Will WE Ever Get Tax Reform? (The Politics of Tax Reform, by GWB Economic Advisor)

Delivered at a Meeting of the National Bureau of Economic Research November 4, 1997

--------------------------------------------------------------------------------

In January 1993, departing Treasury officials left some advice written on a blackboard for their incoming colleagues: "broader—base lower rates". This mantra is now broadly accepted as good advice by the Public Finance community, although admittedly we may disagree on the details of base broadening and on just how low rates should be.

The departing officials had fought the good fight. The measure of their success was the same as all good volunteers who come to Washington for a brief period to fight for the causes of truth and justice—minimize the amount you must retreat. The 1990 budget deal could be judged as a modest retreat in this regard, but a retreat nonetheless.

Regrettably, their successors have been present for the military equivalent of a rout. The 1993 budget deal saw the largest increase in tax rates since the 1930s accompanied by a small modicum of base narrowing. The 1997 budget deal saw plenty of base narrowing but no rate reduction. Let me stress that this result is not the fault of the people at the Office of Tax Analysis. Anyone who has served as an economist in Washington knows that the deck is stacked against you.

Today, eleven years after tax rate cutting accompanied by base broadening seemed to have garnered a bipartisan triumph, codified in the 1986 tax reform act, we appear to have come full circle and are now back where we started. Why is that the case, and will we ever achieve lasting tax reform?

To understand our decade long retreat on taxes, it is important to understand the fundamental dynamic that motivates politicians in this town: the need to appear to be doing something. Consider the recent budget deal. Had the President and Congress not done anything, our nation's fiscal position would have been incomparably better off. In practice, these budgeteers took the soaring capital gains revenue that has resulted from the stock market boom and spent it on higher domestic spending, broader Medicare coverage, and tax reductions which are devoid of incentive effects. If the objective really had been the nation's fiscal health, they should have stuck to oversight hearings. But then neither the President nor the Congress could claim credit for what would have happened anyway. So, they had to be seen as causing the deficit to go down, even though their actions will have the opposite effect.

Now consider the sad fate which awaits the tax reform process over the long run. The Congress, which enacts fundamental and substantive tax reform, can go home to the voters in triumph, rightly proud of actually having done something useful. But what of the next Congress or the one after that? If they just leave well enough alone their election opponents will call them "do-nothings". So there is an endless incentive to tinker.

Note further that this endless incentive to tinker will be especially pressing on the tax writing committees. Their colleagues on appropriations and authorizing committees can tinker to their heart's content. They can always raise some program a few billion, pursue some invariably popular cause like attacking cigarette vending machines, and look like they're doing something without actually doing much damage. But on the tax writing committee all you can do to compete is either chip away at the base with special exemptions or raise one person's taxes to cut someone else's. Neither is conducive to a stable tax code with low rates and a broad base.

To maintain a broad base and low rates we must find a mechanism to make tinkering unpopular. More to the point, we must find a way of making it so that the marginal voter is opposed to changes in the tax code, or at least so that Congress thinks that is the case. We are dealing with political economy, not economics, and our focus should be on the fact that political economy, like markets, is a dynamic process. As we learned from 1986, what may be politically achievable in a one shot deal, may not be sustainable in the long run.

First, let me note two political facts which are key to producing fundamental tax reform: revenue realities and IRS abuse. The revenue reality is that the recent surge in income tax receipts on which the budget deal depends is based on two unsustainable trends in personal income. The first is the enormous increase in the share of pre-tax income going to upper income taxpayers. If the 1995 share of income received by the top 5 percent of the population were to return to its 1985 level, Uncle Sam would be out roughly $45 billion in personal income tax receipts without any decline in the level of personal income. The second revenue reality is that we are in the midst of an unsustainable surge in capital gains tax receipts. In 1995, capital gains realizations were actually 10 percent below their level in 1985, even though the stock market was three times as high and personal wealth had doubled. The 28 percent capital gains rate, as predicted, led to a sustained decline in realizations. In 1996, we saw a swelling of involuntary realizations through mutual fund sales, enough to explain $20 billion of the recent revenue surge. The recent tax rate reduction will, without a doubt, produce a surge in realizations in the next year and a half as a decade long retention of gains is ended. But then what? If the stock market should falter and if the income distribution should return to more normal levels, the pressure on receipts will be enormous. Given that top tax rates are probably already at or near their maximum level, the pressure for reform will be intense from a budgetary point of view sometime shortly after the next Presidential election.

