January 06, 2005

How not to complain about taxes (1)

Anderson on Political Economy, Anderson on Taxes, Elizabeth Anderson: January 6, 2005

"Governments cannot be supported without great charge, and it is fit every one who enjoys his share of the protection, should pay out of his estate his proportion for the maintenance of it."

That's John Locke, the great defender of private property, writing (Second Treatise of Government, ch.  XI, par. 140).  It pays for defenders of private property to listen to Locke, so as to avoid silly complaints about taxation.  Here's one common one I hear:  that government, in taxing my property, is taking away what is really mine.  This complaint is often conjoined with the accusation that liberals, in order to justify taxation, must believe that the government owns all property to begin with and by rights could confiscate it all.  Two points should put these fallacies to rest.

First, a technical point:  the fact that some property is mine does not entail that other people do not have rightful claims to some portion of it.  I am entitled to my salary; it's mine.  But my children have a rightful claim to support from my income.  In some states, such as California, I have a legal obligation to support my parents out of my income, if they cannot support themselves.  I have to pay my bills out of my income.  If I negligently injure someone, I am liable to pay them damages from my income.  The fact that this income is mine does not settle anything about who else might have legitimate claims to some portion of it, and on what grounds.  Note also that I did not have to give my personal consent for some of these others to have a claim on it.

So far, I've just been talking about property as a legal institution.  But perhaps the complaint I am criticizing is talking about supposed "natural" property rights, following theorists such as Locke. So here's my second point:  unless one is a bomb-throwing anarchist, an advocate of natural property rights must concede the legitimacy and indeed necessity of a state, at least as an institution for collective protection and impartial adjudication of claims--the so-called "minimal state."  And such a state will have a legitimate claim on every member's property, to the extent necessary for everyone to pay their fair share for its maintenance, as Locke rightly insisted.  Even in a minimal state, the fact that my income is mine does not constitute an argument against the taxation necessary to support the state.

In fact, Locke himself went much further than this minimal claim. In the Lockean mythology loved by libertarians, it is supposed that individuals, upon joining a minimal state, retain full claim to all of their natural property rights, except to the small extent needed to support a minimal state.   The fallacy here is to suppose that, when people join together to form a state for the protection of their property, they are concerned only to protect their property from the encroachment of others.  According to Locke, however, individuals form a state not just for protection against violations of their negative liberties but for the preservation of their lives (which are part of their property):

"the first and fundamental natural law, which is to govern even the legislature itself, is the preservation of society, and (as far as will consist with the public good) of every person in it." (Locke, Second Treatise, ch. 11, par. 134)

Unless one could show, contrary to fact, that death rates under publicly funded health care systems are higher than under systems that leave people to pay for their health care with whatever resources are at their disposal, some kind of publicly funded health insurance entitlements are compatible with, and may even be required by, Locke's theory of natural property rights.  Moreover, Locke insists on our obligation to provide for the poor:

God hath not left one man so to the mercy of another, that he may starve him if he please: God the Lord and Father of all, has given no one of his children such a property in his peculiar portion of the things of this world, but that he has given his needy brother a right to the surplusage of his goods; so that it cannot justly be denied him, when his pressing wants call for it. . . .  As justice gives every man a title to the product of his honest industry, and the fair acquisitions of his ancestors descended to him; so charity gives every man a title to so much out of another's plenty, as will keep him from extreme want, where he has no means to subsist otherwise: and a man can no more justly make use of another's necessity to force him to become his vassal, by with-holding that relief God requires him to afford to the wants of his brother, than he that has more strength can seize upon a weaker, master him to his obedience, and with a dagger at his throat, offer his death or slavery. (First Treatise, ch. 4, par. 42, emphasis mine)

Locke's point is not just that some kind of entitlement-based welfare system is required by morality and built into the structure of natural property rights (the poor have a title to what they need).  It's also that, to prevent a free property system from degenerating into feudalism, constraints on freedom of contract are required.  Just as contracts into slavery are invalid, contracts into vassalage are.  People are not entitled to use their superior bargaining power to drive others to the wall, or into subjection. 

So, you can't get an argument against a welfare state from Locke's theory of natural property rights.  I won't pretend that Locke was as generous as modern welfare states; his preferred system of provision for the poor was in fact very harsh.  And, given the primitive state of medicine in his day, no one at the time imagined it would have done much good to universalize access to it.  But nothing in his system  prevents a more generous welfare state.

In fact, given the wide scope of the legislature to pass laws for the common good of society (Second Treatise, ch. 11, par. 135), it isn't even clear that the distribution of natural property before people joined a state provides a constraint on the legislative power.  Locke insists that any just government establish some system of private property or other.  But there is no indication that it must mirror the distribution of property people brought with them into civil society.  He never says that people in civil society have a right to the goods which were theirs in the state of nature.  Rather, he says "they have such a right to the goods, which by         the law of the community are their’s" (Second Treatise, ch. 11, par. 138, emphasis mine).

Does it follow that Locke, in accepting the legitimacy of taxation to promote the general welfare, including the establishment of welfare entitlements, really believes that the government owns everything and so could by rights dispose of all property arbitrarily?  Of course not.  He lays out the following constraints on legitimate taxation in ch. 11 of the Second Treatise:

1. It must be consistent with some system of private property or other (par. 138).
2. It cannot confiscate people's private property arbitrarily, but only in accordance with duly passed laws (par. 135-8).
3. The people must consent to these laws, not in the sense that they must obtain the personal consent of each individual, but in the sense that they have the consent of the majority of representatives in the legislature (taxation "must be with his own consent, i.e. the consent of the majority, giving it either by themselves, or their representatives chosen by them") (par. 140).
4. The laws must be for the common good of society, and in particular, promote the preservation of each member in it (par. 134-5).
5. The level of taxation cannot be so great as to reduce anyone to poverty or subjection ("It [the legislative power] . . . can never have a right to destroy, enslave, or designedly to impoverish the subjects" par. 135).

