tax code for reasons extraneous to fair and efficient tax collection--generally to encourage behavioral changes or simply to reward certain activities or taxpayers. As long-time advocates of a simpler, fairer tax system, we at Citizens for Tax Justice have written and testified extensively on this topic. See, for example, The Hidden Entitlements (1996), available from our office or at our web site, www.ctj.org.
Today, however, I want to limit my remarks to a discussion of the rather extraordinary complexity introduced by one, seemingly innocuous new tax provision: the $500 per child tax credit adopted in the 1997 Taxpayer Relief Act. [For a summary of the Act's provisions and its distributive consequences, click here.]
The $500 Child Credit: A Case Study in Complexity
The new child tax credit provision of the tax law had its genesis in the notion that every family in America should get a $500 allowance for each of their children. But other considerations quickly emerged.
Some people didn't think that such an allowance made much sense for well-off families. Others argued that it shouldn't go to lower-income families either (in part based on cost considerations). And because the $500 allowance was conceived from the beginning as a tax credit--apparently so it could be described as a (praiseworthy) tax cut rather than a (wicked) spending increase--many worried about excusing large families with significant incomes from paying any income tax at all.
As a result of all these conflicting goals, we now have a child tax credit that will require potentially qualifying taxpayers to fill out not only a new child-credit tax form, but also an extremely complicated new worksheet and in some cases the Alternative Minimum Tax form--a combination often even more complex than the much-bemoaned new capital gains tax calculation form also mandated by the 1997 tax act.
The new child credit form itself will probably not be hugely complex, and I will not dwell on it here. Essentially, the new form will require taxpayers to list and count their children age 16 and under, multiply that amount by $500 ($400 for tax year 1998) and, if necessary, apply the income phase-out rules at higher income levels.
The real complexity of the new child credit arises when it comes to the "worksheet" that must be completed to calculate various other limits on the credit and to determine whether it is refundable or non-refundable. In broad strokes, this worksheet is designed (1) so that the credit cannot reduce tax liability below what would be owed under the Alternative Minimum Tax for families with one or two children, and (2) for larger families, so the credit cannot exceed combined income and FICA tax liability less the earned-income tax credit.
I have devised an example of a new child credit limit worksheet, based on the statutory goals it must accomplish. The new worksheet can be combined with the old 10-line credit-limit worksheet for the dependent care credit, with the addition of 18 new lines. I've attached a representation of such a new worksheet, along with several examples of filled-income worksheets by taxpayers in different situations.
Here is a brief line-by-line explanation of what taxpayers will face in completing the new child credit (and dependent care credit) limit worksheet:
Lines 1-11 of the new worksheet are based on the design of the current dependent care credit limit worksheet. The purpose is to limit certain non-refundable tax credits, mainly the dependent care credit and the new child credit, so that they do not reduce tax liability below what would be owed under the Alternative Minimum Tax (AMT). To calculate their credit limits, taxpayers claiming the dependent care credit and/or the $500 child credit must fill out either the complex AMT form (a copy is attached) or in some cases a shortened version of the AMT computation provided on the credit limit worksheet itself.
Because of its fairly large exemption--$45,000 for couples and $33,750 for unmarrieds--the AMT doesn't actually affect many child credit claimants (even if they do have to fill out the form). But some middle-income couples with lots of children may find that the AMT can limit the amount of dependent care credit and child credit they're allowed to take--although generally only because their taxes are very, very low. The effects of this rule on the $500 child credit, however, are substantially mitigated by the refundable rule for families with more than two children. See example 7.
