The Tax Policy Center (Joined in progress.)
NADA EISSA: This is the third session in a series on tax fairness, sponsored by the Tax Policy Center. Today's session addresses the question of whether a fair tax system is an oxymoron.
Notions of vertical and horizontal equities are, as we all know, central to debates on tax policy. Generally, it's accepted that one desirable feature of a tax system is that it is progressive, although we might disagree about the degree of progressivity. It's also generally accepted that we should treat equals equally. And, again, we can differ on the definitions of who equals are and what treating equals equally means. In the U.S., for example, we choose to tax the household. In other industrialized countries, the choice of the tax unit is the individual. In that light, I was struck recently when, on a trip to Denmark, I was told that there is a voluntary church tax in Denmark — it's a state-run church — that is paid by 85 percent of the population. And as Len Berman told me, we have a similar example in the U.S. which suggests that the take-up rate on such a voluntary tax might be very different, where on the tax return we can choose to contribute a dollar to the presidential campaign fund. And as we know, the take-up of that is much, much lower. The notions of tax fairness and how they differ across people and across countries is interesting, and hopefully we'll get to touch on that today.
So let me introduce today's featured speaker. Rudolph Penner is a senior fellow at the Urban Institute. He holds the Arjay and Frances Miller Chair in Public Policy. Previously he was a managing director at the Barents Group, a KPMG company, and was director of the Congressional Budget Office from 1983 to 1987. He's also a resident scholar at the American Enterprise Institute.
There are two discussants today: Jeff Rohaly, who is an economist here at the Urban Institute with a background in tax policy. He is very important to the Tax Policy Center. He's the director of Tax Modeling and has built up the tax simulation model here that is so important to so many of the simulations that are done by the Tax Policy Center. And Joe Thorndike will also be a discussant. He is director of the Tax History Project at Tax Analysts and a contributing editor for Tax Notes magazine.
So with that introduction I will turn the table over to Rudy Penner.
RUDOLPH PENNER: Thanks, Nada. The essay that I wrote was originally inspired by the debate over tax fairness initiated by the Bush tax cuts of 2001, but it soon went far a field. I originally called it "Groping for Tax Justice," since I think that's what the Congress actually does, but then Arnold ruined that word for me — (laughter) — so I had to change the title.
The debate over the Bush cuts tended to focus only on one criteria of tax justice: the effects on people arrayed by income class. And liberals seem to imply that in order for a tax change to be fair it had to equalize the distribution of after-tax income. There were moderate conservatives who focused more on the share of government costs paid by different income classes, and they tended to argue that so long as a share of the cost paid by the top went up, all was fair. Purer conservatives who favor a flat tax using consumption as a base provide relief for the poor — like every ideological group does in this country — but they don't care much about how income is distributed between the upper middle class, the rich, and the filthy rich.
Well, a liberal argument carried to ridiculous extremes would imply a steady march to an egalitarian society, which I don't think is something that most Americans believe in, whereas a moderate conservative argument would imply eventually that 100 percent of the cost of government was paid by the top, which is pretty ridiculous as well, so I'm sure neither side would go that far, but where would they stop?
An army of philosophers has written about the issue. I only cite the two extremes here: John Rawls on the one hand, who argues that if we met in some nether land before we were born and decided on an income distribution and if we didn't know what advantages we'd have from birth or natural talent or anything else, we would opt for an egalitarian society in a very risk-averse manner. He then adds the important proviso that we'd accept inequality to the extent that it helped the least well off. Well, that of course opens the door to considerable inequality depending on how fervently you believe in trickle-down theories. Robert Nozick starts at the other extreme. He believes that the only possible organizing principle for a society is based on the notion of entitlement; that is, a notion that everybody is entitled to what they happen to have so long as it's obtained in a morally permissible manner. Income redistribution is permitted but only if it's agreed to voluntarily.
Academic economists tend to base their theories of tax justice on the notion that it hurts less for a rich person to give up a hundred dollars than it does for a poor person to do the same, and that implies obviously that the tax burden should go up as income goes up, but there's nothing in that statement that implies that taxes should go up faster than income. If you equalize the pain inflicted by the last dollar of taxes paid, that would equalize after tax incomes if everybody had the same tastes, but economists quickly note that that sort of egalitarianism would squelch incentives, and they talk about the tradeoff between equity and efficiency. Now, while economists are fuzzy about whether a progressive tax system can be justified objectively — that is, how unequally should we treat unequals — they're crystal clear on one point, and that is that equals should be treated equally. But the practical problem is that they're not so clear on how you define equality.
Three of the most important questions that arise are, first, what should the taxpaying unit be: the individual, the family, the household? What should be the measure of their affluence: income, consumption, or wealth? And over what time period should the measure be applied: one year, several years, a whole lifetime?
Now, while philosophers and academic economists can debate such issues in the abstract, legislators have to make definite decisions that affect people's lives. And one can ask, do their votes logically imply a consistent theory of tax justice. Very obviously the answer is a resounding no. And that's not surprising given that legislators are good at producing compromises not purist solutions. Nevertheless, I think it's remarkable how they've taken every possible criteria of tax justice, mixed them all up in this grand hodge-podge that we call a tax system. Furthermore, the weight given to different criteria of justice by the Congress tends to change from time to time, which I suppose is not surprising given that the ideological makeup of the Congress changes, but even within a single Congress you have shifting coalitions with shifting ideologies that are necessary to get a majority vote.
So taking the tax unit as an example, our most important federal tax, the personal income tax, has different schedules for singles, married couples and their dependents, and households. In contrast, the second most important tax, the payroll tax, very clearly depends on the wages of the individual. Other taxes, like property taxes and the old telephone excise taxes, might be said to be taxes on households depending on your theory of tax incidence.
Within the income tax, legislators have struggled over time as to how to tax married people versus singles, and large versus small families. And the pendulum has swung back and forth through history, but the Bush tax cuts very clearly move the pendulum toward historical extremes in favoring married couples and large versus small families. So I expect singles to start complaining loudly at any minute.
With regard to the tax base, we of course have consumption taxes and taxes on capital income and taxes on different kinds of wealth, but what we call an income tax is in fact far from a pure income tax. There seems to be a political consensus that a pure income tax would tax capital income too heavily, so we have special concessions for retirement saving, medical saving, educational saving, capital gains, dividends, and depreciation deductions.
The concessions on capital income of course imply that in the longer run, consumption must be taxed more heavily, and I'm not sure whether this is the result of a Hobbesian philosophy that argues that it's fairer to tax people on what they extract from an economy, as measured by their consumption, as opposed to what they contribute, as measured by their income, or whether it's purely a pragmatic decision not to dampen incentives for economic growth. A cynic might argue that it could simply be to provide tax relief for the rich in a camouflaged manner. I doubt the last, for reasons I'll come to later.
But whatever the true reason, it is clear that the pendulum swings back and forth on this issue as well. A high-water mark for income type taxation was reached with the Tax Reform Act of 1986. Since then we've seen numerous concessions on capital income. I think, without careful study, that I see a correlation between the top marginal rate and the degree to which you provide concessions for capital. The Tax Reform Act left us with a top marginal rate of 28 percent. It's my impression that people didn't find that bothersome applied to capital income, but as the marginal rate went up again we saw concessions on capital income go up again.