Although I'm no Lockean, I'm happy with these constraints, as I think all liberals are. (Personally, I would add another constraint, that requires the distribution of tax burdens to be fair.  Locke may also implicitly be insisting on fairness in the quote that opens this post.)

So please, stop the silly rhetoric that liberals suppose that the government owns everything already.  Stop the silly rhetoric that supposes that the fact that some property is mine offers any argument whatsoever against the legitimacy of taxing it.

I hasten to add that this still leaves plenty of room for reasonable dispute about proper levels of taxation.  For all I've said so far, it's fine to argue that current levels of government spending are excessive, so that the levels of taxation required to support those levels are unjustified.  It's fine to argue that the tax system we have unfairly distributes its burdens on the rich (I'll be posting later on that subject).  It's fine to argue that our tax system stupidly rigs incentives in unproductive ways.   It's even fine to argue that government welfare entitlements are illegitimate in principle, and hence that taxation to support them is unjust.  (For my point here is narrow:  merely that one can't get any support from Locke's theory of natural property rights, and hence not from the general idea of natural property rights, to argue this point.  I'll be posting later on why I reject theories of natural property rights.  My answer will surprise you.)  This post is simply a plea to focus on real arguments about taxation, not silly rhetoric.  But I wouldn't mind if you also learned a thing or two about Locke.


January 20, 2005

How Not to Complain Against Taxes (II): Against Natural Property Rights

Anderson on Political Economy, Anderson on Taxes, Elizabeth Anderson: January 20, 2005

In a previous post, I argued that the claim "it's mine" does not by itself constitute an argument against taxation.  Nothing follows about the legitimacy of taxation from the mere fact that something is one's own property, nor, as Locke's example shows, even from the fact that something is one's own by a natural property right.  Several comments and a trackback on that post supposed that I was arguing from the truth of Locke's theory of natural property to my conclusions.  This is odd, since I explicitly disavowed theories of natural property rights in that post.  My point was only that "it's mine" is no argument against taxes; it's at best a begged question.  This of course does not rule out anyone's offering an alternative comprehensive theory of natural property that entails the illegitimacy of taxation, or taxation beyond what the minimal state requires.

Now I'd like to take on the idea of such a comprehensive theory of natural property more directly.  In that earlier post, I promised to explain why I reject theories of natural property rights, dropping the teaser that my answer would surprise you.  Here I spill the beans:  I reject theories of natural property rights because they are incompatible with capitalism.  More precisely, they are incompatible with the forms of modern capitalism that have proven so successful in  expanding people's opportunities, prosperity, and the scope of cooperation--that is, the forms worth supporting.  Far from being the foundation of capitalism, natural property rights, construed as putting constraints on state action, are its bane.

By a theory of natural property rights, I mean a theory that (a) identifies first principles by which individuals may initially own or acquire property without the help of a state; (b) upholds principles of virtually unrestricted voluntary transfer (freedom of contract, gift, and inheritance); and (c) limits the state (if it is to exist at all) to enforcing these strict property rights and whatever obligations arise from unrestricted freedom of contract.  Crucially, these rights and obligations may not be abridged, limited, or revised by the state in order to produce various desired consequences.  Some natural property rights theorists allow exceptions to these principles at the margins (Nozick, for example, allows that property rights may be abridged to avoid catastrophe).  But by and large they see free market capitalism as a spontaneous self-sustaining product of systems of property whose logic lies outside state definitions and social engineering.  The great danger to capitalism, on this view, is state "intervention" into a market sphere that runs by its own natural laws.

I think this picture of capitalism is misguided.  The forms of capitalism that exemplify its greatest virtues rest on artificial, not natural property formations.  The state does not "interfere" in a "natural" capitalistic realm; rather, state action constitutes this realm as distinctively capitalistic.  Advanced capitalism requires a vast apparatus of socially engineered institutions to sustain itself.   To put some specificity on these claims, I'll argue as follows.  (1) Certain types of property rights and rules found in advanced capitalism have no sound basis in "natural" property rights but are nonetheless essential to advanced capitalism.  (2) Natural property claims do spontaneously arise independent of state action, but they are incapable of generating the distinctive form of property needed for capitalism--namely, capital.  State action is required to turn property into capital, and such action will inevitably, and rightly, abrogate these "natural" claims.  (3) A pure system of natural property rights with unrestricted freedom of contract contains inherent tendencies to revert to feudalism if the state does not limit freedom of contract by restricting property transfers from the desperate to the well-endowed.

(1) Advanced forms of capitalism depend on types of property that have no natural foundation. 

Consider, for example, the limited liability corporation.  In a natural property regime, groups of people contracting together can enjoy no more rights vis-a-vis third parties than what the sum of their individual property rights already gives them.  Since individuals do not enjoy any "natural" limitation on their liability, they can't naturally acquire such a limitation just by combining their property with others.  Limited liability is justified not because it could arise from a system of natural liberty, but because investment in firms that separate ownership and control will be retarded, and hence overall economic growth will be depressed, if investors don't enjoy it.  Limited liability does leave some rightful claimants uncompensated when firms go bust.  Capitalism makes up for this in part through generally higher growth (which socializes some of the benefits of this property form).  It also partly makes up for this through social insurance schemes that prevent some of the worst costs from being so concentrated as to produce hardship (which socializes some of the costs of this property form).  For example, the state bails out pension funds that bankrupt firms owe to their workers.

