The Committee for Economic Development of The Conference Board (CED) uses cookies to improve our website, enhance your experience, and deliver relevant messages and offers about our products. Detailed information on the use of cookies on this site is provided in our cookie policy. For more information on how CED collects and uses personal data, please visit our privacy policy. By continuing to use this Site or by clicking "OK", you consent to the use of cookies.OK

In the Nation's Interest

The Inevitable Non-Neutrality of Tax Reform

Tax reform has recently re-emerged as a hot topic in Washington, DC. We’re waxing nostalgic about past efforts and successes, and the presidential candidates are trumpeting their own visions of reform.  And tax policy aficionados are hopeful that having a fellow enthusiast, Paul Ryan, in the role of Speaker of the House has improved the prospects for major tax reform in the next Administration and Congress.

What Economists Mean by “Neutrality”

Earlier this month, the Brookings Institution and the Urban-Brookings Tax Policy Center commemorated the 10th anniversary of the report by the (George W.) Bush advisory panel on tax reform. The Bush panel recommended two alternatives to the current federal income tax system: the “Simplified Income Tax Plan” and the “Growth and Investment Tax Plan.”  These proposals claimed to be both revenue- and “distributionally-neutral.”  Revenue neutrality refers to raising the same dollar revenue as the current income tax system. The two Bush panel proposals reduced marginal tax rates but achieved revenue neutrality by broadening the tax base. 

“Distributional neutrality” refers to leaving the distribution of the tax burden unchanged across different income categories of households.  The two Bush panel proposals achieved distributional neutrality because much of the base broadening in the proposals was through the reduction of itemized deductions and other preferences, forms of “tax expenditures” that disproportionately benefit higher-income households in higher marginal tax rate brackets—and these base-broadening increases in tax burdens offset the benefits those same households would disproportionately receive from the proposals’ (i) across-the-board reductions and flattening of the marginal tax rate structure and (ii) moving the tax base closer to a consumption base where the taxation of capital income is reduced.

The achievement of “distributional neutrality” in both of the panel’s proposals, illustrated below (from the panel’s report), was frankly impressive. 


Source: Bush Advisory Panel Report

But I continue to put quotes around “distributional neutrality” because it was only one type of specifically defined neutrality, where households were sorted into particular annual income categories and not according to any other kind of characteristics.  (It was also based on a measure of the tax burden defined as each group’s tax share of total taxes paid, rather than each group’s taxes paid relative to each group’s income—the latter which would be a better reflection of burden relative to ability to pay.)

What Politicians Mean by “Neutrality”

The Bush tax panel’s proposals were labeled “dead on arrival” on the Hill precisely because they were not viewed as “distributionally neutral” at all in the eyes of the leaders of the tax-writing committees. Former Senator John Breaux (D-LA), who had served as vice-chairman of the advisory panel, made this point at the Brookings event:  because both of the panel’s proposals called for the elimination of the state and local tax deduction, they were immediately panned by both of the Ways and Means Committee leads—Chairman Bill Thomas (R from California) and Charlie Rangel (D from New York), both representing high-tax states.  (I can vouch for this, as I was working as Rangel’s chief economist on the committee at the time.)  In other words, geographically, the proposals were explicitly very non-neutral in the distribution of the tax burden and what types of households would end up “winners” vs. “losers.” And yet, given the panel’s first-order goals of developing proposals that would be revenue neutral, reduce marginal tax rates, and yet get the tax distribution “right” across income categories, such “losers” necessarily would be created based on characteristics other than income.  The panel experts were definitely aware of this tension (“between a tax-policy rock and a tax-policy hard place” one could say), because they were looking at numbers like these—the taxpayers who would see tax increases (the “losers”) vs. those who would see tax decreases (the “winners”) under the proposals:

Source: Bush Advisory Panel Report

Note that under either of the panel’s proposals, about one quarter of taxpayers would be worse off in terms of tax burdens, despite the proposals being both revenue neutral and “distributionally neutral.”  That share, however, includes lower-income households that face no federal income tax liability under either current law or the proposals, in the denominator.  Counting only “taxpayers” who actually pay positive federal income taxes, the share of “losers” under the panel’s proposals would rise to closer to one third.  Note as well that the distribution of these “winners” vs. “losers” by income category does not show a clearly progressive outcome, and in the case of the “Growth and Investment Tax Plan” (the version which more sharply reduces effective tax rates on capital income) that the losing share is highest for the second-lowest income category—taxpayers with annual incomes of only $30,000 to $75,000.

The Tax Reform Act of 1986

Let’s look even further back in time to the last successful major (and “model”) tax reform effort, the Tax Reform Act of 1986 (TRA86).  That was a reform that was also intended to be both revenue neutral and “distributionally neutral.”  But it benefitted from the fact that its revenue neutrality came from raising corporate income taxes to finance reducing individual income taxes in equal measure, and thus being able to claim that in theory, all “individuals” (as in households, or “real people”), in aggregate and on average, could be made better off. Not just held harmless, but actually made better off.  But even in the 1986 reform, not all households were made better off under this overall-a-winner reform.  Here is the “winners” vs. “losers” table from the Joint Committee on Taxation’s analysis of the reform at the time (note the vintage look):


(Note that the numbers of tax returns in the table above should show as “thousands.”  And for perspective, the median household income at that time was about $20,000.)