On the other issue I mentioned — it hasn't received much attention recently — is the question of over what time period you look at the measure of a person's resources? A progressive tax system, based on one year's income, tends to hurt people with volatile incomes relative to others. It tends to hurt professionals who spend a long time in education and have a relatively short, well-compensated career. We used to mute these effects by allowing people to average over several years, at least for those whose income went up. We didn't do it for those whose went down. But after being cut back earlier, that privilege disappeared altogether in 1986, and there hasn't been much discussion of it since.
My full paper describes, in some length, efforts to redistribute on the spending side of the budget where we tend to focus on subsidizing different kinds of in-kind consumption. I think that raises all sorts of interesting issues, but alas I can't go into them here.
Legislators of course play to the public, and that has an important impact on outcome. We often judge public opinion through polls. I think there's a danger there. There's usually enough ambiguity in poll results that you can read into them what you like. But in a recent session at the Brookings Institution, separate papers by Bartels and Slemrod looked at public opinion and reached exactly the same conclusions I do, with somewhat different polls. And I think the main features of public opinion are, first, the public is abysmally informed about the details of the tax system. Second, they think it's unfair. They don't think the rich pay their fair of taxes. And they tend to believe that the income tax is actually regressive. They do not, however, understand that on average the top income groups pay a very high average rate of tax.
My own conclusion, based on the polls, is that the tax burden on the rich is probably higher on average than most people think that anyone should pay, but I emphasize "on average" because I think people's confusion is understandable. One of the many reasons that distribution tables are misleading is that they disguise an incredible variation in tax burdens within every income group. In the most recent session here with David Kay Johnson he pointed to extremely low average tax rates among some of the richest people in the country. I'm sure he could have found some high ones as well, but it's the low ones that make the strongest impression, and huge variations can be found within every income class.
However, a belief that the current system is unfair does not translate into a belief that a fair one would be highly redistributed. The National Opinion Research Center has, for a number of years, asked what people think of the statement, quote, "It is the responsibility of government to reduce the differences in income between people with high incomes and those with low incomes." In the year 2000, only 33 percent "strongly" or "somewhat" agreed with that statement, 25 percent neither agreed nor disagreed, and 40 percent disagreed or disagreed "strongly."
In the Journal of Economic Literature that just arrived several days ago, there's an article by James Konow on experimental evidence regarding people's views of justice, and I really wish I would have had it when I wrote this essay because it's highly relevant and very interesting. I think in general terms, the evidence supports the polling evidence cited in my paper. People do not fervently support redistribution. In particular, he has a bunch of tests of Rawlsian notions and people tend to reject them. They also reject Nozick.
There is, however, one message that comes out loud and clear that is missing from my essay. Just as people separate out the deserving and undeserving poor, they seem to separate out the deserving and the undeserving rich. In two of Konow's principles — one he calls the "dessert" theory, as in just desserts and the other the "equity" theory. It's clear that people are not hostile at all to great riches if it's perceived that they were gained from hard work. They're not so sympathetic to riches gained from luck or accident of birth. If riches are gained through making particular choices like getting a good education, attitudes are much more complex, and I can't go into that here.
I find attitudes to riches gained from natural talent to be most interesting. Nozick asked his readers to perform the following mental experiment. Suppose you can dictatorially impose whatever distribution of income on the population you think is just. Then suppose — to date the book, he says Will Chamberlain comes along, but let's say Shaq O'Neal comes along, and it costs an extra $25 a ticket to see him play basketball. He's allowed to keep the proceeds. One million buy tickets so Shaq walks away with $25 million, presumably a more unequal distribution than you started with. But Nozick argues that it must be a just distribution since the first distribution was fair by definition and the new one is completely the result of voluntary transactions — no coercion is involved. Konow's experimental subjects reject Nozick's argument 59 to 41. I must confess I'd probably be in the 41, but I find the result very interesting.
In any case, we can be sure of one thing that the public believes about redistribution: they may favor it weakly in the United States but concerns stop at the border. Although the world's poor make our poor look very rich by comparison, there's very little public support for foreign aid.
My paper discusses the results of all of this by looking at distribution tables of various types; that is, after recounting the many reasons that distribution tables are misleading. But Jeff Rohaly will argue later they aren't and maybe we can discuss the issues further then. But for what they are worth, the tables show a considerable amount of redistribution, which is interesting, given the public's apparent tepid support for such policies.
My overall conclusion is what I call the tax establishment — editorial writers, think tankers, legislators, congressional staff — have disproportionate power in the formation of tax policy because of the misperceptions and general disinterest of the public. And Joe Thorndike will elaborate on that them later. But it's my feeling that the establishment is probably more interested in redistribution than the voters. So will the public regard Bush's tax cuts to be fair? I don't have a clue to tell you the truth, but in judging the issue it must remembered that the tax cuts had three dimensions: they favored married couples, they favored large families, they favored capital income over wages and consumption income.
The last dimension was structured in way that favored many of those, but not all, with very large incomes. Now, I'm not sure how the public feels about that issue, but I'd suggest that Democratic candidates ignore the first two themes at their own peril.
Thank you.
MS. EISSA: Thank you very much. We'll turn now to the discussants, and we'll start with Joe Thorndike. And each discussant will speak for 15 minutes and then we'll open it up to the floor for questions.
JOSEPH J. THORNDIKE: I think it's my job to try to put this discussion in a somewhat larger historical context and I'll begin, as so many people do, with Oliver Wendell Holmes, whose overworked quotes get to start a lot of the time.
"Taxes," he once said, "are what we pay for civilized society." And that's an apt description, capturing the dual role of any tax system. Fundamentally, taxes are the price of government, but they do more than simply raise money; they also serve as a means of social control, encouraging the development of a civilized society by regulating behavior and promoting social justice. Both functions, revenue and regulation, have played a key role in the history of American taxation. Americans have accepted a growing tax burden as a price of a growing government. In addition, however, they've used taxes for non-revenue purposes. They've employed tariff duties to protect American industry, and sin taxes to protect American morals. Perhaps most important, they've embraced income and estate taxes, not just for the revenue they raise but for the apparent fairness that they bring to the tax system.
Understanding the history of American taxation, then, means understanding the willingness of American voters to tax themselves for a variety of purposes, including social justice. As the noted tax historian Elliot Brownlee has demonstrated, the federal revenue system has evolved through a series of major watersheds, most of them sparked by wars or other national crises. But while crisis prompts change, it does not shape the details. Tax reform emerges from a complex political process, one that balances revenue needs against political constraints. In every period of U.S. history, issues of justice, fairness, and equity have figured prominently in tax debates. What seems striking about the history of American taxation is the central role that political leaders have played in the social definition of tax justice. Rarely have political movements arisen unbidden from the grassroots to work major change on the nation's tax system, all in the name of tax justice. More often, political leaders have harnessed inchoate, sometimes confused popular notions of fairness to advance specific programs of fiscal reform.
The U.S. tax regime, the first one, emerged from the crisis of independence. It was marked by an overwhelming reliance on tariff duties — which we don't really consider taxes anymore but they certainly did then — and it was shaped by a powerful ideology of what we historians call "small-r republicanism." This was an ideology that included a commitment to communal responsibility, a dedication to public virtue, and a fear of public corruption.