Intellectual property rights are also both indispensable to capitalism and deeply artificial.  They contain two features that, in combination, cannot be rationalized by a theory of natural property:  (a) they afford monopoly rights to inventors, and (b) they are temporary.  A theory of natural property could reject intellectual property altogether, on the ground that thought, once made public, becomes part of the commons, and no one coerces anyone else in using a valuable idea.  Or it could insist that inventors have an absolute and permanent right to their ideas.  (This would, for example, grant to the heirs of the inventor of the alphabet the permanent right to determine who is permitted to use it, and how much they must pay for the privilege.)  But it is difficult to envision any theory of natural property that could both acknowledge the existence of such rights and insist on their expiration.  There is a sound economic justification for a property regime like this, but it isn't "natural."

Many other examples of property rules important to capitalism, but not rationalizable within a theory of natural property could be cited--for example, bankruptcy (which discharges an insolvent debtor's debts, thereby abrogating the "natural" property claims of creditors), the rule against perpetuities (some version of which is needed to ensure that property rights ultimately vest entirely in living people, to prevent the dead hand of the past from permanently weighing down future uses of property), anti-commons rules (which prevent owners from dividing their property into uselessly small bundles), and rules against shareholder oppression (which limit what majority owners can do to undermine minority shareholder interests).  Some version or other of these types of artificial property rule are vital for dynamic economies. (I don't pretend to defend the details of the ones currently in force.)  Of course, one person's modus ponens is another's modus tollens.  One might reject these types of property on the ground that they violate natural property rights.  But don't pretend that one's preferred system would still be able to sustain capitalism or preserve its observed advantages.

(2) Natural property systems do not generate the distinctive form of property essential to capitalism, namely, capital.

Natural property exists, in the sense that people do successfully establish private property regimes, based on local conventions, that are independent of and often in opposition to the state.  (Like my fellow-blogger Don Herzog, I happily embrace distinctions between state and society!) Indeed, natural property is by far the dominant form of property in the developing world.  The great Peruvian economist Hernando de Soto has documented that in major regions of the developing world, 65-85% of housing is extralegal (that is, it consists of squatter settlements).  The vast majority of retail markets and mass transportation in these regions also exist outside of the formal sector.

Since natural property systems exist as an empirical reality, and not just in the state-of-nature fantasies of philosophers, we can compare their performance with artificial property systems established by capitalist states.  De Soto's verdict is clear: natural property is inferior to artificial property.  The people who have only natural property are poor.  Moreover, they are not poor because the state actively interferes with their natural property systems.  For the most part, developing countries acquiesce in their formation.  Rather, they are poor because natural property systems cannot convert property into capital.  The distinctive feature of capital is that it has market value to strangers, to people who do not belong to the parochial community in which the property exists.  Capital can be sold to strangers and used as collateral on loans.  Capital's value rests on the fact that it is the locus of a massive expansion of the scope of cooperation and trust, beyond the face-to-face community, ultimately encompassing the whole world.  It is a great engine of cosmopolitanism, which is one reason why I support it.

The natural property theorist might insist that all that natural property lacks is formal recognition  by the state, exactly as it has been created by the locals.  This ignores the fact that natural property systems are profoundly idiosyncratic.  They vary in innumerable details from one locale to another, just as languages vary.  And just as language variation puts sand in the gears of cooperation and trust across linguistic groups, variations in natural property conventions impede the conversion of natural property into capital.  Strangers are reluctant to buy it, or accept it as collateral, and not just because they lack confidence that their property claim will be enforced (by either the state or the locals) if a dispute arises.  They are reluctant also because they literally don't know what they are getting.  The local conventions that define and encumber the property are not written down.  Even if they were, they are too idiosyncratic in form to enable effective comparison with other parcels of natural property.  Without easy comparison with other parcels, natural property lacks fungibility and so lacks a market value to strangers.

When the state grants legal recognition to natural property, as when it issues squatters title to the land they have been occupying extralegally, it incoporates that property into a vast system of artificial rules that are not of local making and that may well contradict the local conventions that previously defined the property.  Of course, as Robert Ellickson has shown in his wonderful Order Without Law, the locals remain free to observe the local conventions instead of the formal property rules.  But, as his work also shows, those conventions lose their force when strangers, who care little for local opinion, buy the property.  This is the price of the conversion of natural property into capital.  But the price is worth it.  This means that the state is not, and ought not to be, bound to respect natural property.

This is not to say that there cannot be a compelling case at times for the state to formalize natural property.  My point is rather that the case cannot be made on the ground that people have a fundamental right to whatever property they have acquired naturally.  This can't be the ground, because formalization does not merely recognize the property that was already there, but subjects it to a distinct and often contradictory artificial regime.   The case for formalizing natural property--incorporating it into the artificial capitalist regime--is rather that  in many cases this is the best way, of the available alternatives, to enable the owners to escape poverty, provide for their needs, and gain prosperity and a wider range of opportunities.

Now here's the rub:  those very same grounds also justify, at times, infringing on natural property and instituting new artificial property rights, such as social insurance entitlements, and hence the taxes to support them.  The general justification for any property regime rests on its ability to enable, to the highest feasible degree, everyone under it to live a decent life, enjoying dignity, personal independence, freedom from poverty and oppression, a wide range of opportunities, and the effective power to participate in the social and economic life of the community.  No one has a right, "natural" or otherwise, to a property regime that in fact deprives others of effective access  to such a life, if there is an available alternative property regime that does effectively secure these others such a life.  Such alternatives have been found through experience to require measures such as social insurance entitlements and the taxes needed to support them.

(3)  A pure system of freedom of contract, in which all property is fully alienable, tends to degenerate to feudalism.  Capitalism therefore needs restrictions on freedom of contract.

