Place Ideal Tax Scheme in Constitution

Balanced Budget Amendment

By Alec Rawls © 1998. (6800 words)

How to un-kill ten albatrosses with one stone: use a balanced budget amendment to place the ideal tax structure in the Constitution and completely strip Congress of its tax powers. Economics and moral theory are sufficient to deduce the ideal tax structure, which then would be used to raise whatever amount revenue is expected to be necessary to balance the budget (or the full employment budget) given existing spending programs.

We would get a balanced budget (removing a dead albatross or two) and we would get rid of fully half of the opportunities for corruption in government (removing another eight or nine). Some issues can only be handled through the flexibility of a sitting Congress, but since the ideal tax structure can be definitively specified for all contingencies, Congress can only do worse, and since Congress is prone to corruption (especially where money is concerned) it will certainly do far worse. Using a balanced budget amendment to strip Congress of its tax power is the most important single thing the United States can do to secure the foundations of good government.

Automatically raising taxes to cover expenditures is not by itself a very appealing idea for many taxpayers. The very popular Reagan tax cuts were premised on the idea that the way to reign in spending is to cut off the amount of money that Congress has to spend. The same thinking holds conversely that, so long as revenue is available, government will grow. By this reasoning, the worst thing to do, when faced with burgeoning government spending, is automatically raise taxes to cover spending.

I don't buy this theory. I was on the side of Milton Friedman in the 80's who maintained that politicians would rather borrow and spend than tax and spend because tax increases anger constituents while government borrowing, since it only incurs future tax liability, is relatively invisible to taxpayers. When taxes are cut and the revenues are replaced with borrowing, spending is made less painful for politicians. Thus automatically raising taxes whenever a spending bill is passed would not promote spending but would actually put substantially more restraint on spending than if the cost is passed down to future taxes through borrowing. Ideally, the tax increase necessary to cover any spending proposal would be calculated and specified (by an independent tax accounting office, not by the politicians) and reported at the bill the authorizes the spending, so that every day the newspapers could report whose taxes went up by how much.

Ultimately, neither taxes nor deficits can impose proper restraint on spending. The problem is that politicians just love to spend other people's money. In particular, they love to spend it on their friends (who they define as anyone who contributes to their campaign war-chests). We also need direct controls on the spending side. Most importantly, there should be a public interest test for legislation. Another scheme for controlling the spending side is my earlier proposal for Billing Aid to Account. On the tax side, a balanced budget amendment that places an ideal tax structure in the Constitution and automatically sets tax rates to cover expenditures would be very powerful in limiting government to its proper role. Since taxes exert greater restraint than deficits, the balanced budget aspect imposes restraint, and at the same time, putting the ideal tax structure in the constitution eliminates all opportunities for government corruption to operate through the tax code, which is gigantic. Tumorous government growth, fed by massive corruption and stupidity, is the greatest threat to our survival (greater than China or the risk of environmental degradation, because if we bankrupt ourselves, we won't be up to these other challenges). We need to rationalize both spending and taxes, ASAP.

The ideal tax scheme is almost completely specifiable. What features are not completely resolved by first principles are second order considerations where the range of reasonable solutions are themselves easily narrowed down.

A balanced full employment budget

The concept of a balanced budget needs to be clarified. It is not desirable to have the budget in constant balance because, as the economy falls into a recession, tax revenues automatically fall and safety net expenditures automatically rise. Maintaining a balanced budget throughout a recession would thus require increasing the rate of taxation and and cutting government spending, both of which would exacerbate the recession. What many economists advocate is setting tax rates so that the budget would be in balance if the economy were at full employment. An amendment requiring that the "full employment budget deficit" remain at zero would allow some deficit spending in recessions but keep debt in check over the long run.

Efficiency and distributional objectives

With a balanced "full employment" budget as the revenue target, the basic elements of an ideal tax structure for raising this revenue can be readily articulated. A tax system has both efficiency and distributional goals. Regarding distribution, Bernard Shaw hit the nail on the head when he declared that "lack of money is the root of all evil." The ultimate or long term distributional goal that government policies should be aiming for is a fortune for everyone -- what John Stuart Mill called "a living" -- enabling people to weather their liquidity crises themselves rather than have to rely on government safety net programs. The immediate or short term distributional goal is to enable people to get a leg up on life, both by serving as a lender of last resort to help people weather their liquidity crises when they have no other resources to fall back on, and by trying to keep the bottom rungs of the economic ladder within easy reach, so that people whose circumstances are still precarious can live productively and inexpensively, providing them the wherewithal to improve themselves and their prospects if it is in them to do so. Efficiency goals are also straight-forward. Where markets work they are the most efficient way to produce goods and services and income and wealth. Thus the efficiency goal is to help and allow markets to work, or failing this, to reproduce how efficient markets would work.