In 1787 the new Constitution granted the federal government broad authority to levy taxes, but the power came with important restrictions. The Constitution required that all taxes be uniform throughout the states, and even more important, that direct taxes be apportioned among the states according to population. These limitations grew out of a deep and abiding strain of suspicion among the nation's founders. Inheritors of a classical Republican tradition that stressed communal responsibilities and civic virtue, they harbored fears of centralized power. The new federal government posed a significant threat even as it promised to solve many of the young nation's problems.
In particular, national leaders worried that unrestricted taxing power, and direct taxes in particular, might give rise to corruption and factionalism. Americans, it's important to point out, were not opposed to direct taxes at this time. Property taxes, in fact, were a mainstay of state government. They enjoyed a reputation for fairness, in large part because property was considered a good measure of taxpaying ability. The restriction on federal taxation then had less to do with any fear of such levies than it did with a desire to protect them from encroachment by the federal government.
In practice, the young United States relied on the fiscally adequate, politically popular system of low tariffs for most of its money. Constraints on the federal taxing power ensured that the government would avoid most direct taxes except in dire fiscal straits. Tariffs meanwhile comported with popular notions of tax justice. Spread broadly throughout the country they didn't seem to encourage factionalism by burdening any particular class or geographic area. And to the extent that they were levied on luxury goods — and many of them were — they served a popular commitment to vague ability-to-pay principles.
The second American tax regime emerged from the Civil War, marked by a dependence, again, on tariffs, but this time on much steeper tariffs, and on numerous excise levies. It lasted for more than 50 years. But the Civil War tax regime is more notable for a much less durable fiscal innovation: the nation's first income tax. Conceived as a tool for promoting social justice, the tax found support from the Union's Republican leadership. Legislators understood that excises, especially when coupled with steep tariffs, made for a regressive tax system. As a result they decided to supplement consumption taxes with a new income tax.
And here I'll quote from Justin Morill, one of the prominent Republican leaders of the time. "Ought not men with large incomes to pay more in proportion to what they have than those with limited means who live by the work of their hands or that of their families?" And Thaddeus Stevens, another prominent leader in the House, justified the tax as a necessary balance: "It would be manifestly unjust to allow large money operators and wealthy merchants, whose incomes might reach hundreds of thousands of dollars, to escape from their due proportion of the burden." The income tax, then, was designed to be a progressive counterweight to more regressive consumption taxes. It was not the product, it's important to note, of popular demand, but it did represent an leadership effort to forestall popular discontent with regressive taxation.
After the war, lawmakers cut taxes. Excises were the first to go. Policymakers held on to a few on alcohol and tobacco and luxury goods, partly because these brought in a lot of money and partly because, as sin taxes, they enjoyed a certain amount of legitimacy. The income tax was next in line. Drawing on a persistent strain of Republicanism, the same one we saw in the Early Republic, critics went after the Bureau of Internal Revenue for its army of officials and vilified the income tax as an inquisitorial levy. Even more important, though, the repeal of excise taxes had eliminated the counterweight argument for the income tax, so the need for a progressive addition to the tax system seemed to be significantly less serious.
In 1872, Congress allowed the income tax to expire, but the tax was still an important experiment. It emerged from an effort to forestall popular dissatisfaction, and it demonstrated that ability-to-pay arguments resonated with lawmakers, if only in the midst of a national crisis.
The third American tax regime emerged from World War I. Officials again faced a wartime crunch and they again turned to the income tax. Fairness remained the central issue, but this time Republicans soon came to appreciate that the fairness qualities that the income tax brought to the revenue system had something to be said for them. Democrats had tried to implement an income tax in the late 19th century without success. They met with bitter opposition. Senator John Sherman, who had supported an income tax during the Civil War as a fairness tool, complained, "In a republic like ours where all men are equal, this attempt to array the rich against the poor or the poor against the rich is socialism, communism, devlism." But by 1913, Democrats and Republicans had won approval for the 16th Amendment, ensuring that income taxes would have a place in the federal system, and Democrats soon crafted a new one with a narrow base and low rates. They believed that it would make the system more fair. And this was the principal reason for its adoption, not revenue. They did not believe that it would ever become a major source of federal finance.
World War I prompted Congress, however, to expand the levy. Lawmakers doubled the rates and established an estate and excess profits tax. Altogether the tax system got substantially more progressive and by 1918 the top marginal rate had reached 77 percent. Republican victories in 1920 seemed to hold out the prospect for tax reduction. Treasury Secretary Andrew Mellon insisted that rates were too high, distorting the economy and hampering tax collection. Over the course of his long tenure — and I should point out here that he served for more than a decade, prompting one person to observe that "three presidents served under Mellon." (Laughter.)
But the secretary orchestrated a series of major cuts. These cuts, however, did not destroy the World War I tax regime. In fact, they helped ensure its longevity by softening its more radical edges. Elimination of the excess profits tax, which was deeply loathed by business leaders, combined with reductions in the high marginal rates, deprived income tax foes of their rhetorical thunder. Woodrow Wilson may have made the world safe for democracy, but Andrew Mellon definitely made it safe for income taxation.
The Great Depression brought the fourth tax regime. Once again, consumption leapt to the forefront and Franklin Roosevelt was the unlikely steward of this transformation. He was compelled by revenue needs to accept regressive excise taxes, but to compensate he championed new taxes on wealth and income. Perhaps most important, he also developed a highly moralistic rhetoric of tax fairness.
A drop in revenue brought on by the Depression forced lawmakers to enact a variety of consumption taxes. Roosevelt did almost nothing to change this when he arrived in office, or at least nothing to eliminate these, but he did try to restore some of the progressivity that had been taken from the tax system by Mellon in the '20s. The president admonished Congress, when he was proposing the Wealth Tax Act of 1935, "Our revenue laws have operated in many ways to the unfair advantage of the few, and they have done little to prevent an unjust concentration of wealth and economic power."
Roosevelt was not actually much inclined to redistribute wealth, despite his rhetoric, and his tax changes have been criticized for being more symbolic than substantive, but I think that they did reveal a deep and abiding belief, on Roosevelt's part, in the idea — however vague, of ability to pay. And he also tended to put tax fairness in intensely personal terms. He conducted a long-running campaign against tax avoidance, convinced that the question of individual morality figured prominently in the notion of tax fairness. He denounced tax avoidance for being technically legal but morally bankrupt. He sold this idea, moreover, to the American public quite effectively. Polls showed a deep conviction that rich Americans should shoulder more of the tax burden, and many Americans even supported the public release of tax returns, but only for rich people. The numbers dropped substantially when people were asked about releasing returns for average Americans, although at this point, very few average Americans were actually paying the income tax. It was still a very narrow levy affecting only the rich and perhaps the very upper reaches of the middle class.
In any case, the tax system did not really change — the Depression tax system did not change substantially until World War II brought another crisis and the nation's fifth tax regime, its last, and the one we still live with today. Again, policymakers turned to the income tax. They realized, however, that the narrow tax of World War I was inadequate. Of necessity they chose to broaden it, transforming it, in this much overused phrase, from a class tax to a mass tax.