Feudalism is based on the principle that private property in land confers political power over whoever is on the land.  One's landlord is one's lord.  Feudalism permits the conversion of property over things into subjection over people.  Theories of natural property rights, which suppose that people have property in themselves, and that all property is alienable in contracts, permit the same conversion.  If the state places no limits on such conversions, then, given the volatile nature of capitalism, many people will be pushed into a poor bargaining position.  In such a position, many will sell their personal independence for subsistence.  Thus arise oppressive forms of contract feudalism such as those into slavery, bonded labor, and debt peonage.  Thus arose company towns in the U.S., which issued scrip instead of cash wages to their workers (redeemable only in company-owned stores), required workers to rent company-owned houses (with leases enforcing what the company deemed a proper lifestyle), and crowned the firm owner as mayor for life (without those pesky elections).  Thus arise modern factories in quasi-capitalistic China today, which keep their workers locked up in factory-owned dormitories, forbidden by contract to wander outside, lest they hire out their labor (enhanced by the training provided by the firm) to competitors.

Theories of natural property misidentify the objection to feudalism.  They suppose that what made feudalism objectionable is that the landlord's political power lacked the consent of the people.  On this view, contractual forms of feudalism, such as debt peonage and company towns, are legitimate.  But if fedualism is objectionable, it is not for lack of consent.  As Hume observed, the people did consent to the rule of their lords.  Their consent, however, had no legitimating force, because they had no reasonable alternative:

Can we seriously say, that a poor peasant or artisan has a free choice to leave his country, when he knows no foreign language or manners, and lives, from day to day, by the small wages which he acquires? We may as well assert that a man, by remaining in a vessel, freely consents to the dominion of the master; though he was carried on board while asleep, and must leap into the ocean and perish, the moment he leaves her.

The same objection applies to contemporary forms of contractual feudalism.  But even this objection does not get to the core of the issue, which is not consent, but rather the content of the relationship.  Feudalism, whether contractual or not, is objectionable because it constitutes a relation of personal subjection, in which one party enjoys arbitrary power over another.  Because personal independence is essential for liberty, it should be considered an inalienable right.

This entails limitations on freedom of contract that go well beyond the abolition of contract slavery, debt peonage, and company towns.  It justifies legal regulation of rental contracts, for instance, so as to guarantee tenants a right to privacy against unannounced or too-frequent invasions by their landlord.  It justifies laws against quid-pro-quo sexual harassment.  (In light of this analysis, quid-pro-quo sexual harassment should not be viewed as any ordinary contractual term, but as contract feudalism's analogue to the droit du seigneur, now applied directly to the vassals, whether male or female, rather than their wives.)  It even justifies laws restricting corporate contributions to politicians.  Without such restraints on contract, we don't get capitalism.  We get contract feudalism.  Capitalism can't survive as a distinctive formation without restrictions on the conversion of property into political power.

Under the capitalist system we have today, people's claims to property arise from a vast artificial system that has no natural foundation, and that rightly contradicts many natural property claims.   The system couldn't be capitalistic if it didn't.  Within such a system, people have no property claim against other parts of the artificial property system, including social insurance entitlements and the taxation needed to support it.  Since their prosperity arises from artificial property, no less than the economic security of those receiving social insurance entitlements does, their property claims enjoy no superior or prior status that could constrain the state's constitution of new entitlements and taxes that advance the proper goal of any property system--effectively securing a decent life for all.

February 03, 2005

Bush hasn't cut taxes

Anderson on Taxes, Elizabeth Anderson: February 3, 2005

This is a plea for terminological honesty, as a prerequisite for fiscal sanity.  President Bush boasts, as one of his major policy achievements, that he has "cut" taxes.  Virtually all media outlets and partisan sources, including Bush's critics, follow Bush in calling his tax policies tax "cuts."  But Bush has not cut taxes.  He has merely postponed them.

Suppose you buy a $200,000 house by putting $100,000 down and taking out a loan for the remaining $100,000, to be paid off at 5.5% compound interest over the next 30 years.  Would it be accurate to say that you had to pay only $100,00 for the house?  Of course not.  You've simply postponed payment of the $100,000 balance.  According to this handy mortgage calculator, the cost of doing so--your total interest payment--will be $104,403, for a total 30 year housing cost of $304,403.

Ok, then.  Suppose, in 2004, you bought $2,292.2 billion of government outlays by putting $1,880.1 billion in tax revenues down, selling $412.1 billion in IOU's to the general public, and $151.1 billion in IOU's to the Social Security trust fund (in return for spending an equivalent amount of Social Security tax revenues on non-social security projects, included in the $1,880 billion figure above).  Would it be accurate to say that you had to pay only $1,880.1 billion for government?  (Data from the Congressional Budget Office.) Of course not.  You will have to pay, on top of that, $563.2 billion ($412 billion borrowed from the general public + $151 billion borrowed from Social Security), plus interest (which, according to the Bureau of Public Debt, was running on average at 4.5% at the end of 2004) in future years to pay for government spending in 2004.

Let's face the facts.  The nominal "price" of government (not the same as its total cost) is equal to the amount of government spending, not the amount of tax revenues.  If the spending isn't paid for today, it will have to be paid for tomorrow, with interest.  Bush has been steadily increasing government spending, as you can see here.  Falling tax revenues today mean rising tax bills tomorrow.  To be honest, we should call Bush's tax policies, given the absence of commensurate spending cuts, tax postponements.  They are not tax "cuts."

Against this, supply-siders have, since Reagan, promised that cuts in current tax rates will increase future tax revenues enough to close the deficit, by stimulating economic growth.  But that didn't happen in the Reagan years, and it didn't happen with Bush's tax policies, either, as the data show.