The second political fact is the unpopularity of the IRS. This unpopularity is hard to overstate, and has become much more intense recently. In recent years managerial practice at the IRS has come to stress production quotas from agents. While the recent hearings on the Hill have forced the President to suspend this egregious practice, the lingering malignant bureaucratic culture, and public perception of it is intense. Outside of the Beltway this is a very live political issue. In addition, circumstantial evidence is building that the IRS has been used for political purposes to an extent which we have not seen since Nixon. While this is not the source of mass discontent, it has energized both political and policy making elites inside the Beltway to favor reform to an extent which I have not seen in my years in Washington.

Now let me be clear. The rhetorical phrase "ending the IRS as we know it" is of exactly the same quality as the phrase "ending welfare as we know it". A national revenue collecting agency is inevitable. But the level of discontent is such that I believe the path is irreversible. Like the revenue realities, this discontent has not yet reached critical mass. But, I do think that these forces will coalesce around the time of the next Presidential election to produce the momentum needed for reform.

If these factors are building toward creating an environment conducive to reform, what should supporters of reform do to get ready? Interestingly, the answers are the same for the academic public finance economist desiring to write a policy relevant paper as for the political tactician planning his candidate's next campaign positions on tax reform.

Factor One: The Distribution Argument Has Changed. For as many years as I can remember, tax economists have been generating distributional tables designed to show the impact of the tax cut on various income groups. The academic debate has centered on how to define income. At the simplest level, some tax-relevant definition such as adjusted gross income has been used. More sophisticated models impute various kinds of income to produce such notions as expanded income. The central political push behind all of this activity has been for Democrats to say that a given tax change "favors the rich" and the Republicans to produce a table which says, "no it doesn't". One need only watch the wrangling over the budget deal to see this.

I do not believe that such tables will go out of fashion. Envy is too important a human emotion for politicians to abandon. But, the tables really miss the key point: the American household is much more diverse than it was when Stan Surrey and Richard Musgrave first had the profession stress distribution. Today, the type of household is a key determinant of that household's income. It is also a key determinant of how that household votes. So, the distributional tables that will really matter will be those that separate households by marital status and the presence of children.

Intellectually, it is certainly appropriate that we do this even if it had no political relevance. One of the factoids I remember from Samuelson's Eighth Edition was that "two can live as cheaply as 1.6." Samuelson of course was debunking the 1950s argument that young people were in a hurry to get married because two could live as cheaply as one. Today, the "two" are more likely to be a mother and her baby with Dad no where to be found. Regardless of the details, a tax system based on the "ability to pay" should make appropriate adjustments for family size.

Distributional tables which take no account of family type also really lose their policy relevance. For example, the median income of a four person married couple family was roughly $60,000 in 1995. This was enough to put them at the cutoff of the top quintile as defined by the traditional tables. And using expanded income concepts which include the imputed rent on the family's home and the buildup on its life insurance and 401(k) plan, the family can be made to appear rich for political purposes, and in fact they were during the last budget debate.

But it is really a stretch to equate this median income family with a single yuppie lawyer making the same nominal income. Economically and intellectually, our traditional distributional tables need reform which takes household type into account. But the need for change doesn't stop with intellectual honesty. Politically, family type is, along with church attendance, one of the most likely predictors of party affiliation. I would refer you to no less an expert than Dick Morris, who although not the type of guy you would like your daughter to bring home for dinner, has been one of the most successful political strategists of our time. Paraphrased, his advice to Clinton in 1995 was, "if they're married with kids we've lost them, if they're not we can get them". In the end, while married people with children voted as a group for Dole, the rest of the country voted overwhelmingly for Clinton. Apparently, if you're in a position to worry about who your daughter might bring home for dinner, you have different tastes in who you want in the White House.