Most economists didn't like the Depression-era tax system, recognizing that it depended too heavily on excise revenue. Better, they argued, to develop a broad-based income tax with progressive rates. Roosevelt, however, was not eager to endorse this reform since he, like most Democrats, believed that the income tax should remain a rich man's burden. World War II forced his hand, however. Officials were confronted with soaring revenue needs, and they considered several options, including a sales tax. The sales tax held out the promise not only of new money but also the chance that it might reduce consumer spending and help thwart inflation.
Roosevelt, however, took a strong stand against the sales tax. He and his Treasury officials insisted that it violated the concept of ability to pay. "It falls more heavily on the poor," one Treasury official wrote. "It is, in fact, a spare-the-rich tax." After a vigorous fight, lawmakers rejected it, choosing instead to expand the income tax. The class-charged rhetoric that Roosevelt brought to this debate proved pivotal, not only derailing the sales tax but also ensuring that the income tax would be steeply progressive. By 1944, the top rate stood at 97 percent, and it remained in the 90s for more than a decade after that.
World War II, then, transformed the tax. Lower exemptions and higher rates made it a fiscal workhorse. Providing just 16 percent of total revenue in 1939 it delivered 45 percent in 1945. And by contrast, excise taxes delivered just 11 percent after the war, whereas they had been providing 30 percent before the war. Mass income taxes had become the mainstay of federal finance, where they remain today.
This quick overview of American tax history can be a little depressing in the sense that major tax changes occur so infrequently it's easy to lose faith in the quest for tax reform, at least tax reform that's engineered deliberately. Absent a major societal crisis, it seems unlikely that major changes will occur, so why even debate issues of tax reform generally or tax fairness in particular? I think that the debate is valuable because change does happen, and when watersheds occur, they reward both preparation and innovation. The major tax reforms of World War II, for instance, were conceived long before the war began. Experts championed the income tax for more than a decade before the political stars aligned, but when they did, advocates for broad-based income taxation were ready with well-formed ideas and proposals.
The income tax so far has turned out to be a political winner. Americans have complained about the modern tax system, but it's enjoyed remarkably widespread acceptance, perhaps because of inertia, but nonetheless — and for politicians it's proven to be a bonanza, providing ample opportunity for granting special favors to valued constituents, much like the tariff did in the 19th century. But unhappiness with the income tax seems undeniable, and the real question today is whether such unhappiness by itself, whether questions of tax justice and fairness I think, cast more broadly by themselves, can lead to major change. From my perspective, it does seem unlikely. In any case, however, it behooves supporters of the income tax and progressive taxation more generally, to prepare for the debate that must eventually occur. Candid and cogent ideas on tax justice, so often central to the development of new tax regimes, are the obvious place to start.
MS. EISSA: Thank you. Let's turn now to Jeff Rohaly.
JEFFREY ROHALY: Thanks. Today we're focusing on the fairness of the tax system, so what I'm going to try to talk about is how the Tax Policy Center attempts to measure fairness through the use of our distributional analysis and the tables that we release on our website. Rudy already talked about the fact that distributional analysis measure is one of the aspects by which you should judge tax policy. Equity — in particular, vertical equity, the notion that those with a higher level of economic well-being, those with a greater ability to pay should pay proportionately more of their income in taxes. So distributional analysis can be used to get a sense of whether a tax proposal is regressive, progressive, or proportional. And obviously I believe distributional analysis is important, but it's also important to keep in mind that it's only one piece of the puzzle. Proposals should also be judged on their efficiency and their simplicity, and we've actually used our tax model to analyze both of those aspects, but today I'm going to concentrate on equity.
First, just a bit of the background on our model that we used to produce our estimates. Virtually all of our distribution tables are produced by our TPC micro-simulation model, which is based on a large database produced by the IRS. It's a nationally representative sample of about 130,000 tax returns, and if you'd like a detailed description of our model and our methodology, there's a handout in the packet that will talk about that.
One other thing I want to mention is that in addition to that information from tax returns, through the current population survey we also have information on those typically very low-income people who don't file federal individual income tax returns. So that we're able to do distributional analysis on the entire population when we produce our tables.
Right now in our distribution tables, we typically look at the impact of proposals on the individual income tax, and we also have the ability to distribute payroll taxes, and we've done so at certain points. Those two taxes account for well over 80 percent of all federal income taxes, but we are at the moment trying to work on, or almost completed, a model of estate and gift taxes. We also plan to include corporate income taxes in our distribution tables in the coming years so that we can give a more complete picture of the distribution of all federal taxes.
Before doing what I essentially want to do, which is dissect a typical Tax Policy table, let me just start with a bit of a background on distributional analysis. At the most basic level, why do we want to do distributional analysis? What's the purpose of distributional analysis? It's an attempt to provide information on how a tax proposal changes the economic well being of different groups of people which obviously leads to two questions, the first being how should you measure the change in economic well being from a tax proposal; and secondly, how should you group individuals into meaningful categories to figure out who benefits or who does not benefit from a tax policy change.
I'm going to refer to a handout that you got in your packet of a standard Tax Policy Center distribution table. It's a table labeled T03-0123, which means it was the 123rd distribution table I got to produce last year. This particular distribution table shows the combined effect of both the 2001 and 2003 tax acts, EGTRRA and JGTRRA, for calendar year 2003. Then, just like Treasury, we distribute only the major provisions of both of these bills. The details, if you want to read them, are in the footnotes.
And the first thing I want to talk about is this idea of how you group individuals into categories. Our classifier, you can see from the first column, is adjusted gross income, as reported on federal tax returns. In this particular table we're grouping individuals into dollar income categories: those with incomes less than $10,000; $10,000 to $20,000, and so on all the way up to those in the more than $1 million of income category. We also produce tables where we break the population up into equal quintiles, equal shifts of the income distribution, but I want to concentrate on this particular table.
So in terms of our classifiers, the advantage of using adjusted gross income is that it's an easy to calculate, easily recognizable measure of income. It comes straight off a federal tax form, and so it's easy for individuals to basically find themselves in our table and see what happens to them. The disadvantage of AGI is that it's a fairly narrow measure of income. Since what we're really trying to do here is we're trying to group people by their ability to pay, their economic well-being. The problem is that AGI obviously leaves out a lot of forms of income. It leaves out nontaxable government transfer payments. It leaves out nontaxable interest income. It doesn't count accrued but unrealized capital gains, and so forth.
At a more fundamental level — and Rudy talked about this — but this is a problem, not just with AGI, our income measure, it's also a problem with the measures used by CBO, JCT, and Treasury. It's a one-year snapshot of income. So obviously people's incomes do vary over their lifetimes, and theoretically it might be nice to use a measure of permanent of lifetime income. We, at the present time, just don't have access to a panel of tax returns across years that would allow us to do that. But given the limitations of AGI that we recognize, we are currently working on a couple of more broad measures of income that we can use in our tables: one, a measure that's more like Treasury's definition of cash income that adds back into AGI some level of resources that should matter, like non-taxable government transfers and a few other items. Then, we're also working on a much broader measure of economic income that's going to include forms of unrealized but accrued capital gains for example.
