JUSTIFYING TAXATION

Abstract

Taxation is more than one thing. Taxes can be levied in various ways on various things, with varying effects on a culture and an economy, and raising different challenges of justification.

I. Introduction

A common-sense first thought about justifying taxation: public spending is necessary to sustain a community. Taxation is necessary in turn to sustain public spending. To justify taxation thus conceived is to justify it as the least of evils rather than as good in itself. It is a dirty job that must be done.

In that case, whether a given regime of taxation is justified is largely an empirical matter, not a matter of armchair philosophy. To be justified, taxation must in fact sustain crucial community infrastructure. Further, there must be no cleaner way to sustain crucial community infrastructure: no more efficient way, no more voluntary way, no more unobtrusive way, or something like that.

If a country needs a given flow of revenue to finance a given government expenditure, then there is a question about how to raise the revenue. Even after it has been decided that taxation is the least of evils, questions remain about what form the taxation should take: What kind of event should trigger taxation? Should it be triggered by the occasion of earning income? The occasion of receiving passive income from investments? Making a retail purchase? Owning property? Should an employer be taxed for carrying a payroll? For meeting some other business expense? Should taxation be based on wealth, on income, on sale or transfer of property? In what proportions? For what ends?

Part of the issue is whether some instruments are less likely than others to be weaponized. If there is discretion in the use of those instruments, then access to them will occasion prolonged struggles between rival factions. A realist has to wonder: Do we justify creating a power to tax in terms of what we can imagine that power would ideally be used for? Or in terms of what it predictably will be used for? ¹ Part of that question concerns whether there is anything to limit the scope of the power we are creating. Can the power to raise revenue for a given purpose be defined, settled, and serve as a durable basis for making plans? Or must the limits of the power to tax inevitably be a political football?

The question of how to raise revenue of course turns on where revenue can be found. That is, the question comes around to who can pay, how much they can pay, and what it will take for tax collectors to collect. As Richard A. Epstein notes, patterns of taxation of wealth are always sensitive to the dating of when the wealth is realized as a taxable event and the form in which the wealth is received. For example, there is an argument against taxing new wealth received in exchange for old wealth when new wealth is not in the form of cash or marketable securities, as in many corporation reorganizations; in that case, there is reason to defer marking the gain as a taxable event until the parties cash out of the transaction. Jeffrey Paul offers a history of the income tax and of the question of whether taxes should be justified by reference to the benefits taxpayers have received, or alternatively, in proportion to taxpayers’ ability to pay. Paul also notes that a felt imperative to treat taxation as a tool for redistributing wealth is a distinctly different justification, and not always as widely accepted as it is today.

II. Proportionality and Justice

It often makes perfect sense to say every consumer should pay the same price per unit consumed. If you were to walk onto a used car lot and immediately be confronted by a salesperson saying that for “people like you” the actual price is double what it says on the sticker, you would feel violated. But even when we say everyone should pay the same, we do not mean everyone should pay the same regardless of benefit received. Rather, we mean roughly that, for example, consumers of bread should pay the same per loaf consumed.

But there are complications. There are club goods. There are public goods. There are common carriers and public utilities for whom the cost of service for given consumer groups varies widely. Scarcity and crowding may have to be managed, but there are many goods whose consumption we do not want to monitor too closely. We want some goods to be more or less freely available. Someone has to pay, of course, but sometimes we do not want the paying to be at the point of consumption. Alternatively, we may not want limited ability to pay to stop people from joining in, so we divorce the paying from the consuming. Instead, we sometimes charge people for earning an income. We charge people for being involved in producing the product rather than for being involved in consuming it. We then in effect use the proceeds to subsidize consumption. Or to subsidize investment. Note that it is not only governments that subsidize consumption at the margin. Amusement parks have learned that it pays to charge patrons a lump-sum entrance fee high enough that the attractions to which entrants gain access can subsequently be made available to entrants at no further charge. It makes for a different experience and represents a choice that needs to be made in public and private sectors alike.