The flat income tax

These efficiency and distributional goals can be optimally achieved a properly designed income tax and the right division of labor between income, micro-economic, and inheritance taxes. To allow markets to work, what is needed are low marginal tax rates on income. This can be best achieved by a flat rate income tax with a personal deduction large enough to cover necessities, and no other deductions allowed. By limiting deductions to necessities, the tax base is maximized, which allows the tax rate to be minimized, which minimizes the the disincentive to productivity that all tax systems incur.

(The idea of a flat tax with an exemption for necessities goes back to J. S. Mill's Principles of Political Economy, where Mill claimed to be following Adam Smith. See Principles, BookV, chapter II, especially §3. The flat tax idea was brought to modern prominence by Robert Hall and Alvin Rabushka of Stanford and the Hoover Institution, who in their book The Flat Tax (Hoover, 1985) offer a fully spelled out flat tax plan that highlights the simplicity and efficiency that can be achieved by eliminating all deductions beyond the personal one. The several flat tax plans that have been at the center of political debate in recent years -- the plan that Forbes made the centerpiece of his '96 bid for the presidency, the Armey-Shelby flat tax bill, along with several others -- are all versions of the Hall-Rabushka proposal.)

In addition to optimizing efficiency, a flat tax with only a personal deduction is also ideal for pursuing the short term distributional goal of making life as easy as possible at the bottom of the economic ladder. All of the flat tax's progressivity is achieved through the large standard personal deduction, after which amount the flat tax begins to apply. This concentrates progressivity at the bottom, where it is most needed. A graduated tax, in contrast, spreads its progressivity further up the income scale where it is less needed, but where the distortionary effects are more severe. (It isn't just that those who are earning more, being farther from need, have more options to work less, but that these are the people who create jobs for others, so distortions here are magnified by spillover effects.) Thus distributional and efficiency concerns both call for concentrating progressivity at the bottom. No matter how much progressivity one thinks there should be, a flat tax with personal exemptions is the best way to achieve it.

The job of determining the proper amount of progessivity in this case is transmuted into the problem of how to define and measure the concept of "necessities." Ball-park figures from the various flat tax proposals run from an exemption of $25,000 to $35,000 for a family of four. Exactly where to set the level of progressivity is essentially a political decision that the nation would have to make. It could either remain constant in real terms (i.e. be adjusted over time for inflation), or it could be revised over time according to articulated principles for defining what is to constitute a "necessity". If principles were articulated, they would then be followed by the same tax office that was charged with setting tax rates so as to achieve a balanced full employment budget. Since necessities are by definition more basic needs and hence not very prone to change over time, I would think that the best approach here would be to settle the political question definitively at the time the amendment was passed by settling on a simple real dollar amount for the personal deduction. If it turned out to be unsatisfactory, it could be altered by future constitutional amendment. But if a clear enough definition of necessities could be articulated and agreed upon, it could be better to put this adjustment mechanism in the original amendment.

As for allowing no deductions beyond the personal one (or one's, when there are dependents), this is utterly compelling on several grounds having to do with the proper role of government, economic efficiency, political corruption, and the division of labor between income taxes and micro-economic taxes. Consider the latter first.

Micro-economic taxes

One necessary feature of an ideal tax structure is micro-economic taxes to correct the market failure that occurs when market prices fail to account all costs. The price of gasoline, for instance, reflects only the costs of bringing gasoline to market. It does not account the cost of the air pollution that results from using gasoline. When people decide whether using a gallon of gasoline for some purpose is worth the cost, the price should ideally reflect all costs, so that if the value to the person of using the gas is higher than the price, his private decision to use the gas will be the socially optimal decision too. Thus we ought to have a micro-economic tax on gasoline equal to the pollution costs of burning a gallon of gas.