This political fact drove the 1997 budget agreement over taxes. I have yet to meet an economist who has a nice thing to say about the $500 child tax credit. But, the reason that it was the centerpiece of the Republican agenda was that the credit went to people inclined to be Republican voters. The response of the Democrats was two fold. First, it was to point out that too much of the Republican plan went to the "rich". This is politically useful rhetoric at securing the largely unmarried and childless Democratic base. But as noted above, this rhetoric loses something in the translation to actual married families with children. So, the Democrats lost this part of the argument and the credit was extended fairly far up the income scale. But, in turn, they won something for some of their constituents - that is, single parents, many of whom pay no taxes, or who receive the earned income tax credit. Making the $500 credit largely refundable was a key Democratic victory in the last budget deal.

In 1998, the Republicans are likely to come back with a plan to abolish the marriage tax. In its purest form, this proposal would allow a married couple to choose whether it wanted to file jointly or as two singles. This plan makes the new distribution argument - based on family, not income - most clear. After all, it is only married people who either pay a marriage tax or get a marriage bonus. The Republican plan is ingenious. Core Republican voters, married people with a primary wage earner and the kids at home, will continue to elect the current joint return and are held harmless. Two earner married couples may still have kids at home, but a great many of them may be empty nesters. So most of this Republican tax cut will be targeted at a voting group which is inclined toward Republicans but somewhat vulnerable to poaching by the Dick Morris strategy.

I highlight this to stress how important the demographic factors will be in any type of substantive tax reform which will happen after the next election. In all likelihood the next tax reform will be roughly distributionally neutral from a traditional income point of view and different factions will be able to produce their own distributional chart to prove their point. But, the real distributional argument will have changed. Will married families with children continue to shoulder their historically high share of the burden, or will it be shifted to others, and if so, to what extent?

Factor Two: Labor Will Be Favored Over Capital. I have sat at many an NBER conference table where the yellow papers under discussion have been variants on the theme of the excessive taxation of capital. Double taxation, taxation of inflation induced capital income, insufficient depreciation allowances, you name it; small forests have been felled to generate the resulting working papers. It won't matter. The next tax reform will reduce the effective tax rate on labor income relative to the effective tax rate on capital income.

There's a good economic reason for this. All of those working papers have had already their effect. The tax rate on capital income, which was clearly excessive during the late 1970s has dropped sharply relative to the tax rate on labor income. Depreciation has been improved. Subchapter S Corporations have become more widespread. Inflation has been reduced. The expansion of IRAs and 401(k)s has allowed taxpayers to shelter retirement saving from current capital income taxation. And, in the last tax bill, the capital gains rate has been cut. At the same time, the tax rate on labor income has risen. The 1993 tax bill not only raised the top rate by more than 11 percentage points, it also added nearly 3 points of tax on labor income alone by removing the income cap on the Medicare tax.

While the economics literature has helped level the playing field between capital and labor, it has failed to establish an incentive-based case for low taxation of capital. It is certainly reasonable to assert that the substitution effect outweighs the income effect with regard to the taxation of capital income and that therefore, a higher real after-tax return to saving will raise the saving rate. But, this fact has not been empirically established.

Complicating this has been the support for the notion of Ricardian equivalence provided by our recent experience with the budget. The budget deficit was reduced largely by higher tax receipts from the rich. While government dissaving was reduced, private saving was reduced still more and our net national saving rate has actually declined. In this regard, the 1993 budget changes have failed, as has our profession's claim to know how to boost national saving.

On the political side, I do not have to stress how many more voters are affected by labor taxes than capital taxes. As an example, I would point to Steve Forbes' campaign last year on the flat tax. Politically, the idea of a flat tax is enormously popular. But, Forbes proposed exempting capital income on the grounds that it had already been taxed once when it was earned. In the run-up to the New Hampshire primary where Forbes had been leading, Dole campaign ads which pointed out that rich people like Forbes would pay no taxes, were a key to turning the race around. The next tax reform will therefore be much more likely to raise taxes on capital rather than lower them.

Factor Number 3: Don't Forget Transfers. Comprehensive tax reform is going to involve substantially closer integration of our tax and transfer system. This is not only good tax policy, it is economically inevitable, and if done right, politically attractive as well. The economic facts of life are that transfers have become too important to ignore. Transfer payments now amount to $1.1 trillion dollars, $4400 per person, one dollar in six of personal income. They are bigger than interest and dividends combined and as big as corporate profits and proprietors' income taken together.