Of course, the government could avoid raising taxes by defaulting on its debt.  But doing so would be far more costly than raising the taxes needed to pay it off.  We'd face a run on the dollar, soaring interest rates and inflation, and likely economic collapse.  Indeed, at current rates of borrowing, we are likely to face these consequences long before default is imminent.

It could also be argued that it's ok to call Bush's policies tax "cuts" if they will be paid for in the future with spending cuts, rather than future tax increases.  But Bush's own spending policies, notably including the new Medicare drug benefit, are plainly inconsistent with the kinds of spending cuts that would be needed to prevent future tax increases.  In fact, according to the Government Accounting Office's latest report, "today’s fiscal policy Is unsustainable" (p. 5).  Among the nuggets reported there:

  • "In fiscal year 2004 the federal government added $13 trillion in new liabilities, unfunded commitments, and other obligations, principally due to the new Medicare prescription drug program" (p. 7).
  • "The federal government’s net liabilities, unfunded commitments, and other obligations now amount to more than $43 trillion, or about $350,000 for every full-time worker, and these unfunded commitments are growing larger every day" (p. 7).
  • If Bush carries out his promise to make his tax "cuts" "permanent," and discretionary spending [only 1/3 of total federal spending] grows at the same rate as the GDP, "by 2040 the government would have only enough money to pay interest on the federal debt" (p. 8).
  • "Social Security is a relatively small part of the long-term fiscal challenge when compared to spending on health care. . . . [T]he estimated Social Security shortfall is about one-third the estimated cost of recent tax cuts if made permanent" (p. 12).

Bush's fiscal policies are leading us to disaster.  It's time to call Bush out on his fraudulent bid to make his tax "cuts" "permanent."  A good start would be to stop calling them "cuts" and start calling them tax postponements.  That would make the fraud transparent.  You can't postpone bills forever.

I'm not claiming that tax increases can solve this mess by themselves.  Clearly, strong steps will be needed to rein in spending increases on health care entitlements as well, since they are a chief cause of the problem.  (For the reasons stated here, Social Security presents no more than a minor problem, and Bush's plan would make it worse.  Why are we wasting time on a hangnail, when heart failure looms?)  But please, let's stop pretending that Bush has "cut" taxes.  And let's stop taking seriously his talk of making tax "cuts" "permanent." 

March 12, 2005

How Not to Complain About Taxes (IV): Productive Contributions, Self-Sufficiency, and Individualism

Anderson on Political Economy, Anderson on Taxes, Elizabeth Anderson: March 12, 2005

My previous posts on taxation have been negative:  they've argued that certain arguments against taxation don't work.  It doesn't follow from them (nor do I believe) that the state really owns everything or that it can tax as much as it likes.  Despite having embraced capitalism, along with a list of Lockean constraints as a minimal starting point, it seems I must say something stronger to prevent readers from leaping to conclusions like this.  So, I'll drop a hint as to my positive view:  while pre-tax conceptions of property do not set legitimate constraints on taxation, the practical requirements of establishing a free society do set constraints on taxation.  More precisely, we can move from considerations of what it takes for each person to live freely, to a joint specification of constraints on both private property and taxation.  I hope that's tantalizing enough to keep you reading my subsequent posts on political economy, without trying to stick me with crazy views. (Further hint: I'm going to endorse F. A. Hayek's argument for a system of pure procedural justice.) For this post, however, I'm still clearing the ground with a negative argument.

In my series of posts on taxation, I've also been exploring the nature of capitalism and the rationale for social insurance.  I've been looking for a sound argument that people have such a strong claim to their pre-tax income that it would be wrong to tax it to fund social insurance.  I have previously argued that the none of these claims support that conclusion:  that "it's mine," that "it's mine by a natural property right," and that "I deserve it."  This time I'll be looking at 2 related arguments for the same conclusion. One asserts a claim of justice, based on a quasi-Lockean labor theory of value:  "I'm entitled to my pre-tax income because it's what I've produced."  The other asserts a claim of virtue:  people ought to be self-sufficient.  On this view, which I have criticized before, social insurance is bad because it undermines individual self-sufficiency.

My argumentative strategy is the same as with the prior claims:  I'll argue (1a) that our capitalist economic system does not in fact reward people according to the proposed principle--in this case, according to their individual productive contributions.  Far from tying each individual's fate solely to his or her own personal productive contribution (or to the pooled product of each individual's family, plus whatever they can beg from charity), capitalist markets commit us, at a deep structural level, to (b) extracting uncompensated benefits from others, (c) imposing involuntary costs on others, and (d) sharing our fates with one another.  (2) Nor should we revise the system of rewards so that it does conform to the proposed principle, because such revisions (supposing they were possible) would be incompatible with advanced capitalism and destroy its virtues.  (3) The self-sufficiency critique of social insurance implicitly relies on the view that the market rewards people in accord with their productive contributions.  Since that view is mistaken, so is the self-sufficiency critique.

1a. The claim that people are entitled to what they have made has strong intuitive support.  In a world of independent, non-cooperating producers, it's easy to see what this entails: as Locke famously argued (Second Treatise of Government, ch. 5, par. 27-30), I'm entitled to the nuts I have harvested from trees in the commons, to the previously unowned land I have cleared and prepared for cultivation, etc.  (This principle is distinct from the principle of desert, since it allows the justice of claiming a bumper harvest produced with the help of undeserved good weather.)