External or unaccounted costs are common. Many kinds and sources of pollution should be taxed. Alcohol and drug use (which should be legal) carry social costs which should be accounted through micro-economic taxes. These few taxes alone, if properly assessed, would bring in one to two hundred billion dollars of revenue (in addition to the halving the cost of law enforcement and the criminal justice system, if drugs were legalized). In the context of a balanced budget amendment, where the flat tax rate is set so that total tax revenue will target a (full employment) balanced budget given the level of government spending, revenues from micro-economic taxes automatically go toward lowering the flat income tax rate. That is, they are revenue neutral, so that their only effect is to increase efficiency on all fronts. Internalizing externalities increases efficiency by bringing private costs in line with true total costs, so that private decisions are socially optimal, and efficiency is also increased when the revenue from micro-economic taxes lowers the flat tax rate, reducing the disincentive to productivity.

The win win nature of micro-economic taxes cannot be taken as lisence to try to internalize all exernalities. Everything affects everything else in some way and to try to enforce accounting of every miniscule harm would be a disaster. The contravailing priority is liberty, the highest of all priorities, and this puts severe limits on what externalities can be internalized with taxes in a utilitarian framework. Whenever liberty is at stake all gray areas must go to liberty. Utilitarianism will only approve a particular tax to internalize externalities where it is absolutely certain, when all considerations are accounted, that the sum of progress in the discovery and attainment of value will increase. This will allow quite a few externalities to be internalized, because when they are appropriate, micro-economic taxes are a win win proposition. The high utilitarian hurdle will be cleared. At the same time, the gray area rule will rule out internalizing the great mass of minor externalities. Utilitarian property rights will include rights to use one's property in all kinds of ways that affect others. But particularly where coordination problems exist, making it likely that private agreement (liberty) will fail to reap available efficiencies, public measures are likely to pass the utilitarian public interest test. For a thorough discussion of the boundaries of economic liberty in a utilitarian framework see Richard Epstein's Principles for a Free Society: Reconciling Individual Liberty with the Common Good (Persus Books, Reading MA).

Despite the win win nature of appropriate micro-economic taxes, in the absence of a balanced budget amendment to render them revenue neutral, they have been and will remain politically infeasible. When government spending is out of control, people are loathe to let the government get its hands on any new source of tax revenue, because the presumption is that the money will just be spent, not returned through the lowering of other taxes. Once revenue neutrality is insured via a balanced budget amendment, we can start reaping the huge advantages that micro-economic taxes confer.

Micro-economic taxes would be set by the same tax accounting office that would be established to set the flat tax rate so as to target a balanced full employment budget. Unlike the definition of "necessities", which would have to be a matter of political agreement, the measurement of externalities is for the most part a well defined economic science, building on well understood concepts of cost benefit analysis. Boundaries would have to be drawn. At what point do externalities from an activity become a matter for criminal rather than tax law, and at what point do they become nobody's business but the actor's, not properly subject to governmentally imposed incentives of any kind? (The correct criteria here, I believe, is given by Mill's principle of liberty and where that principle can be seen to give rise to the concept of protected privacy. Interested readers are here referred to my article series on how to re-frame our system of liberty, and in particular to Article 6 of that series, which addresses privacy.) Once these questions are settled, they define the proper extent of tax incentives for particular activities.

Tax exemptions beyond the personal one are just a kind of micro-economic tax. As such, they should be set by the tax accounting office according to the neutral cost benefit principles that can be established for internalizing externalities. To circumvent these standards, by allowing some expenditures to be deducted from the income tax, is to introduce just the inefficiencies and opportunities for opportunities for corruption that agreeing on a flat tax and placing it in the Constitution is aimed at eliminating. If a tax deduction is given to each constituency's pet cause, that narrows the tax base and drives marginal tax rates up. Then there is the principle that the government ought to be neutral between different legitimate private activities and not favor one over another by offering "tax subsidies". Micro-economic taxes, derived according to neutral economic principles, satisfy this principle because they work to establish the proper concept of equal treatment, or not favoring one activity over another. All activities should face their true total costs.

That is the proper role of government: to establish that market ideal where prices accurately reflect costs. Beyond that, government should stay out until the boundary of illegitimate, or criminal, activity is reached. This is essential to our concept of liberty: that the government only step in to favor one private behavior over another when it has to. Up to that point, it just sets the level playing field where prices accurately reflect costs. In contrast to micro-economic taxes that internalize externalities, tax exemptions for pet activities, distort true costs and get government into the business of favoring one private activity over another in violation of the principle of liberty. Thus the proper place to incorporate tax incentives is in micro-economic taxes that internalize externalities on neutral economic principles, and no income tax exemptions beyond the personal exemption should be allowed.