Politically, it seems unlikely that fundamental tax reform can be undertaken without also revamping our middle class entitlement schemes. Some form of fix of these schemes is urgently needed. As a country we are going to have to decide whether to rely more on compulsory private saving or continue with the current pay-as-you-go approach. The answer to that question will decide what type of fundamental tax reform we undertake. But whichever type that is, a closer legal integration of taxes and transfers seems likely, and a more rigorous examination of the net effects of these government policies by the political process seems inevitable.

What type of tax reform will it be? There are three main approaches now on the table: a value added tax, a progressive consumption tax, and a flat or flatter rate income tax with a much broader base. The decision making criterion will be short run feasibility. But an equally important criterion should be long run sustainability, meaning, as I said in my introduction, that Congress not have an incentive to make frequent changes. By those two standards, I think the process of elimination leads us to one clear path for reform.

Let's start with the VAT. Last year, Nick Bull and I wrote a piece on the monetary policy implications of a VAT for the National Tax Association's conference. We reached a fairly unambiguous conclusion: the monetary authority should not facilitate a rise in the retail price level to accommodate the effects of switching to the VAT. This really should not be a surprising conclusion. From a macroeconomic standpoint, would anyone recommend a change in monetary policy to account for a revenue neutral change in fiscal policy? Of course not.

But the reaction to our paper from the tax community here in Washington was striking. Everyone had assumed that such a change would involve a shift in monetary policy. Indeed, most found our conclusion unthinkable. This reaction convinced me that a VAT really is unthinkable. Accommodated by monetary policy, a VAT quickly becomes a one time wealth tax coupled with a prospective labor income tax. It is the one time wealth tax that actually gives it the political cache to make it a viable option. Since a significant fraction of the tax falls on existing wealth which is held in relatively few hands, the politicians can, in effect, promise the great majority of both workers and capitalists lower taxes on their future earnings and saving than would otherwise be the case. Take away the option of wealth confiscation through inflation, and the political appeal of a VAT vanishes.

Nick Bull and I also concluded that the adjustment process was actually less distortionary if the monetary authority makes this clear from the outset. Thus, our conclusion was based on the long run adjustment process. Of course, the short run transition from an income-based tax to a VAT is also complicated from a macroeconomic point of view. I would, therefore, have to conclude that the VAT option does not pass the feasibility criterion I mentioned earlier.

I also doubt that it is really sustainable. The history of VATs is that they are introduced at low rates and then raised. It is also doubtful that a switch from an income tax to a VAT could be complete. Politically, the VAT would just be viewed as yet another tax. The party which implemented it would therefore bear the onus of having added yet another tax to our system. So, no, there probably is not a VAT in our future, at least as the centerpiece of fundamental tax reform.

The progressive consumption tax has a different set of problems. First, it is based on the old distributional system. The result is excessively high rates which makes it unattractive on an incentive basis. At the same time, it does nothing to resolve the new distributional problem of the tax treatment of families with children. Notably, it is largely pushed by the business community which has not appreciated the important political and economic shift among our nation's families.

Second, it distinctly favors capital over labor. Its fundamental premise is to encourage national saving by taxing only consumption. But, as I noted, this is still an unproven point intellectually. Third, it is not a sustainable program because it looks too much like our current tax system. There will always be the incentive to raise the other guy's tax rate in order to lower the rate of people who vote for you. I therefore find this option of largely intellectual interest only.

This brings us to the flat tax. Initial feasibility is an open question; it will only be feasible if a national election produces a mandate to make such a change. But, unlike any other proposal, a flat tax introduces an element of sustainability into the equation. The key is the single rate. All taxpayers can understand it and all have a stake in it. Really tinkering with the rate structure will require breaking the single rate commitment, and the case can be easily made that when politicians break this kind of commitment, everyone should hold onto their wallets. Of course, with a substantial zero bracket formed by personal exemptions, there is always room for tinkering. But the chances of raising the basic tax rate and affecting that median voter seems quite remote.