Our preliminary analysis, and from what I understand to be the case at Treasury when they, for a time, produced tables by both cash income and this broad measure of economic income, is that in most cases — not in all, but in most cases it doesn't make a substantial difference. A progressive tax cut is still going to look like a progressive tax cut regardless of your income classifier, but we do plan on, in the next year, releasing tables, at least in situations where it does seem to matter, by a broader definition of income to get a better measure of people's well being.
So that's the classifier. The other issue is, okay, how do you measure the change in well being from a tax proposal? What are our preferred measures of distributional impact? Obviously there's no perfect measure. Ideally what you'd like in this table is a column that tells you the change in economic well being of all of these different groups, but that's obviously too abstract a concept that we can't measure directly. We proceed on the notion that welfare is closely linked to after-tax income. After-tax income is the resources that people have available to them in a given year to consume or to save, so our preferred measure is the middle column in the table, the percentage change in after-tax income, the idea being that if a tax cut gives everyone the same percentage increase in after-tax income, the distribution of economic resources is the same before and after the tax proposal.
So by that standard, if you look at the table, EGTTRA and JGTRRA, the 2001 and '03 cuts are regressive. The percentage change in after-tax income is substantially higher for those at the top of the income scale than for those in the middle or the bottom. For example, those in the more than $1 million income category get an average increase in their after-tax income of 5.4 percent. Those, for example, in the $30,000 to $40,000 get 2.5 percent. One thing to note is that this doesn't include the estate tax provisions in these bills. It doesn't include the retirement pension portions of the bills, which presumable also disproportionately benefit those with higher incomes.
So percentage change in after-tax income is the measure I tend to concentrate on, but we do provide a couple of other measures. We also provide the percent of the total income tax change that goes to each income class, and we also present the average tax change in dollar terms. The latter, the average tax cut in dollars, tends to be the one that garners us the most press attention, but it can be misleading, as I'll explain in a second. If you look at the average tax change from these two bills, for those in the million-and-over category it's just under $113,000, substantially more than individuals in the lower-income categories. But of course the problem is that — not the problem but the fact is that those making more than a million dollars also pay a much higher amount of income tax in the first place, so it might not be surprising that they are the ones who get the largest tax cut in dollar terms. So a natural question then is why not show - instead of the average tax change in absolute dollar terms, why not the average change in tax as a percentage of tax liability, which is what Treasury showed in the table that it released for last year's tax bill.
The next page on the handout goes over a very simple example of why we have chosen not to do that. Suppose there's just two people in the economy, Person A, the low-income person; Person B, the high-income person. Person A has income of $20,000 and pays a dollar in taxes. The proposal comes along and cuts their taxes by $1, wiping out their tax liability, so their percentage cut in tax liability is 100 percent. Person B, the high-income person making $2 million pays $500,000 in tax, gets a $100,000 tax cut, which represents a percentage cut in their tax liability of 20 percent. So using the percentage change in tax liability makes it look as though this particular proposal is tilted towards the low-income individual. They get a 100 percent tax cut. The high-income individual gets a 20 percent tax cut. But as the handout shows, in terms of what you should care about or what should matter to welfare, after-tax resources or after-tax income, in percentage terms, the increase for the high-income person is 5 percent, for the low-income person is a fraction of a percent, so that the high-income person gains economic resources that are about 1,000 times more than the low-income person in percentage terms.
Another common measure that we don't show in this particular table is the share of the overall tax burden paid by different income classes, which is obviously a useful concept on its own but can also be misleading if the overall tax burden is changing and you're using that to compare two situations where the overall tax burden is different. For example — I'll come back to what I was just talking about, but for example, in the debate over last year's tax bill, it was put forth that JGTRRA would result in high-income individuals paying a greater share of the individual income tax burden than they did before the bill, which is true and which is what Treasury showed in its distribution tables that are released. But that ignores two things. The first is that you need to look at the distribution not only of individual income taxes, but the entire tax system, and even at a more fundamental level it's that even if high-income individuals end up paying a greater share of the overall tax burden, it doesn't mean that they benefited proportionately less than everyone else from this tax cut.
Look at the example that I was just talking about. Again, after this proposal, the second person, Person B, the high-income individual pays a higher share of the tax burden. They pay 100 percent of the tax burden because the other person had their tax liability wiped out, and that's a higher percentage of the tax burden than they paid before but as I just explained, in terms of after-tax income, in terms of economic resources, that high-income person actually, from this proposal, benefited 1,000 times more than the low-income person.
So that's our standard table where we show a number of measures of distributional impact with what I think the most informative being the percentage change in after-tax income, but to address some of the concerns that Rudy brought up, for example the variability in average amounts within income classes. We also provide other tables that supplement our standard table. Number one, we provide separate distributions by filing status, which will get at the issue of, for example last year's tax bill benefiting married couples and those with children proportionately more than the rest of the population. We also show tables that give the impact on hypothetical families: what's the tax savings for a family of four with $50,000 in income with two kids? What's the tax savings for a single person with $30,000, $40,000 income?
We also provide tables that show the number and percent of returns that receive various sizes of tax cuts. In other words, we put out tables that show what percent of the population gets no tax cut from this proposal, what percent of the population gets $1 to $100, what percent get $101 to $500 and so one. You have examples in the handout of some of those tables that we produce.
So let me just finish by talking briefly about how we plan on enhancing the model that we have now and improving the estimates that we put out. The first thing, like I mentioned at the beginning, is we intend on including estate taxes and corporate income taxes in our distribution tables, along with individual income and payroll taxes, to give a better sense of the effect on the entire distribution of all federal taxes.
Secondly, we're going to try to use a broader measure of income to better classify households by their level of economic well being, and in some tables, at least in situations where we feel that it matter, we're going to use a broad economic income measure that's adjusted for family size, which is something I didn't really talk about but it gets at the issue that a family of four with an income of $30,000 doesn't really have the same level of economic well being as a single person with $30,000 of income. And yet, in a standard distribution table, those two households would be lumped together in the same income category. So we're going to experiment with adjustments for family size, much as CBO does in the distribution tables that it puts out.
And finally, we're going to continue to, in addition to our standard table, put out other tables that show the effect on hypothetical families, the effects on different filing statuses, the effects by size of tax cut, and in so doing, hopefully we will provide a complete description of the fairness of tax proposals. Thank you.
MS. EISSA: Thank you all very much. We will open it up for questions - if you could speak into the microphone and identify yourself.
Q: Hi, I'm Jason Fichtner with the Joint Economic Committee. I have a question for Jeff. Going back to your tables, Jeff, I really appreciate all you've done to put out more information on these tax policies in the models, and other organizations have in the past on their tables. But the question I have is on your focus on the percent change in after-tax income, and you clearly pointed out why you are concerned with the distribution of well being, but as Joe pointed out in his history lesson, we're concerned with the distribution of tax burden, not distribution of tax well being. And I think it's important to consider everything, and you did a good job of highlighting how, in your second table, the percent change in taxes; how focusing on just the percent increase in after-tax income can be misleading. It makes it look like the tax cut — like the Bush plan was more beneficial to the poor than it was to the rich.