When it comes to paying taxes, people talk a lot about what would be a fair share. It is common for people to say the rich do not pay their fair share, or do not pay taxes at all. One gets the sense that complaining about how little the rich pay is like complaining about the weather. It would be rude to treat such claims as factual claims needing to be supported by real data. Thus, one cannot listen to the news without hearing a story about how the one percent should start paying their fair share. If those who say this are asked what a fair share is, they typically have no clue what the rich are paying now, so a fair share is simply a way of saying “more.” When I ask people to quantify “fair,” a typical answer is that everyone should pay at least 10 percent of their income. The Congressional Budget Office sums up its data by saying that, “In general, higher-income households paid a higher average federal tax rate than lower-income households. Average federal tax rates fell between 1979 and 2017 across the income distribution, with the sharpest decline in the lowest quintile.” ² The richest one percent of American households now pay roughly 38 percent of federal income taxes paid. ³ This number is never mentioned by the popular press.

Most people imagine something radically different. Upon being told how much the rich already pay, people sometimes have a visceral reaction. They do not want to believe it and deem it vicious even to hint at doubts about whether the one percent should pay far more than whatever they are paying now. Congressional Budget Office data, though, suggest that the one percent are not freeloading; rather, if we could induce the one percent to move to Monaco, tax receipts would crash. Statistical inequality as we measure it in the United States today would decrease, but the 99 percent would be worse off.

In any case, we have to choose whether to tax production or consumption. We can suppose Jeff Bezos devised such an extraordinarily efficient way of reducing transaction costs that his firm, Amazon, generated a cooperative surplus worth trillions of dollars. Moreover, Bezos managed to capture upwards of two hundred billion dollars of that surplus for himself. What would we have to ignore to make it appear as if this development increased inequality? Transacting with Amazon is not what gives people in the bottom income quintile a sense of being left behind. On the contrary, when people order packages from Amazon, they do so in expectation that their experience will be like previous experiences they have had with Amazon, when receiving packages made them feel enriched. If they feel that Bezos has impoverished them, they get that feeling from reading what the media say about Bezos, not from dealing with him.

Meanwhile, because of people like Jeff Bezos, ordinary shoppers did not need to travel to the marketplace to do their shopping during the COVID-19 pandemic. Real data is scarce and controversial, but being able to order online over the past few years may have saved thousands of lives. Moreover, even in a pandemic, shoppers did not need a lot of money to induce a vast network of people to act as their personal valets, assembling and delivering packages to their door. This new system served as an equalizer in ways that poor consumers care about, regardless of how rich people (that is, the richest 5 percent, or whatever number is large enough to include most academics) choose to express the deprivation they feel when they see Jeff Bezos figuring out how to market a service that customers actually want.

To Jason Brennan, the fable of the ant and the grasshopper, familiar though it may be, has an underexplored nuance. Political philosophers talk about the ant having a duty to pay more, but now suppose that if ants are social animals who work in cooperative social worlds, then ants have obligations. Yes, but to whom? What if we said these obligations are not to other ants, but to grasshoppers. People who say, “you didn’t build that bridge” are not insinuating that the identifiable workers who did in fact build that bridge never got paid. They aren’t saying cooperators owe it to other cooperators to pay for services they are exchanging. Rather, they are saying cooperators’ tax dollars were not used to help build bridges from which they derive immense benefits. Or perhaps instead their idea is that bridges are what a previous generation built, suggesting an obligation to pay it forward by building the next generation’s bridges. That may be plausible, but it won’t imply that ants have obligations to grasshoppers.

Brennan’s work on ideal theory is important, and its implication is not nearly as prosaic as the familiar fable makes it sound. Brennan argues that there is a logic to ideal theory that can sort out which visions are worthy of aspiration and which are not. However, Brennan says, there is a kind of equality and a kind of reciprocity deep in the logic of utopian thought. Nick Cowen might agree. Cowen draws attention to concerns about day-to-day processes of taxation and to the ultimate distributive results and aims of taxation. He emphasizes a need to evaluate the ongoing process, not only the ultimate outcome. The upshot, as he sees it, is that a true egalitarian utopia should be as attractive to conservatives as it is to liberals, insofar as a truly egalitarian utopia is not so much an equality of consumers but rather an equality among producers, that is, an equality among contributors—one that gives each an equal risk-adjusted rate of return on their private investments. Among producers, equality is an achievement, not a given, and no one would have either a right or the ability to shirk if Cowen’s last constraint were binding, as it ideally would be. No one has a right to be a grasshopper—a right to be nothing more than another mouth to feed—and ideally, in a utopia, no one would purport to have such a right. Everyone has a duty to be part of a grand effort to extend that most estimable of all equalities—equality among workers who give back to their communities—as universally as possible. And ideally, no one would pretend otherwise.