Even in TRA86 almost 20 percent of taxpayers were “losers.”  Why are there winners and losers even in all these supposedly distributionally neutral (or even net-winning) reforms?  Because we aren’t starting from scratch and a level playing field.  The existing tax code that screams for reform is screaming because its tax base is full of holes and molehills, craters and mountains.  The way it treats different people differently is not just based on levels of income or other measures of “ability to pay” (the measures we keep closest tabs on in our distributional tables)—it’s based on characteristics like your family composition, the state you live in, whether or not you own a home and hold a mortgage, and whether you receive employer-provided health insurance.  And when a goal of tax reform is, on both efficiency and fairness grounds, to “level the playing field” and to even out the tax treatment of different forms or uses of income, an inevitable consequence is that the people who previously received preferential treatment from the tax system will become losers, while people who previously did not qualify for those special preferences will become winners—the latter because of the lower marginal tax rates that a revenue-neutral, base-broadening tax reform will allow.

And Back to Today

Now let’s move along to the present day and the 2016 presidential candidates’ talk about tax reform…

On the one side we have the Democrats’ repeated intentions to not raise taxes on anyone who’s not “rich”—commonly defined as households with annual incomes less than $250K.  But the only way to literally not increase tax burdens on anyone with income of less than $250K would be to either fully exempt those households (i.e., about 95 percent of households) from any base-broadening features of tax reform, or raise marginal tax rates on only those households with incomes over $250K (only those in the top 5 percent).  While such policies would be tailor-made to further the narrow distributional goal of not raising tax burdens at all on anyone who is not “rich,” they would move us away from the other economic goals of tax reform: to enhance economic efficiency, to promote economic growth, and to increase the tax system’s “horizontal equity” by treating households with similar abilities to pay (incomes) more similarly—rather than differently depending on whether their other characteristics put them on the “winning” or “losing” end of special (rather than neutral) tax treatment. And regarding the most basic goal (or essential point!) of tax policy overall, generating revenue to pay for public goods and services, if part of the Democrats’ tax reform agenda is to raise federal revenues as a share of our economy (either for deficit reduction or for expanded government services), then holding 95 percent of American households harmless will certainly handicap that effort, both economically and politically.

Meanwhile within the Republican ranks, we see candidates take the familiar, seemingly easy way to avoid creating “losers” in tax reform by proposing only massive tax cuts where everyone can be a “winner.”  This conveniently ignores the fact that there are actually costs (and “losers”) from revenue-losing tax reform, too.  Revenue-raising offsets are routinely accounted for (and distributed) by economists in evaluating revenue-neutral tax reform proposals.  But with revenue-losing tax reform proposals, which imply either larger budget deficits or offsetting spending cuts, the economic analyses of such tax proposals typically will leave out the identification or distribution of the burdens of those alternative (and unmentioned but real) offsets.  (Larger budget deficits place burdens on future generations who will eventually face higher taxes and/or lower benefits; offsetting spending cuts place burdens on anyone who currently benefits, personally or collectively, from the spending programs that are cut.)

We tax economists probably do not help the cause of tax reform by routinely taking ourselves out of the debate about the “fairness” of tax policy and sticking to our comfort zone of analyzing “revenue-neutral” and “distributionally-neutral” reform options.  Taking the issue of winners vs. losers off the table means we economists end up emphasizing the other economic effects of tax reform: economic or allocative efficiency, or how tax policy affects the allocation of resources across different sectors of the economy; and macroeconomic growth, or how tax policy affects the aggregate demand for goods and services (outputs) or the aggregate supply of labor and capital (inputs).  These are frankly the economist’s “go to” effects to measure and talk about, but, to be honest, they are pretty small in the grand scheme of all the different forces that influence our economy as a whole and the individual economic experiences of real people.  Crucially, these potential (but small) efficiency and growth effects are certainly not the factors that would make or break a tax reform proposal on the Hill.  With politicians, the distributional effects of alternative tax policies are everything to be focused on when it comes to the good and the bad of tax reform. 

So rather than glossing over it, tax economists and policy analysts should be helping politicians (and the general public) to better understand all the actual non-neutralities (winners and losers) that every single tax reform proposal contains, so that everyone can have an honest and fully-informed debate and discussion about which of these distributional tradeoffs we are willing to make to achieve which particular economic goals—distributional and otherwise. We should be better highlighting the more specific distributional effects of tax reform proposals to go beyond the rough and incomplete justice that is deemed satisfied through our mere categorization of people into income quintiles. There is no such thing as a (strictly) revenue-neutral and (strictly) distributionally-neutral tax reform proposal—other than the (strictly) do nothing option. The inevitable and unavoidable distributional effects of alternative tax reform proposals should be openly and thoroughly debated rather than either brushed under the rug under the guise of (imprecise) claims of “distributional neutrality” or over-generalized in terms of who would and would not be burdened under the proposals. The take-away is this: If a politician makes his or her tax reform proposal sound very simply appealing in terms of who would win and who would lose, voters beware.  Insist on seeing the fine print, grab a tax economist or two to help sort it out, and let’s all discuss.

Diane Lim is CED's Vice President of Economic Research. Image from the Wall Street Journal by Martin Kozlowski.