Distributional concerns do not create any exceptions to this principle, but, where legitimate are to be captured by the personal exemption. Egalitarians, for instance, have argued against gas and other micro-economic taxes on the grounds that they fall disproportionately on the poor, but this concern does not hold up in the context of the system discussed here. The income and inheritance taxes (to be discussed next) are already optimally designed to handle distributional concerns on the premise that prices accurately reflect social costs. All micro-economic taxes do is fulfill this premise. In today's world, gasoline is certainly a necessity for most people and the cost of gasoline would be included in the setting of the personal deduction amount (whether it was set once and for all, in real terms, or whether it was set according to formulas for defining need) and this cost should obviously include any micro-economic taxes in the price of gasoline. Micro-economic taxes, far from being at odds with distributional concerns, are a part of the system for properly accounting those concerns.

Inheritance taxes

The long range distributional goal -- a fortune for everyone -- is best served by promoting efficiency goals, allowing the economy to function as an engine of income and wealth creation, but the long range distributional goal can also be targeted directly by inheritances taxes. At first glance, inheritances taxes look like double taxation, taxing monies that were already taxed when they are earned. But in the context of a balanced budget amendment with a mandated tax scheme, an inheritance tax is properly thought of as achieving a distribution of labor with the income tax. Without the inheritance tax, the income tax would have to be higher to achieve the mandated revenue target (equal to spending). It is not double taxation because the amount taken at death reduces the amount taken when the money is first earned. What we get is two episodes of partial taxation, adding up to the needed single full amount of taxation, and there turn out to be a good distributional reasons to hold off some taxation until inheritance. First, it allows the goal of a fortune for everyone to be promoted directly by allowing individuals to inherit up to a fortune, or "a living", tax free. "A living" today would probably be something like a half or three quarters of a million dollars (enough principal to afford a modest life, free from need, on the interest). Second, an inheritance tax can be used to encourage the dispersal of larger fortunes so that we move towards "a fortune for everybody" instead of huge fortunes for a few.

This is easily accomplished by Following Mill's idea of offering the exemption, not on the size of the estate that is left, but on the amounts received by individuals. Each person could receive up to a half million dollars of inheritance tax free over his life. Wealthy people who hate the idea of leaving anything to Uncle Sam could leave their money to many individuals. Once an individual's inheritance from all sources has totalled a half a million dollars (in constant dollars), a high marginal tax, say thirty or forty cents on the dollar (certainly no more than fifty cents), would be imposed on all additional inheritance. This money would go to reducing the flat rate on the income tax, which would in turn further promote income and wealth generation. The idea is to encourage people to make money, then let them pass it along tax free to the extent that their doing so promotes the distributional goal of a fortune for everyone. Inheritance beyond this goes partly to the distributional goal of an even bigger fortune for everyone, and partly to promoting basic wealth accumulation by reducing income tax rates.

Eliminate all business taxes

A basic principle of tax efficiency and neutrality is that all forms of income should be taxed the same. Breaking taxation into business taxation and income taxation greatly complicates attainment of this goal. Further, there is no reason to have a division of labor between business taxes and income taxes, as there is between income and inheritance taxes. Both target current earnings. Business taxes intercept those earnings before they leave businesses and get sent to households. Income taxes intercept them when they arrive at households. The different taxes are doing the same thing, just at different places, unlike inheritance and income taxes which are able to accomplish different things. Since dealing with taxes is an enormous burden, we should obviously eliminate any arbitrary breaking up of taxation into multiple stages, especially when that breaking up of taxation complicates the job of taxing all forms of income equally.

Consider capital gains taxes. Under our current system, capital gains are subject to business taxes in a full but round about way. Stock prices, for example, reflect the capitalized price of a company's expected future profits, because a share of stock gives the owner a claim on a share of all future profits, or more precisely, all future after tax profits. When business taxes go up, stock prices go down by the amount that the government takes in taxes, reducing capital gains by that amount. Thus when stock is sold, the capital gain already in effect has business taxes taken out. If these capital gains are then taxed at the household at the same rate that labor income is, they will in sum be getting taxed more. To try to keep the rate of taxation roughly the same for different kinds of income, our current tax code allows a partial deduction for capital gains, but it bears no accurate relation to the size of business taxes and is subject to constant political haggling. Of course once the haggling starts, principle has already been obliterated from consciousness. For instance, capital gains obviously must be measured in real terms. An increase in asset price equal to the rate of inflation means no gain in asset value for the owner, yet our current system measures capital gains in nominal terms. This obvious departure from simple honesty about when there is or is not a capital gain demonstrates how easily and completely the tax system stops having anything to do with principle, once tax law devolves into haggling over favors.