Miranda Perry Fleischer notes that there is another target for taxation, namely inheritance, and that inheritance raises a new set of questions about proportionality and justice. Is there a need to impose a tax at death or in anticipation of death if the other taxes in the system are properly calculated? Will a “death tax” distort activities taken throughout life?

III. Creation Myths

Fabian Wendt ponders background questions about the nature and scope of property rights as a way of framing questions about justifying the taxation of property. He contrasts the conventionalist interpretation of property offered by Liam Murphy and Thomas Nagel with that found in John Locke. In keeping with the theme of Murphy and Nagel, Stephen Holmes and Cass Sunstein write, “No sharp line can be drawn between markets and government: the two entities have no existence detached from each other.” ⁴ Murphy and Nagel follow up by inferring that because “[t]here is no market without government and no government without taxes … [i]t is therefore logically impossible that people should have any kind of entitlement to all their pretax income.” ⁵ This appears to be a maximally provocative (and implausible) way of making a simple point that governments do not function well in the absence of a price mechanism, and markets do not function well except in the context of enforceable contracts. Provocation aside, the simple point is fair enough.

Murphy and Nagel write: “The state does not own its citizens, nor do they own each other collectively. But individual citizens don’t own anything except through laws that are enacted and enforced by the state.” ⁶ This would of course be a non sequitur to John Locke, who thought that the state is brought into being to protect entitlements created by labor, not to be the source of those entitlements in the first instance (unless we interpret entitlements in an anachronistically positivist sense). To Locke, the human condition is sufficiently lawlike to make the consequences of legislating one way rather than another somewhat predictable. Locke also believed in something that we think of today as self-ownership but which, in any case, is one of the pillars of Western liberalism. Liberalism is, at its core, a view that individuals have a right to say “no” and that the proper holder of this right is each individual person. When somone says “no,” the implication is not that we should put great weight on her refusal. The actual implication is: “no” means “no.” End of story. Locke would say that this much is built into the human condition, and the bare fact of capturing the reins of government (as the Taliban has done in Afghanistan, for example) does not entitle anyone to decide otherwise.

In any case, to say government decides what the law is does not imply that government decides what morality is. There comes a point when a government acknowledges that “no” means “no,” but the acknowledgment is not what first makes it wrong to ignore a person’s “no!” The acknowledgment is merely an acknowledgment. If a government were to conceive of itself as being at liberty to go either way on the question of whether citizens have a right to live their own lives in peace, that would not be liberalism. Some societies have a transcendent virtue of making people willing and able to accept each other as neighbors, each with dreams of his or her own. Justice in those societies is not guaranteed, but the point is, those are the societies that have a chance.

Accordingly, while property is in some ways conventional, it is not arbitrary. Or more precisely, to call property conventional is not to call it random. There are correct and incorrect answers to questions about where to draw the line. There are morally nonarbitrary questions about which ways of managing traffic actually work to help people get where they want to go, with reasonable hope of avoiding collision along the way, and with reasonable hope of being of service to other people—fellow travelers—along the way.

These rights include the exclusive possession of oneself and one’s talents; the ability to acquire property in land, animals, and chattels, by initial occupation; and the ability to gain access to public waters and public roads along with all others. The formalities that create and enforce these entitlements will vary from place to place. For traffic to move, people have to drive only on one side of the road. Whether to drive on the right rather than the left may seem arbitrary in abstraction from convention and from legal history, but once a decision is made, the further decision to respect an emerging convention of driving on the right is not arbitrary. And conventions for transferring goods, whether by deed or delivery, are less arbitrary than that. (How far can a theorist go with the idea that there are no conventions without government? Do Holmes and Sunstein think languages did not exist until created by government?) There are compelling (even if not universally decisive) reasons to treat the person responsible for growing a given crop as the crop’s owner rather than, say, tying ownership to being the next person to introduce crop disease. Letting those who sow reap fosters efforts to create a needed supply of foodstuffs. Letting those who don’t sow reap is to slide in the direction of ending agricultural production.