How do we translate this principle to cooperative production under an extensive division of labor? How do we measure each person's productive contribution to the outcome jointly produced by all participants working together?  P.T. Bauer, in Reality and Rhetoric: Studies in the Economics of Development, argued that individual productive contributions are measured by the marginal product of their factor (labor, land, capital) times the number of units of that factor they dedicate to production.  Since, in equilibrium, markets equate the price of a factor with its marginal product, markets therefore reward people in proportion to their productive contributions.

This idea doesn't work.  The sum of the marginal products do not add up to the total product under increasing or decreasing returns to scale.  Moreover, the ubiquity of declining marginal returns means that the nth unit of labor generally contributes less than the (n-1)th unit.  More fundamentally, the function of factor prices is to direct production decisions at the margin, not to reward people for their total productive contributions.

b. When the division of labor is efficient, we are systematically enhancing each other's productivity.  The situation is like that observed by a high school field hockey coach, who once had the top two players in the league on his team.  While both were fine players, he thought the second-ranked player was not as good as she was ranked:  she was scoring lots of goals because the defensive players on the other team were ganging up on her first-ranked teammate, giving her more opportunities to score.  On a more mediocre team, she wouldn't have ranked that highly.  In the economy at large, lower-ranked workers enhance the productivity of higher-ranked ones as well as vice-versa:  big-time executives couldn't be cutting so many lucrative deals if they didn't have secretaries to screen their calls.  Even the unemployed contribute to the productivity of employed workers.  If firms couldn't count on being able to hire quickly from a pool of readily employable people, they'd have to hire more of them up front in anticipation of expansion, so as to avoid bottlenecks later on.  But doing so would dilute the productivity of all workers in a firm before the expansion has taken place.  Many people not in the labor force also contribute.  Full-time homemakers relieve their partners of the time, stress, and bother of managing the household themselves, thereby enabling them to be more productive at work.  Full-time parents contribute to production by raising the next generation of workers.

It might be argued that these positive productive externalities that people are conferring on others are captured by each person's market wage.  The cases of the unemployed and unpaid homemakers and parents show otherwise. Their productive contributions are not rewarded by the market.  The contributions of the unemployed are not rewarded by any private orderings.

More generally, market wages don't track any physical measure of positive productive benefits conferred on others, since they vary with factors independent of the firm's production technology.  Wages vary with the number of other people competing for the same job, the power of competitors to hold out for better terms, the costs to the worker of changing jobs, the availability of alternatives, and the costs the worker can impose on the firm by quitting.  These factors don't track physical measures of a worker's contribution.  Rather, they track people's bargaining power.  "It's my productive contribution" sounds like an intuitively compelling ground for recognizing an entitlement of people to keep their market-derived income.  "It's my threat advantage" doesn't sound so compelling. (Thanks to fellow-blogger Don Herzog for that phrase).

c. Not only are we systematically contributing benefits on one another, many of them uncompensated, we are also imposing involuntary costs.  Negative externalities such as pollution and traffic jams are pervasive.  Notably, capitalism imposes its worst negative externalities on those least able to avoid or afford them.  Poor people are subject to considerably higher levels of pollution than the well-off.  Our negligence system of torts entails that people are imposing costs on others without having to pay for them, as they would in a system of strict liability.

d. Capitalism, through its business cycles, forces us all to share each other's fates.  Our fortunes rise and fall together (albeit unequally) in virtue of our interdependent interactions.

2. Should we try to arrange our laws so as to ensure that people reap only the productive contributions attributable to their own factor inputs, and internalize all their negative externalities? Such a system would promote the individualist dream of each living only off of their own production, none being a net uncompensated creditor or debtor to others.  The objection to this is the same as Hayek's objection to distributing rewards according to deserts:  trying to jigger rewards according to an external standard of justice such as productive contribution will distort prices away from their critical informational function.  It would destroy the virtues of capitalism.  Moreover, just as Hayek argued that the idea of just deserts is illusory in the case of factor compensation, so is the idea of individually assignable productive contributions.  There is no physical fact of the matter regarding what any individual's productive contribution is, in an economic system structured by an extensive division of labor rather than by indepedent producers.

The libertarian Richard Epstein, following through on individualist intuitions about justice, has argued in favor of replacing our negligence system of torts with a system of strict liability ("A Theory of Strict Liability," 2 Journal of Legal Studies 151 (1973)).  Such a system would ensure that whenever one person caused damage to another, they'd have to compensate them for it.  Long before we reached this point, I'd be joining the critics of the tort system in crying "crisis!"  The problem wouldn't be frivolous lawsuits.   It's that when negative externalities are systematic, it's grossly inefficient to deal with them on a case-by-case basis, as the tort system does.  The costs of litigation are too high, not in the amounts of judgments, but in lost production opportunities, as the valuable time of defendants and plaintiffs gets sopped up in lawsuits.  That's why I prefer alternatives to litigation, such as regulations for worker safety and pollution, workers' compensation, and no-fault auto insurance, to cover systematic costs imposed on people by our advanced capitalist system of production.  Excessive reliance on Individualized solutions puts too much sand in the gears of capitalist production.

3. These reflections offer support for my much-maligned post on social insurance and self-sufficiency.  The self-sufficiency critique of social insurance implicitly relies on a rugged individualist ideal.  According to this ideal, each person should rely only on their individual production, none being a net lifetime debtor to others. The self-sufficiency critique supposes that pre-tax capitalist distributions embody this ideal, in that they reward people according to their personal productive contributions.  If this were true, and the ideal were sound, then it could make sense to call upon each person to rely only on their market income to survive.  But as Hayek has argued, market prices don't track any notion of virtue.  Once we detach market rewards from any moralized conception, such as deserts or productive contributions, it's hard to see how they could set a coherent standard of self-sufficiency.  What's the virtue in living within the constraints of one's threat advantage?  (What's the virtue in exercising one's threat advantage so as to drive the less fortunate to the wall?)