The flat tax also could address the three main factors described above. Most of the problem with our differential tax treatment of marrieds and singles stems from the system's progressive rate structure. Under any tax, there might still be a "marriage bonus" in that a non-working spouse taking care of the children can have his or her "zero tax" amount applied to the family's primary earner. But I doubt that anti-natalist sentiment could actually grow so much as to consider this a problem. I would see an exemption amount of roughly $12,000 per adult and $7000 per dependent as viable. A four person family would therefore not pay any taxes until they reached $38,000 in income.

Second, the flat tax would end most of the existing differential treatment of capital and labor. This would likely include capital gains. With a flat rate in the low-to-mid 20s, this would mean a modest increase in the capital gains tax rate. On the other hand, such an increase would very much assist making the old-style distributional tables look more neutral. One could also imagine limiting the exemption to labor income. A single rate on all capital income, no matter who receives it, would take care of a lot of the complexity in the tax code including such bizarre concoctions as the "kiddie tax". It would also modestly improve the distributional tables and be decidedly "pro-family". Another issue in the capital taxation area would be the means of ending the special treatment of retirement saving. Including the income from such investments in the base would allow a substantial reduction in the effective flat rate. On the corporate side, there is no reason why the corporate rate should not be lowered to the personal rate level. The fiscal offset would come from ending the option to both deduct interest and depreciate the capital purchased with borrowed money. This would mean some move to a cash-flow tax base, but in which direction is far from clear.

Finally, such a flat tax would have all transfers in the tax base. Politically, this can only be accomplished with the fairly high zero bracket exemptions described above. Should such a flat tax include an earned income credit option, then a lot of these transfers could be used in lieu of the credit. The highly variable taxation of low and moderate income families, and the resulting high tax rates on initial labor income could be eliminated in the process.

In sum, I do see fundamental tax reform in our future. The concept of a broad base and a low rate is an intellectually attractive one. While far from the theoretical ideal in defining the base, a sufficiently low tax rate can make up for the errors which result from a less than ideal definition of the base. In the past, we have failed to cement the gains of such a regime because we left too much room for tinkering. To be sustainable, the next tax reform must be quite sweeping in its scope and must establish a clear benchmark from which politicians are not tempted to deviate. What has changed of late is the political feasibility of creating such a system. Depending on the will of the voters, a broad based flat rate income tax with a rate in the low 20s, and a generous exemption amount, may well be on the table after the next Presidential election.

To maintain a broad base and low rates we must find a mechanism to make tinkering unpopular. More to the point, we must find a way of making it so that the marginal voter is opposed to changes in the tax code, or at least so that Congress thinks that is the case. We are dealing with political economy, not economics, and our focus should be on the fact that political economy, like markets, is a dynamic process.

What I believe to be the key of Lindsey's points.

HR2525 provides the mechanism by being a single rate visible tax with a pre-paid monthly tax payment offset paid to everyone rich and poor.

The monthly offset meets the requirement for the bottom end wage income folks to meet their tax payments yet participate in and resist increases in the levied tax rate. Any shifts in the tax system incidence causes an increase in rates for everyone.

The key is maximum participation and perception of the tax, while providing a basis of self interest to keep the system in place.

The Flat Tax system, by maintaining personal exemptions, taxes on business and not removing payroll taxes, provides the illusion of free lunch for those most dependant on government largess and panders to the idea of the rich guy and nasty of businesses paying the piper.

As youknow I can easily live with HR2525 (ot the Tauzen NRSTIF the 16th amendment is repealed). I do however the battle with the socialist will never end, ("The Price of Freedom is Eternal Vigilance").

I live in a State with both an Income Tax and a Sales Tax. Special interest are constantly fiddling with what is covered. In the case of the NRST, there will be calls to exempt cerain items from the tax, Food will be one... to 'help the starving children' and 'for working families'. I realize that there is an exemption for a rebate that would make this unecessary, but since when would that matter to rat politician like Clinton?

In the case of the Flat Tax the battle will be to reintroduce other rates for the "Rich" so they pay their "Fair Share".

I don't believe the rats will ever give up on class warfare. Class Warfare the Race Card and Entitlement programs are who they get their base to vote.

gives a clear explanation of the difference between a Flat Tax and a VAT, So I must disagree about a Flat Tax being a VAT.

Perhaps you will explain the difference to us?