However, I would argue that your table was also misleading because you're failing to highlight the other options. And so my question to you is, why doesn't the Tax Policy Center be academically open and put all three measures — or even add a fourth which shows how many people don't pay taxes currently. And a lot of people actually get negative taxes, get a refund or social transfer from the government. How do you give a tax cut to those who don't pay taxes? So my question is why not, in a next Tax Policy Center table you put out, include percent change in after-tax income, but then also include percent change in tax liability, the share of tax burden paid before and after, put it all in one table. You can put a narrative why you think one is more important than the other, but put it all out for the public to see so they can make their own decision based on all measures that various economists have said are important.
Thank you.
MR. ROHALY: We don't actually put it all in one table, but we have — and I think I put in your packet tables that show the pre-EGTRRA, pre-2001 tax cut and post-2001 tax cut distribution of income and payroll taxes so that we can get at this notion that some people prefer to use that who pays the overall share, what income class pays a greater share of the overall tax burden before and after a tax proposal. And in fact, we do show that for example, high-income people do pay a higher share of overall income taxes after the Bush tax cuts than before, but they also in fact pay a lower share of income and payroll taxes.
Q: My point is I think that's great to put it all out there, but what usually gets out to the media is your one table, and that's what gets published in the Washington Post, the New York Times, everything else, and I think that's very misleading. You're kind of smiling about it, but what happens is one table gets out there and gets used in all the debates because the rest of the stuff is sort of hidden in a packet or it's hidden on your website. Why not — if you've got one table that shows all these different tax burdens, take one, whether it's income tax, payroll, add it all together, put it all on one table with that information so, again, the public can see it all on one table, not to go searching for it.
MS. EISSA: I think that the point is valid that there are many ways to measure the distributional effects of a tax change, and I think the point —
Q: If you could show it all at once —
MS. EISSA: If we could — point well taken. I think we'd like to have the discussion be broader than that, but point well taken.
Q: Jim Klumpner from the Senate Budget Committee. I want to stay on that point as a matter of fact. (Laughter.) Do you know of any economic theory that says people's utility is a function of their share of the tax burden, or do most economists think that utility depends on a person's income and consumption?
MR. PENNER: Well, I'm not sure that's really relevant. Some may judge the fairness of a tax change by the distribution of after-tax income and the "utils" that people receive. But others ask, who's paying for government? That's a different question and it has nothing to do with the distribution of utils, but I think it is a fair question.
MS. EISSA: There's a question over here.
Q: Edward Cowan — I'm a retired New York Times correspondent who covered taxes for a while, and I'm an independent editor and writer now.
Rudy, thank you for a very interesting paper. It reminds me a little bit of one of the interesting passages of parenthood. When children get to be about eight of nine years old they discover what they hope is going to be a very effective protest, that something is not fair. And so the challenge for the parent is to explain that fair, like tax justice, is a somewhat elusive concept - it can mean different things — and that there are other considerations, and of course that applies in taxes.
I want to ask you if, in preparing this paper or in your other work, you came across anything that, even in your mind, amounts to a time series of public feelings about tax justice or tax fairness. Is there any point at which, as far as you know, the public was more satisfied with the fairness of the federal tax system than at other times, or than now?
MR. PENNER: Well, the one poll that I cited that asked people's attitudes to redistribution has been run since the early '80s, and you see virtually no change in the results over time. I suspect people's cynicism about the tax system relates to their intense distrust of government, and that has changed over time.
There is a poll that asks the following question. "If government tackles a problem, are they likely to do more harm than good?" And the people who were positive toward government were in a majority in the early '60s, and ever since, that has deteriorated and deteriorated badly. So the number of people who think government is more likely to do harm has grown enormously, and I personally attribute that to Vietnam, Watergate, and the failures of the Great Society. And we've never recovered from that.
MR. THORNDIKE: I might be able to add something to that. There are poll questions going back at least to the '30s that would try to answer that question. They're not necessarily consistent poll questions, and polling in the '30s — the early years of public opinion polling — were not always great, not always accurate. But I do think that there was a substantially — well, there are similarities and differences to be seen. On the one hand, I think attitudes toward the tax system generally were more positive. People regarded it as more fair. On the other hand there was a great misunderstanding — many great misunderstandings about what the tax system actually was, much like there are today. People had a vastly different idea of what tax rates were. They didn't have a good sense of what people were actually paying, and in fact, when asked to choose what tax structure that they would like, they routinely chose rates that were much lower than anyone was actually paying.
So, that being said — I think that the constant here is confusion and lack of information. What has changed, again, is about this trust in government issue that has brought the tax system into increasing disrepute along with the rest of government.
Q: Hi. Julie Kosterlitz with National Journal. Joe Thorndike had said that you don't usually get major tax reform or tax change by sort of open debate; you get things sort of by accident, and I'm wondering — last year the administration proposed, and may again propose, large tax exempt savings vehicles, RSAs, LSAs, et cetera, that I guess are widely viewed as over time basically exempting all capital income from taxation, and that would be a fairly major change in the progressive tax system that we have. Does anybody consider it likely that you could have major change in the tax system through what some have even — the advocates have called stealth, and would this rank as a major change, in your book, if it were to go through?
MR. THORNDIKE: Is this a question for me?
Q: Anybody who wants to take it.
MR. PENNER: Well, as I tried to point out, I think we've been moving in that direction, toward a wage tax or a consumption tax, since the tax reform of '86, which moved the system with a jerk toward an income tax system. It would be another step in the same direction. I think those of us who feel favorably toward progressive consumption taxation feel that the transition problems are just gargantuan. They're just very hard to deal with.
So the advocates of that kind of movement within the administration favor proceeding gradually. The problem when you do it gradually is that you do all the easy things, like exempt the return from certain types of accounts, but you don't do the hard things which are an essential part of a consumption tax concept, and that is to punish borrowing by not allowing interest deductions. And indeed, if you do one without the other you create the danger of a lot of tax arbitrage, people taking bigger mortgages than they would otherwise so they can put the money in these favored accounts. I don't know how many people actually do that, but it is a big risk to the system.
Q: This is a question for Rudy Penner. I was wondering if you could walk us through Table 5 in your study. That looks really — (audio break, tape change) - your assessment of fairness here. And if you could just kind of make sure we understand some of your headings here?
MR. PENNER: Well, maybe I should direct the question to Jeff because that's his table. I'm not sure - what's your specific question? Sorry.
Q: I guess I would like to make sure I understand what you mean by "proposal." I'm not sure what proposal that is. And I want to make sure I understand what you mean by "percent of total income tax change," what the — I don't know. Maybe everyone else understands this; I don't know for sure.
MS. EISSA: Could you also identify yourself?
Q: Oh, I'm sorry. I'm Jill Barshay at Congressional Quarterly.
MR. ROHALY: Sure you don't want to handle it?
MR. PENNER: Sure.
MR. ROHALY: Table 5 shows the distribution of income tax changes due to both the 2001 and 2003 tax acts, which had in them a variety of sunsets. In other words, the 2001 Tax Act is supposed to expire at the end of 2010. Various portions of the 2003 act are supposed to expire between now and 2008, I think. This assumes that all of those sunsets are eliminated, so what would happen in 2013 if all of the provisions of the 2001 and 2003 Tax Acts were, in fact, extended? And therefore, what it shows is that percent of total income tax change shows you for each one of the income groups, where here the income groups are quintiles, who would receive the benefit from the income tax change. So for example, 3.8 percent of the total income tax changes from extending all these sunsets would go to those in the second quintile, those with incomes in the 20th to 40th percentile; 28.5 percent would go to the top 1 percent.