Ghen v. Rich ⁷ is illustrative. In 1880, in Ghen v. Rich, the court learns that Ghen fired a bomb lance into a fin-back whale. The dead whale sank, washed up on the coast of Massachusetts, and was found by Ellis. Ellis sold the carcass to Rich who extracted oil from the blubber. Hearing rumors of this, Ghen filed a claim for the value of the oil extracted. Ghen’s case rested on the whaling community’s custom of treating the person who first harpoons a whale as establishing possession and thus ownership. Judge Nelson of the Massachusetts District Court ruled in favor of Ghen, acknowledging that whaling ports were close-knit communities that had worked out simple, transparent norms regarding how to establish possession, and thus presumptive ownership, and that these simple customs helped whalers to be of service to the larger community. Also by custom, one who found a whale washed ashore was paid a fee to induce him to report the location of the landed whale to its owner. Both parties thus gained in stable fashion. Customs for dividing the spoils of whales can differ by location, but in all cases the participant in the capture of the whale receives a return proportionate to labor, or the entire project will founder. Specifically, where commercially valuable whales are slow and docile, the mark of possession is “fast fish, loose fish,” meaning possession is established by fastening the whale to the boat. ⁸ In other communities where the most commercially valuable whales are too dangerous to consider attaching to one’s boat, the operative rule is the less stringent “iron holds the whale,” meaning that fatally harpooning a whale suffices to establish possession. This rule applied to the fin-back, which made the first possessor Ghen, not Ellis.

Carol Rose ⁹ identifies two background principles. First, establishing possession should involve doing something unambiguous to notify the world of what one is claiming. Second, establishing possession should involve labor that adds to the value of what one is claiming. As Judge Nelson saw, he was not judging in a blank-slate state of nature. On the contrary, when the case came before his court, there already were property conventions, and reasons for those conventions. Nelson added something to the body of judge-made common law of property, but his job was not to arbitrarily make up something to fit his personal vision of justice. Instead, his job was to resolve conflict by coming up with a verdict that the litigants should have seen coming and that everyone should be able to live with as a framework for knowing what to expect from each other going forward.

Holmes and Sunstein refer to “the obvious truth that rights depend on government.” ¹⁰ But they know that this is not obvious. To quote the paragraph from the beginning: “The Declaration of Independence states that ‘to secure these rights, Governments are established among men.’ To the obvious truth that rights depend on government must be added a logical corollary, one rich with implications: rights cost money.” ¹¹

Notice that the Declaration of Independence, far from stating that this is obvious, denies that rights depend on government. It says that part of what can justify establishing a government is its efficacy in securing rights; the explicitly stated point is to secure rights, not create them. Laws obviously are creations in some sense, but there remain truths about what should be recognized as a legal right and what should not. There was mutual understanding about what people ought to be able to expect from each other, and a hope that codifying laws, and eventually creating governments, would help to secure such expectations. (This is not to deny, of course, that expectations change, and one factor driving change is a sense that government itself has developed a capacity to drive expectations in a direction that a particular faction favors.)

People have a record of prospering when they treat each other as self-owners (which is to say, as having a right to say “no”) and accordingly treat the coercing of other selves—disposing of their life, liberty, or property without their consent—as paradigms of what needs justifying. That’s the relatively simple truth about why normal people see taxation as justifiable in principle, while at the same time seeing taxation as a paradigm of what needs justifying.

IV. When Plans Meet Reality

People talk as if a tax is “regressive” just in case it leaves people with unequal after-tax income. But that is incorrect. “Regressive” is not a way of expressing a feeling. It is not a way of saying “boo” to inequality. A tax on income is regressive, as a matter of technical fact, only in cases where those who earn less pay a higher percentage of their income. A tax on income is flat only in cases where everyone pays the same percentage. A tax on income is progressive only in cases where those with higher earnings pay a higher percentage. This is what the words mean when used correctly. If one’s real complaint is not that taxes are regressive but that tax brackets are less progressive than they should be, then being honest about the nature of that complaint is basic.