The case with dividend income is even worse. Dividends are paid out of after tax profits. That is, they have already been fully taxed at the business end, but then under our current system they are also fully taxed when received by households as income. There is not even an attempt to achieve equal rates of taxation on different forms of income, but simple bald double taxation. The effect is to give stockholders an incentive to prefer companies that do not pay out dividends but hold on to profits and use them to grow, so that profits can be taken in the form of capital gains instead of dividends. We get an artificial impetus for firms to become large, with no regard to whether this is socially desirable or not, all so that we can violate the basic principle that government should treat different forms of income the same.

Instead of breaking income taxes into two parts and hoping to reconcile them at the same time as we have multiplied the opportunities for corruption, it makes much more sense to tax all income when it is recieved by households and eliminate business taxes entirely. Capital gains would automatically be taxed at exactly the same rate as all other forms of income because they would all just get lumped together. We would have a fairer and more efficient system, and at the same time eliminate half of the apparatus of tax collection.

Hall and Rabushka did not choose this route. Instead they chose to minimize the problem of reconciling income and business taxes by applying the same flat tax rate to disbursements at the business end as is applied to income received by households. This eliminates the need to impose a partial taxes at one end to make up for partial taxation at the other. Each bit of income is taxed either when it is paid out by a business or when received by a household, then so long as no bit of income is taxed at both ends, all forms of income will be taxed the same. With this provision, Hall and Rabushka advocate business taxes because they see businesses as a place where book-keeping is already centralized and taxes can be levied while imposing relatively little burden. I think this is a mistake. Even though businesses are well placed to take out taxes, there is really no advantage to having them take out taxes, and there are several huge drawbacks.

One is just missing the chance to eliminate half of the tax levying apparatus and all the opportunity for mischief that it affords, both in the writing of the tax law and in compliance. Second, business taxes complicate the process of evenhandedly accounting for personal exemptions. Since there is no way to allow for personal exemptions in the business tax itself, the income tax must take account of how much tax was paid at the business on dividend and other investment income received, at least for those whose earned income was not greater than the personal deduction amount. Consider a family that is due $25,000 in personal exemptions and earns $25,000, but all in the form of dividends, taxed at the business end. If the flat tax rate is 20% they will have paid $5,000 in taxes that will have to be reimbursed if they are to receive their personal exemptions. This complication is eliminated if all income is taxed only when it is received by households.

The most important reason to eliminate business taxes may be one of political perception. As noted, when taxes are levied at the business end, taxation of capital gains is accomplished in a round about fashion such that they are never taxed explicitly. Rather, the capital gain on the sale of business holdings is implicitly taxed in that (as the capitalized value of future income) the sale price is lower by whatever percentage of business income is taxed away. It looks like capital gains are not being taxed because the actual capital gain must be exempted from taxation to account for the fact that it is already taxed implicitly. This is a tremendous political liability. It is a great mistake to ask the electorate to believe that a system of taxation is fair when only an economist is able to understand that it is fair. This political mistake already cost Forbes his run at the presidency. His proposal to eliminate capital gains from taxation was seen (even by myself until I looked further into the details) as a thinly rationalized attempt to exempt capital income and only tax labor, on the grounds that government should favor investment over consumption, by exempting income from investment. (This confusion was spawned by Hall and Rabushka, who described their tax as a consumption tax and defended it for favoring investment over consumption when in fact it does not do this at all.)

Eliminating businesses taxes eliminates all this confusion. Capital gains taxes would no longer be implicit but would be levied explicitly at the household. Capital gains would just be lumped together with all other sources of income and all would be taxed at the single flat tax rate after the personal exemption amount of income was reached. This might seem to raise another perception problem: the idea of "business" not be subject to taxation, which is instead shifted entirely onto "people". But I think it is much easier for voters to grasp the idea that all business income goes to people, where it can be taxed when it is received as personal income, than it is for them to grasp the subtleties of implicit taxation.