I forget what your other question is. I'm sorry. The average income tax columns show the average tax rate, average tax as a percent of income under current law, in which all of these things expire. The proposal would be extending all of the provisions of those tax cuts.
Q: (Off mike) — sunsets.
MR. ROHALY: Right.
Q: Ellen Nissenbaum at the Center on Budget and Policy Priorities. Very much appreciate all of your comments and particularly appreciate the modeling TPC has done. Following on Julie's comment, one of the changes that did not occur in major tax legislation that was focused on by a number of us working on health care policy but not focused on as much by folks following tax policy was the establishment, for I believe the first time, in the Code of Health Savings Accounts as universally accessible, where you do not pay tax either going in or coming out. So for the first time, you have a new mechanism that's universally available, where you basically pay no tax on income. Those I think may be followed on the heels by other proposals that are coming forward.
In addition to the LSAs and RSAs that Julie mentioned, there is a proposal floating called the RMBA, the Retirement Medical Benefit Account, that came up in — it's a great alphabet soup, but — that came up in Medicare drugs, where it's a sub-account in IRAs and 401s, and so there is no deduction in and no deduction out. I didn't know, Jeffrey, in your modeling, whether you're picking up some of these health tax proposals, and if not is that something that could be done because I think there's no question what the distribution would look like. There's fiscal implications that I think worry a lot of us, but there has also never really been much looking at the distribution of some of these health tax proposals.
MR. ROHALY: And the current answer is no, we don't at the current time have the capacity to do that, but obviously we're looking at all of the things that could show up in the budget this year, and we would like to be able to model some of these things. It's not a short-term project, unfortunately. We have modeled and we have produced estimates in the past of the Lifetime Savings Accounts that were proposed last year and will probably be proposed again in the budget this year.
Q: Tom Field, Tax Analysts. This question is for Joe Thorndike. You sketched for us, Joe, five periods, major periods in American fiscal history, and I wondered if you could tell us whether in your view the concept of tax fairness has varied in those five periods, and if so, how?
MR. THORNDIKE: I definitely think it varies. I mean, it's — the danger is to not get — get into thinking that there is a single definition of fairness in any given period because it's really all about argument. But in terms of what concept of fairness is by and large embodied in the tax regime, I think that there have been a variety. At the risk of sounding a little Whigish, I would say — (chuckles) — that the trend here is toward a faith in progressive taxation. You know, this was something new — not unheard of in, say, the early republic, but over the next 200 years or 150 years, it was increasingly the touchstone for good tax policy. By the late 19th century, the income tax had emerged among economists and political scientists as the benchmark for what constituted fair tax policy, and consumption taxes of all types were on the decline for the most part, except to the extent that they were called sin taxes or something like that and had some other claim to legitimacy.
But yes, I think in every period it has been a little different. I mean, the republicanism of the early United States is not a trend that I think you see — is not an ideology I see much at work today. I think by the time you get to the Great Depression and World War II, issues of redistribution, ability to pay — these have really taken center stage in most tax debates. In the years since then, while the tax regime has not changed, I think that ideological debate about fairness has changed substantially. And there is a good paper by Dennis Ventry that explains the tradeoff arguments about equity and efficiency that occur after World War II and the ways in which efficiency have sort of triumphed and have — that they — at last among economists.
And I think that that — we're still sorting out the results of that today. We still have an income tax that's based very strongly on notions of vertical equity, but it seems to me that most of the heavy lifting in academic circles is really done on horizontal equity. And it's — I think that has put the income tax at some danger at this point.
MR. PENNER: One thing we haven't talked about that I talked about only briefly in my paper is the whole philosophy of benefit taxation. And I think, Joe, that has been much more important in past historical periods than it is now. I think I remember John Stuart Mill debating the issue, and if I remember correctly he saw the main function of government as preventing the rich from exploiting the poor. But if that was the benefit derived from government, then maybe the poor should pay a higher tax. (Laughter.) So if I remember, he struggled with this point some. (Chuckles.)
MS. EISSA: Gene?
Q: Gene Steuerle. My question — I have long argued that the income tax is in some ways a conservative instrument of policy, but among the reasons are is that it's very visible, and I wonder if someone might want to comment on the extent to which the public views the fairness of a tax system by how much they see themselves paying.
MR. PENNER: Well, all I can say is that the polls seem to suggest that most people think that whatever they're paying the rich pay less. That seems to be a very strong view that comes out of the polls. And as I said, while that is apparently untrue — although I suppose David Kay Johnson might argue with the point — as shown in the usual distribution table, that's untrue. I find the public confusion very understandable given the huge variations that you see in how much tax is paid by those within each income group.
MR. THORNDIKE: I've seen a lot of numbers that would suggest, in really very contradictory fashion, that Americans, at least for the last 60 years or so, have almost all felt that they pay too much in taxes and yet are almost uniformly underestimating how much they are paying in taxes. And that's true almost straight through from when you first have poll data.
Q: Underestimating how much they pay in income tax or —
MR. THORNDIKE: Yes, and in all taxes together. You know, the questions vary, but — and I'm just generalizing on a huge number of polls that have been asked over time. But they underestimate what their total tax burden is, all taxes included. They also underestimate what they pay in income taxes because it's not such an easy question to answer necessarily when filling out your tax forms. But they're quite certain that they're paying too much and that nobody else is paying enough.
So I mean, you know, these are not surprising, I suppose, right? I mean, these are what you would expect people to say. I think what's interesting is that leaders have done such a poor job of actually communicating information to people in any kind of active fashion, and I think that's just the nature of tax policy. That's politics.
Q: Hi. Larry Haas. I worked at OMB in the last administration. I have two questions, one for Jeff and one for Rudy.
Jeff, speaking of Rudy's reference to David Kay Johnson, if he were here I suspect what he would say is that your distribution table is very interesting, but doesn't get at a very interesting phenomenon going on, which is you have to drill down within the top 1 percent and really see what's happening at that level. And in essence, that's kind of what the book is about: what's really going on with the rich, the very rich, the filthy rich, and why. And I'm just kind of wondering whether you're doing work to drill further into that group of people.
And then the second question for Rudy — and I apologize because I was very late, so if you went over this, again, my apologies, but I'm kind of wondering about whether you have a perspective as to whether we're spending, frankly, too much time in this town and in the tax debate in general talking about the income tax when three-quarters of Americans are paying more in payroll taxes; that is, that people have a skewed view of the tax system because of the way it's discussed in this town, without in essence absorbing a much larger phenomenon at work.
MR. ROHALY: To answer the first question our tables are basically limited to the data that we have. The data that we have allows us to at least look at, say, people making a million dollars or more, and we plan on looking a bit higher up the entrance scale than that. Part of the problem with our data is that a lot of the really interesting things that we would like to look at, we just don't get in the data file that's available to the public.