The self-sufficiency critique imagines that capitalism embodies a society of rugged individualists, like some imagined self-sufficient isolated producers in the state of nature.  In fact, capitalism embodies a society of deeply interdependent people who share one another's fates.  Social insurance simply makes this fact explicit.

May 27, 2005

So You Want to Live in a Free Society (1): What Hayek Saw

Anderson on Political Economy, Anderson on Taxes, Elizabeth Anderson: May 27, 2005

So far in my posts on taxation and political economy, I've mainly been making negative arguments--that this or that case against taxation to support social insurance doesn't work.  It's time now to start building a positive case for social insurance and the taxation needed to support it.  Most of you have heard arguments for social insurance based on ideas such as equality and compassion for the less fortunate.  For many, such arguments cut little ice because they view the system being defended as incompatible with freedom.  Ok, then, let's take freedom as our starting point and foundational value.  Suppose you want to live in a free society--one in which everyone is free.  What institutions, what types of distributive rules, what kinds of constraints on coercive action, what sort of property regime, should you support?  Over this new series of posts, I'm going to lay out my view of what's needed to have a free society.

This first post in the series is dedicated to F. A. Hayek, who had a deep insight into what's needed.  In Law, Legislation, and Liberty, vol. 2, Hayek argued that, for a society to secure the liberty of all, its distributive rules cannot aim at achieving some pre-established pattern of distribution based on individual need, desert, or merit.  Instead, they should be purely procedural in form.  Set up a system of fair, impersonal rules governing our interactions and applicable to all, let people choose freely from among the opportunities generated by acting within the constraints of the rules, and whatever distributions of goods result from following the rules will be just.

Hayek likened the procedural rules constitutive of a free society to the rules of a game:

namely a game partly of skill and partly of chance. . . . It proceeds, like all games, according to rules guiding the actions of individual participants whose aims, skills, and knowledge are different, with the consequence that the outcome will be unpredictable and that there will regularly be winners and losers.  And while, as in a game, we are right in insisting that it be fair and that nobody cheat, it would be nonsensical to demand that the results for the different players be just.  They will of necessity be determined partly by skill and partly by luck.  Some of the circumstances which make the services of a person more or less valuable to his fellows, or which may make it desirable that he change the direction of his efforts, are not of human design or foreseeable by men. (Law, Legislation, and Liberty, vol. 2: The Mirage of Social Justice, p. 71)

In this passage, Hayek denies that the concept of justice can even apply to the outcomes of procedurally fair rules, for two reasons.  One is that, because luck is inevitably involved in the outcomes of actions following fair procedures, the outcomes can't be relied upon to track individual merit or desert.  The other is that the concept of justice can apply only to things that are deliberately willed, but the outcomes of free individual interactions within procedurally fair rules are unintended consequences of everyone's behavior.  I think Hayek was mistaken on the latter point of usage.  When we say that the winner of a contest won it "fair and square," we imply that justice would be served by awarding the prize to her, so it is just that she receive it.  This is just a verbal quibble, however.  The key point, on which Hayek was correct, is that the just outcome can't be determined ex ante, before people have played the game.

Why ought a society, to be free, distribute goods according to purely procedural rules?  First, consider the  leading alternative:  what would a society be like if it tried to distribute goods according to some notion of individual merit or desert?  Given that the outcomes of free exchanges inveitably include some element of chance, to adjust the outcomes so that they reflect some prior notion of merit or desert would require that the state look over everyone's shoulders to see how they are using their liberties.  If, in the state's judgment, an individual used her liberties poorly or irresponsibly, then she is responsible for whatever disadvantages come her way and society will not compensate her for them.  But if the state judges that her disadvantages were the result of mere luck, which is undeserved, then society will compensate her.  There are of course other ways to draw the line between deserved and undeserved outcomes--indeed, too many ways, which put people into endless disputation over which way is the right way. (A look at recent literature on egalitarianism, full of disputation about how to draw the line between luck and desert, confirms this.) But all of the ways of drawing the line and redistributing goods accordingly require the state to make and enforce intrusive judgments about how people are using their freedom.  People can't be free under such a system, where the state is monitoring their choices and passing moral judgment on them, with attendant material consequences.  This is the ultimate busybody state.

Of course, not any random set of procedural rules will enhance freedom.  Distributing all income according to a lottery, for instance, would be an instance of pure procedural justice.  But that would be a crazy system to implement.  What is needed is a set of rules that leave people free to offer mutually advantageous exchanges, so as to systematically give people incentives to behave in ways that overall enhance the liberty and opportunities of everyone else.  Markets play an indispensable role in this, because prices signal to people where their productive efforts will be most valued by other people.  In contrast with a command economy, individuals in a market system are free to take or leave any particular opportunity open to them, free to respond to or ignore any particular bargain or incentive offered to them.  Moreover, market prices reflect the aggregate result of everyone's free decision to demand this or that, rather than some bureaucrat's notion of what they ought to be consuming.  These are two extremely important ways in which a system of procedural justice based on voluntary market exchange secures everyone's freedom.  However, the most important way in which reliance on markets enhances everyone's freedom concerns the dynamic effects of market competition in a private property regime in producing ever-expanding opportunities.  I'll postpone to a later post an explanation of this, which I believe gives us the core freedom-based argument for private property.