The Flat Tax is a VAT implemented with an individual Wage Tax.

The computation of a VAT is :

Taxrate*(Sales less purchases) = Taxrate*(GrossIncome - investment)

You must apparently disagree with Hall & Rabushka, the architechs of the Armey Flat Tax.

The Flat Tax; Chapter 3, by Robert Hall and Alvin Rabushka

In our system, all income is classified as either business income or wages (including salaries and retirement benefits). The system is airtight. Taxes on both types of income are equal. The wage tax has features to make the overall system progressive. Both taxes have postcard forms. The low tax rate of 19 percent is enough to match the revenue of the federal tax system as it existed in 1993, the last full year of data available as we write.

Here is the logic of our system, stripped to basics: We want to tax consumption. The public does one of two things with its income—spends it or invests it. We can measure consumption as income minus investment. A really simple tax would just have each firm pay tax on the total amount of income generated by the firm less that firm’s investment in plant and equipment. The value-added tax works just that way. But a value-added tax is unfair because it is not progressive. That’s why we break the tax in two. The firm pays tax on all the income generated at the firm except the income paid to its workers. The workers pay tax on what they earn, and the tax they pay is progressive.

To measure the total amount of income generated at a business, the best approach is to take the total receipts of the firm over the year and subtract the payments the firm has made to its workers and suppliers. This approach guarantees a comprehensive tax base. The successful value-added taxes in Europe work this way.

***

The other piece is the wage tax. Each family pays 19 percent of its wage, salary, and pension income over a family allowance (the allowance makes the system progressive). The base for the compensation tax is total wages, salaries, and retirement benefits less the total amount of family allowances.

Concerning Proposals for a Flat-Rate Consumption Tax
Before the Joint Economic Committee, Statement of Robert S. McIntyre
Director, Citizens for Tax Justice May 17, 1995

The Hall-Rabushka flat tax, and its Armey and Specter variants, would replace the current personal and corporate income taxes with a new tax that is conceptually identical to a "subtraction-method" value-added tax, a version of a national sales tax, with two major modifications: First, imports would be exempt from the flat tax, while exports would be subject to tax. Second, to mitigate the regressivity of the VAT, cash wages, pensions received and other cash earned income would be taken out of the VAT base (i.e., deducted by businesses) and taxed directly to individuals, with exemptions. Thus, structurally, the flat tax is exactly equal to a value-added tax with an import incentive, an export disincentive and a personal rebate based on a percentage of a capped amount of cash wages, pensions and other earned income.Businesses compute a typical "subtraction-method" value-added tax by adding up their taxable gross receipts and subtracting the cost of previously taxed items. Thus, in computing the VAT, businesses deduct their purchases from other businesses, whether for supplies, services, machines, land or whatever.

When I refered to the VAT (and I believe when Lindsey refers to a VAT) I am refering to a Eurpoean style VAT where products have a tax added at each stage of the production process. THe tax is at a lower rate and is "hidden" from the consumer. A single item is taxed at multiple points. A NRST like 2525 avoids this problem by taxing the item only at the point of sale.

In the Case of the Flat Tax it depends on How the tax is implemented (and in any tax reform proposal the devil is in the details!) Some proposals, tax Corporate profits (gross revenue minus plant, equipment and wages) then taxes the employee wages, In this case the revenue is taxed only once at it's point of origin. Ohter variations remove all business taxes, and tax the vusiness owner (shareholder) dividends (or in the case of stock sales capital gains) at the same rate as any other income. This is to avoid the political canard of the 'rich' not paying taxes on their interest income.

The current system is a real VAT also in the case of business income and captal where revenue is taxed 2-3 times and rates going well over 50%.

It is interesting that Hall Rabushka and Mitchell describe the Flat Tax as a Consumption tax the same as an NRST such as HR2525 or the Tauzen NRST.

Thanks for the link!

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.

Concerning Proposals for a Flat-Rate Consumption Tax Before the Joint Economic Committee, Statement of Robert S. McIntyre Director, Citizens for Tax Justice May 17, 1995

Concerning Proposals for a Flat-Rate Consumption Tax
Before the Joint Economic Committee, Statement of Robert S. McIntyre
Director, Citizens for Tax Justice May 17, 1995