One thing that I am interested in working on is the notion that once you go high enough up the entrance scale, effective tax rates actually begin to fall because most of the income at the upper end is coming from capital gains primarily and dividends. And I think one of the tables in there actually shows that for those making more than a million dollars, the average effective income and payroll tax rate is lower than it is for the group just below. So stuff like that I am interested in looking at, with the data limitations, obviously.
MR. PENNER: Well, I think you're right that we probably do pay an inordinate amount of attention to the income tax, but I think it's all part of a very important point that Joe makes about how misleading the debate is. Just the fact we call the payroll tax a contribution on the wage slip is a pretty weird thing. I don't know how many times I have heard it said by politicians that the payroll tax is our most regressive tax, and I think that's the excise tax on beer, actually. (Laughter.)
The political rhetoric is really pretty awful. I think we all have to admit that.
Q: Jim Klumpner again. The Tax Policy Center and the tables are just wonderful. It's a major contribution, but as you acknowledged, they only are a snapshot of a particular year and they ignore any economic or political feedbacks that might result from a tax cut. That's the preamble to a question, which is what are the major fairness issues regarding debt-financed tax cuts?
MR. PENNER: Well, I think that's a really important point and a complicated point, and it's so complicated I didn't mention it in the paper. There are a lot of different ways of looking at it. On the one hand, you can think it's terribly unfair to be stealing from future generations who don't vote. On the other hand, you might say well, those guys are going to be richer than us anyway, so why not steal from them? So I think those are the two different strains of the debate. I ask, what have our children done for us lately. (Laughter.)
Q: Just an ambiguity struck me looking at your tables. Jeffrey, you're probably including the Earned Income Tax Credit in your tables. And wondering about the issue of — that really seems a lot like an outlay program. And you don't include other outlay-type programs, but you're struggling to include more types of taxes: payroll and state tax. And it's hard to draw a line, but I wonder if you might consider whether the Earned Income Tax Credit belongs in a table about tax burdens?
MR. ROHALY: Good question. Actually, and getting back to the last point, one interesting thing would be to look at the distribution of the spending cuts that are going to be required to finance the tax cuts that were passed in the last couple of years, but obviously our tables look only at the tax side.
I think just essentially we're following the standard that other people follow, and I think by leaving that our would give a misleading picture of what the tax system gives to low-income people. Of course, like you said, where do you draw the line between the gradual movement over the last few years of trying to format every new spending program because you can't have it funded as a spending program, as instead a tax credit, and therefore it's going to show up in our distribution tables. It's tough to draw the line.
Q: I wonder if I can follow through on that. It's actually one of the essays in the book here, but I wonder if we really count all tax and spending together, isn't it very likely that in the end, every tax increase that finances spending increases is probably going to redistribute for the most part to low-income people, assuming spending is a little more equally distributed than taxes, even the most regressive tax? And vice versa, isn't it very likely that most tax cuts financed by spending cuts are going to accomplish less redistribution? And if so — if I'm right, what are the implications for talking about equity of tax and expenditure changes, or don't we eventually have to ask questions about what's the ultimate size of government we need to achieve or we want to achieve?
MR. PENNER: Well, I guess I doubt your basic proposition because so little of the government budget goes to means-tested type welfare programs. And I think that it's very hard to distribute the benefits of most government spending, like defense most notably. So I wonder if the spending side is really as progressive as you imply. I don't know the answer to that.
Q: Julie Kosterlitz again. Following up on the question about moving more towards a tax on wages and/or consumption, I'm not sure I understood — I think you made a couple allusions to this being a sort of general trend, and I'm wondering maybe both for you and Mr. Thorndike why so much more emphasis on lowering in this era or I guess even just the past few years on moving towards wages and/or consumption?
MR. PENNER: I didn't mean it was a really long-term trend. I meant since '86. And I think as Joe pointed out earlier, we tended to tax consumption much, much heavier.
Q: Why now, then? Why -- (inaudible)?
MR. PENNER: Well, I struggle with giving you an explanation of that. I gave the three explanations in my paper, one a purely philosophical judgment that is 400 years old and comes originally from Thomas Hobbes. That is the notion that it is fair to tax people on what they extract from an economy as opposed to what they contribute to an economy. So that's a deeply philosophical argument which I find pretty persuasive.
The second is a more pragmatic argument. Legislators worry about dampening savings and investment too much. There's another pragmatic argument that I didn't mention. I think it's very difficult to measure capital income in a fair way. There are problems such as adjusting for inflation and measuring real economic depreciation.
And the third more cynical argument is that you seem to get more relief for capital income when the top marginal tax rate is high, Then the tax structure makes it appear that the system's very progressive, but the Congress makes exemptions for capital income that provides tax relief at the top which is disguised.
I'm not sure why politicians would play that game because, as I said later, there doesn't seem to be a real strong feeling in the country that the rich should be effectively punished or that there should be a great deal of redistribution in the tax system. But in any case, all I can say is that it is certainly a revealed preference of the Congress not to tax the economic definition of income.
MR. THORNDIKE: I do think that this — the interest in consumption-based taxes has certainly risen dramatically over the last 20, 30, 40 years. It's worth point out, though, that it's not entirely new. There was a proposal during World War II to establish a progressive consumption tax that operated in administrative terms much like an income tax, but that would have exempted all savings from taxation. It was definitely an issue — and even then it wasn't new. The idea had been kicking around for 20 years that applied progressive rate structures to a consumption-based tax. And it never went anywhere. I mean, it was called "Morgenthau's morning glory" because it opened in the morning and it closed before noon — (laughter) — which is actually a literally accurate description of it. It died in a single day.
But there was even then I think a growing appreciation among economists that a consumption-based tax that involved progressive rates might actually be a fair, and in the case of the war, economically advantageous tax structure because they were most worried about stemming inflation and they believed that consumption taxes might help with that.
MS. EISSA: I have actually a follow up question to Gene's, a question about accounting for the spending side. And it is difficult in many cases to account for spending in terms of where it would fall on the income distribution, but in some cases it's not. For example, with Social Security, if you looked simply at the tax side, we do have a regressive tax, but there are clear benefits from the Social Security system. And when you account for the benefit side, the distributional analysis might look different because the benefit formula — the benefit's schedules is very progressive.
And that brings up another point, which is when one looks over their life cycle — and that's a point that both you and Jeff have brought up. When one looks over their life cycle, one can get different distributional numbers in terms of what the tax system does. And so the point still stands. It's — the question is still valid, I think. Should we make an attempt to incorporate the spending side when we look at the distributional analysis of tax systems?
MR. PENNER: Well, I think in theory we should. What I find interesting about the spending side is that we have chosen to try to help the poor with grants that seem to be associated with various kinds of in-kind consumption. So we don't just give them money. We give them something called food stamps or we give them housing subsidies or we give them Medicaid. And I think certainly in the case of food and housing, the way the programs are structured, they depend on your income and not on how much food or housing you buy. So I think you can make a case that they have very, very little effect on the consumption of those items, and yet it's clear that that sort of support to the poor is much more acceptable to the public than just giving them money.
I wonder how well the policymakers really understand exactly what they're doing here. Is it just to make aid for the poor more acceptable — or do they really think they're increasing the consumption of food and housing?
MS. EISSA: Thank you very much. I think we will stop there. Thank you all for coming.
(Applause and end of event.)
No comments:
Post a Comment