The perception of businesses as entities in themselves, rather than just producers of income for households, is correct and does raise problems. Firms in general are places where consumption as well as investment takes place. To properly measure payment to persons, we would want to monitor firms so that we can distinguish these kinds of expenditures. It would seem that on this front, eliminating business taxes is a step in the wrong direction. We would be moving back from scrutiny. But in actuality, getting rid of business taxes actually lessens this concern about consumption expenditures at work. When business taxes are in place, the government participates in every tax write-off. If a meal can be written off as a business meal, the government in effect pays a percentage of the tab equal to the rate of business taxation. This reduces the amount that the stockholders have to pay for consumption at work and therefore increases the level of it that they are willing to tolerate. Getting rid of business taxes makes the full cost of consumption at work fall on owners, who then have full incentive to eliminate it where it is not an efficient form of remuneration.

The one place where this does not hold is in the case of proprietorships and other privately owned businesses where, where the people who would be doing the consumption at work are the owners themselves. They will have an incentive to try to classify their personal consumption expenditures as business expenditures, so that the amount of net income they can say they received from the business will be lower by that amount, reducing the amount of income they have to pay taxes on. In the case of proprietorships, it will be necessary to retain "business taxes" and get into the nuts and bolts of what is to count as a business expense and what as a personal expense (to be accounted as income paid by the business to the household), exactly as our current tax code attempts to do. No matter what system of taxation is settled on, this complication must be entered into. As far as I know, this aspect of our current system is reasonably sound and could be taken as a starting point for a constitutionally imposed set of tax rules for proprietorships. The alternative, as with the definition of "need", would be to articulate principles for defining what is and is not a business expense and leave it up to the tax rate setting authority to locate the lines that those principles draw.

Eliminate all payroll taxes

Another key element of the optimal tax structure is that all payroll taxes should be eliminated. No one should be forced to be the tax collector for anyone else. (It is amazing, in all the flurry of Clinton appointee "nanny-gates", that no one questions why nanny's shouldn't be doing their own with-holding and filing their own taxes, just like every other person who works for clients instead of for a company.) Getting rid of all payroll taxes would have the salutary side effect of wiping out the current payroll tax incentive for firms to offer pay in the form of health insurance, which keeps employees from deciding for themselves what health coverages are worth buying, which keeps the market for medical services from working. [Button here to health care article.]

Eliminating all payroll taxes would also require eliminating the social security tax. Luckily, there would be no need to relocate this tax elsewhere, since having a separate social security tax only perpetuates the illusion that people are paying into a funded retirement plan. Social security is in fact a safety net program, not a disbursement of savings, and it is time to make it explicitly a safety net program, doling out aid strictly according to need. The present business of doling out welfare to every member of the wealthiest age cohort in American society is an unconscionable disgrace.

Supplemental "taxes" to repay debts of aid

Which brings up the last major component of an ideal tax scheme: a supplemental tax by which individuals who receive safety net help, or other government largess, would repay with interest the dollar cost of that help according to an ability to pay formula for as long as they live. This is the only way to offer a broad safety net without destroying incentives for responsible behavior. By making the repayment scheme part of the tax system, repayment requirements could be tailored to ability to pay, as indicated by income statements. Repayment requirements could be as generous as We the People decide, but it is absolutely necessary to keep the books straight. Welfare recipients (including recipients of upper class welfare, like social security) are not owed something by society. They owe, and they must be expected to pay back, to the extent that they are able, both during their lives and in probate.

To see the best way to structure repayment for aid, it is necessary to go back momentarily to the problem of how to evenhandedly allow for personal deductions. To be really evenhanded, it will be necessary to allow carry forwards of unused personal deductions. Then two people who earn the same amount over two years will pay the same amount of taxes, when without carry forwards, a person who has no income one year and a large income the next only gets half the personal deduction of a person who receives his income spread evenly over the two years.

This system of carry forwards provides a structure that can be used to set up a system of repayment for aid that has been billed to the account of the recipients (a key part of the scheme of spending controls, and the only way to set up safety nets without destroying incentives for responsible behavior). Interestingly, carry forwards offer a way to give a break to poor people without violating the principles of properly accounting what is owed to whom. People receiving aid will typically be earning substantially less than the personal deduction amount. (Aid that was billed to account would not be counted as income, and neither would private borrowing, because it is accompanied by an equivalent liability and hence does not constitute an inflow. Gifts would be subject to taxation, but only after total gifts to an individual recipient add up to more than the inheritance exemption amount). Allowing the shortfall of income from the allowed deduction amount to carry forward would reduce the taxes that these borrowers would have to pay in the future, when they get on their feet and start earning more money (more than the personal deduction) if they ever do. It is obvious that billing aid to account and charging market interest on that debt is a strict discipline. Here we see that a full and proper accounting of who is owed is generous as well as demanding.