If we think flat taxes are less fair than progressive taxes would be, that would be good to know. If we think a head tax is regressive, while thinking that a flat tax is by definition neither progressive nor regressive, that would be good to know. If we think income taxes are not regressive but our current way of levying Social Security tax is, because we cap how much income is subject to it, then that is different again.

Shruti Rajagopalan comments on the project of theorizing about what would be a fairer tax in theory, if we were building a rule of law from the ground up, then treating that thought experiment as weighing in favor of imposing a fair tax structure on our present situation:

So, in addition to many federal and state taxes becoming a single [Goods and Services tax (GST)], the effect of the GST was to turn India into one large free trade zone, with a single tax rate. It was marketed to citizens, businesses, investors, and multilaterals as the “Good and Simple Tax.” However, by the time the reform was complete, it was neither good nor simple. Pursuing several different goals, all masked as increasing or furthering equity, the GST has furthered interest groups seeking rents, tax concessions, and privileges. ¹²

A coalition of power brokers had to be assembled. In practice, there was no other way for a law to pass. Therein lies a bracing dose of realism for proposals to replace our current system with a system that would in theory be fairer, less regressive, or whatever the theoretical aim may be.

Rajagopalan thus warns us, on the basis of real experience, of the gap between what a tax officially does in theory versus how it actually works once complexities and exemptions get inserted into the bill’s footnotes as the price of assembling a coalition of legislators that can pass changes into law. Ostensibly progressive legislators will go along with it in order to get elected. To them, that won’t be corruption; it will be a straightforward concession to political reality.

Wouldn’t fairness demand that the rich pay more? How else to get the rich to pay more other than by simply demanding more? Under a simple flat tax, the rich would in fact pay more, by virtue of paying the same percentage of a higher income. Few taxation schemes are this simple, of course, because the politically connected rich always find a way to mobilize political support for complexity. Complexity works to their advantage because they are better positioned to game the system. They have the lawyers.

Rajagopalan’s ultimate point is that complexity ends up being regressive. It will be marketed as making the system more progressive. When Warren Buffett complains that he pays less than his secretary, he is not making it up; yet, he then lobbies for the very complexity that enables him to pay less, even as he champions nominal progressivity.

Charles Delmotte likewise worries that rent-seeking “is endogenous to democratic capitalism, as it creates institutions and incentives that make collusion between political power and economic power more profitable.” ¹³ Cooperation for mutual advantage is an ideal for the kind of social and political animals we are. Trade is the paradigm of this ideal—at least ideally. But where rent-seeking becomes possible—that is, where the commodity that traders are trading is access to “other people’s money”—then the situation departs from the ideal. In practice, there are conditions under which trade approximates its theoretical promise, but the opportunity to buy and sell the game’s referees is a disturbing and by no means merely theoretical departure from the ideal. Allison Christians extends such concerns to the realm of international trade, and to what turn out to be strikingly difficult questions about who should tax multinational corporations. Linda D. Jellum likewise sees hard interpretive questions regarding the authority to resolve ambiguities in tax statutes, even regarding something as seemingly simple as deciding how to date the delivery of a document to the IRS. Should we be looking to the courts or to the Treasury for resolution? Jellum argues that the authority that regulatory agencies currently have to reject judicial interpretations of ambiguous statutes is too broad and does not work.

V. Is Efficiency a Bias?

Jonathan H. Choi considers the idea that tax policy should efficiently raise revenue with which to finance activities of state. But, he notes, from the perspective of people who think that the point of tax policy should be to level income classes, the idea of limiting ourselves to tax policies that are efficient could seem like an ideological bias. Sometimes, this objection is supplemented by an argument that that concept of efficiency is incoherent in any case, because any maximization problem is defined with respect to a set of constraints. The idea is that since policies themselves are part of the set of constraints, there is no independent set of constraints with which to judge whether policies themselves are efficient. Choi sorts through the theory and practice of these issues. But suffice it to say, whatever else we can say by way of comparing alternative sets of problem-defining constraints, the fact remains that if the players don’t know the rules, or if all they know is that they should hire lobbyists to help them strategize the rewriting of the rules, then the ceiling of human possibility will be lower to that extent.