Lines 12-14 of the new worksheet list other non-refundable credits that come into play in calculating the limits on the $500 child credit. It should be noted that the foreign tax credit and certain business credits are treated differently than other non-refundable credits in calculating the limit on the $500 child credit for families with one or two children. In effect, some families who have these kinds of credits may be eligible for a partially or fully refundable child credit, while families with similar amounts of other types of non-refundable credits would not be eligible for a refundable child credit. See example 6. One rationale for special treatment of the foreign tax credit may be the principle that we do not intend to tax foreign income twice.(1)
Oddly, for families with 3 or more children, it is sometimes advisable not to claim an otherwise claimable foreign tax credit or business credit, because these credits do affect refundability of the $500 child credit under the alternative limit calculation (limit B). Not many taxpayers take the foreign tax credit, but for those who do, making this determination will probably require professional assistance. See example 4.
Lines 15-20 compute the $500 child credit under limit B, which can generate a refundable credit. For families with three or more children, the portion of the credit that is refundable is equal to the excess of the sum of the regular non-refundable credit limit and FICA taxes, less the earned-income tax credit, over credit limit A (see below). Although generally, these calculations are applicable only to families with three or more children, they can also be necessary for taxpayers with one or two children who also have foreign tax credits or certain business credits.(2)
Line 21 calculates the $500 child credit under limit A. Under this limit, the child credit cannot reduce taxes (after other non-refundable credits, such as the dependent care credit) below what would be owed under the Alternative Minimum Tax. The credit computed under limit A is generally a non-refundable credit, but if the limit calculated here is bigger than the amount computed under limit B, then the excess will be refundable. Generally, this only affects taxpayers with foreign tax credits. See example 6.(3)
Lines 22-26 get to the final child tax credit, parsing it between non-refundable and refundable portions. This distinction does not always matter, but often will, especially for families with large numbers of children.
So there we have it. An apparently simple idea--to give every family $500 for each child--has been transformed into one of the more complicated items in the tax code from the point of view of affected tax filers.
What can we learn from this? For one thing, it illustrates the danger, even folly, of trying to implement non-tax policies through the tax code. Our instinctive insistence that tax laws be fair makes it difficult to use the tax code for non-tax policy purposes without adding much more complexity than a direct spending program would typically entail. Similar criticisms, of course, can be leveled at many other tax-based spending programs, including many adopted in last year's tax act.
Of course, even a tax code that is limited as much as possible to the goal of collecting enough revenue to pay for government programs fairly and efficiently will have some complexity. Business taxation, in particular, will always be complicated, both because of the inherent complexity of business itself and because overly simple rules can often be gamed by aggressive taxpayers. Tax proposals that pretend otherwise, such as the flat tax, are frauds.
But we can and should make our tax laws much simpler than they are today for the vast majority of taxpayers. Indeed, the tax code after the Tax Reform Act of 1986 was far simpler for most people than it had been. It is only in recent years that a bipartisan coalition of tax complexifiers has moved us so far away from the principle of tax simplicity.
Tax simplification, does not, by the way, require compromising progressivity. On the contrary, those who promote radical tax redistribution proposals like the flat tax or a national sales tax get in the way of true simplification. It's not just that these half-baked proposals are actually far more complex than advertised. More important, middle- and low-income voters are not likely to tolerate paying a much larger share of the tax burden than they do now in the name of simplicity (as the single rate plans would necessarily require). The recent, widely criticized House vote to abolish the tax code without naming a replacement illustrates the widespread understanding by members of Congress of the politically disastrous nature of these flat-rate proposals.
Although recent history makes one wary, the right kind of comprehensive, progressive tax reform remains an excellent idea. Proposals like Representative Gephardt's plan show that it is at least theoretically possible. In the meantime, members of the tax-writing committees ought to focus more on using the tax code for its primary purpose of raising revenue fairly, rather than as a tool of economic and social policy. Or if that's a pipe dream in the short run, then what Congress wrought in enacting last year's child credit should at least encourage taxwriters to pay a lot more attention to complexity issues when they pass tax laws.
Back To Reports
1. Different technical corrections included in the pending House and Senate IRS restructuring bills rewrite this refundability rule, and appear to make it inoperative for families with fewer than three children and largely meaningless, albeit still quite complicated, for larger families.
2. See footnote 1.
3. See footnote 1.