The structure of carry forwards could further be used to coordinate schedules of repayment for aid billed to account. After those with carry forwards get above the standard deduction amount of income, they could nevertheless be required to immediately start paying to the government a share of all income above the standard deduction. The difference is, these payments would not be taxes,, but would be repayment on the aid that was billed to account. The personal deduction in this case provides a handy way to gauge the "ability to pay" criterion of the repayment scheme, because this is just what the concept of the personal deduction is intended to capture: the cost of necessities. Notice, however, that the rate at which people who owe debts of aid would be required to pay, once they earned more than the exemption amount of income, would have to be higher than the flat tax rate, because in addition to becoming taxpayers, they would have to start repayment on their debt of aid. Their overall rate of repayment should be at least half again as much as the flat tax rate, in my estimation, and not more than twice as much, and this should start as soon as those with aid to repay start making more than the personal deduction, whether they have personal deduction carry forwards or not. All the carry forwards do is determine whether the payments are tax payments or payments on one's personal account. (Note that since borrowing is not counted as income, it does not affect the size of the carry forward, so no distortionary incentives to borrow are created.)

In sum

Tax systems by nature distort incentives and must settle for second best solutions, getting as close to undistorted ideals as possible. An exception is microeconomic taxes which can kill two birds with one stone here, removing distortions while raising money to pay for public goods. Answering claims of need by billling aid to account also minimizes both government expenditure (the amount of the subsidy is only a fraction of the liquidity provided) and distortion to incentives. When the remaining necessary and proper govenment expenditures are financed by a broad based flat tax, the marginal tax rates necessary to cover expenditures should be low enough to cause little distortion. This necessary amount of taxation can be reduced still further by inheritance taxes designed to target long run distributional goals. All of these elements are clearly optimizing and bring us very close to a first best situation.

While it is clear what the main outlines of an ideal system must me, there remain several points of execution where the ideal is not determinate. It is not clear just how the line should be drawn to separate business from personal expenses in the case of proprietorships. Similarly with figuring out what micro-economic taxes should be applied. Theoretically, optimal boundaries in these areas are partly determined by incentives to comply, which bring up classic second best problems that can be difficult to solve. Plus they can be "messy" in the sense of trying to separate elements that are intimately entwined. Sometimes business is personal and vice versa. But these problems are few, and reasonable solutions seem to be fairly easy to arrive at. It is not clear exactly how to define what is a necessity, for purposes of deducing the appropriate personal deduction. But at the margin (after clear necessities have been accounted) this is essentially a political problem, about how charitable we want to be. If we can articulate a generally accepted concept of necessity we can decide it once and for all, or leave it to the legislature. Theoretically necessity is not that hard a concept to solve, and politically it certainly would be resolved. A problem that is both "messy" and "second best" is figuring out what the rate of repayment of aid should be, out of income above the personal deduction amount. To high and the ability to give aid is undercut. Too low and incentives are destroyed. Between theory and politics the problem is not that difficult. The important thing with all these problems of definition and second best considerations is that none of them seem to be particularly difficult to solve in the sense of having fairly narrow ball-park of reasonable answers. In all, the ideal tax scheme seems to be almost completely specifiable, certainly in its structure, and within narrow margins in its details.

Thus we can with confidence take this most propitious step of placing an ideal tax code in the Constitution and stripping Congress of its tax power. With Congress out of the tax writing business henceforth, it would no longer be possible for representatives to log-roll votes that shift the burden of taxation off of favored parties and onto everyone else. (A few questions about how generous we want to be in answering claims of need could be left in the hands of Congress if we prefer, but this would offer no opportunity for corruption. If representatives did try to favor monied interests under the guise of need, those monied interests would be required to pay back in full, since aid would all be billed to account.) The reason we have such a corrupt Congress now is because Congress is perpetually faced with such huge opportunities for corruption. With half of its grossest opportunities for corruption gone, we might even see an upstanding Congress in our lifetimes. We just need to do it, and in one stroke get both a much more efficient and a much fairer economy and safety net system. Good government is ours for the having.

Next article in Utilitarianism volume of Moral Science: Institutions of Liberty Require Immigration Crackdown

Top of Page