A market system does not preclude all consideration of individual deserts.  Importantly, when people violate the rules of a free society, we enter the realm of retributive justice.  Here, we do know ex ante what the just results of a trial should be:  namely, that all and only those guilty of violating the rules be punished (or, in a civil trial, compensate those they have harmed).  The quality of a person's intentions--whether they did something intentionally, or unwittingly--matter here.  But as long as people are abiding by the rules, the state takes no interest in their individual deserts.  Consideration of individual deserts may also play a role in local distributions--say, within a firm.  Employers often voluntarily implement merit-based pay structures, for example.  But in a market system, that one's rewards reflected one's merits relative to the other employees in one's firm is no guarantee that one's rewards reflect one's merits globally--that is, relative to employees of other firms.  For the size of the total compensation pie available to any given firm to divide among its employees is typically determined by chance factors--for example, an unanticipated shortage of some input, or sudden surge in demand for a product--independent of anyone's merit.

It might be thought that a system of pure procedural justice must place no constraints on possible outcomes for individuals, lest the constraints intefere with individual liberty.  So pure procedural justice must permit catastrophe to befall the unlucky.  Robert Nozick famously argued for this position in Anarchy, State, and Utopia (1977), encapsulated in his slogan that "liberty upsets patterns."  This is the main reason (concern for individual deserts aside) that many egalitarians have objected to letting free markets determine distributions.  But there is nothing in the idea of pure procedural justice, nor in the liberty it secures--to freely choose any of the opportunities generated by spontaneous interactions within the constraints of the rules--that precludes placing constraints on the outcomes.

To see this, we can pursue Hayek's analogy of markets with games by looking at the rules of some actual games.  Games provide the paradigm of pure procedural justice, because there is no notion ex ante of who should be the winner, the same rules apply to all, and the rules are designed to be fair to all, in the sense of giving everyone a basically equal ex ante chance to win, supposing they play with equal skill.  (Sometimes unavoidable asymmetries in a game give a slight advantage to a particular player--for instance, the one who gets the first move.  But the rules of games are typically designed to prevent this advantage from being decisive, lest the game be boring for lack of uncertaintly about the outcome; and access to that advantage is itself typically allocated by a fair procedure, such as a coin toss or roll of the dice.)

The game of Monopoly illustrates a system of pure procedural justice that matches Nozick's ideal of unconstrained outcomes.  In Monopoly, each player's objective is to drive all of the other players into bankruptcy, and to end up owning all of the property in play.  Monopoly is a game that does not constrain how low people can go, or how high they can go, within its rules.

Milton Bradley's game of Life illustrates a system of pure procedural justice with constrained outcomes.  In the game of Life, each player's objective is to retire with the most money.  Although wealth inequality is inherent to the game of Life, it constrains the outcomes in three ways.  First, nobody goes bankrupt; everyone retires with something.  (I suppose it's technically possible to retire with a negative net worth, but I've never seen it happen.  I suspect that that the rules are designed so as to virtually preclude this possibility.)  Second, in the course of the game, players collect "Life tiles," which give them windfalls.  When the draw pile of Life tiles runs out, a player who lands on a "Life space" gets to take a Life tile from any opponent.  Strategically rational players will take their tiles from the richest opponent.  Thus, the game of Life contains a redistributive element that in practice constrains how wealthy the richest player will get.  Third, once people acquire assets (a house or a car), they can protect them by buying insurance.  Insurance is a device that constrains middling outcomes by means of a ratchet--that is, it protects people who have already acquired some assets from losing them.

The game of Life illustrates how a system of pure procedural justice can consistently constrain outcomes at the bottom, at the top, and in the middle.  It can even implement these constraints by way of redistributions from the top to those below.  I don't pretend to have offered an argument that we should prefer a system that implements such constraints.  I just want to point out that there is nothing in the idea of pure procedural justice, even one based on granting free markets a large role in determining distributions, that precludes setting constraints on possible outcomes.

Robert Nozick famously objected to John Rawls' egalitarian Theory of Justice that it was a "pattern-based" theory of justice that, because it identified just distributions independently of how people play by the rules, is incompatible with a free society.  He was wrong.  Rawls, the leading egalitarian theorist of the 20th century, in fact endorsed a system of pure procedural justice that insisted on constraining the top and bottom outcomes of a market-based "property-owning democracy."  As he made clear in the revised edition of his book, the idea of a "property-owning democracy" is to use devices such as progressive taxation and rules promoting competition so as "to disperse the ownership of wealth and capital, and thus to prevent a small part of society from controlling the economy and indirectly political life itself" (Theory of Justice, rev. ed., xiv-xv; thanks to Amit Ron for drawing this passage to my attention).  (By the way, Rawls on these pages contrasted his preferred system of "property-owning democracy" with a "welfare state," which aims to protect the unlucky from the worst misfortunes.  The goal of a property- owning democracy is rather to secure the material conditions for democracy, in part against the threat of plutocracy.  I'm not arguing for Rawls' position here; just highlighting the fact that egalitarians have more than one reason for constraining market outcomes.  A concern for protecting the material conditions of democracy and equal citizenship is utterly distinct from compassion for the less fortunate.)

Hayek saw what Nozick failed to see: that Rawls' egalitarianism, while it contrained possible outcomes at the top and bottom, is in fact a system of pure procedural justice.  It was not a "pattern-based" theory, and hence not subject to Nozick's objection that a free society will invariably upset patterns.  Here's what Hayek said about Rawls' Theory of Justice:

the differences between us seemed more verbal than substantial.  Though the first impression of readers may be different, Rawls' statement which I quote later . . . seems to me to show that we agree on what is to me the essential point [that distributive justice in a free society must take a purely procedural form].  Indeed . . . it appears to me that Rawls has been widely misunderstood on this central issue (L, L, L vol. 2, xiii).

Widely misunderstood, not least by Nozick.  Hayek continued his observations on Rawls as follows:

there unquestionably also exists a genuine problem of justice in connection with the deliberate design of political institutions, the problem to which Professor John Rawls has recently devoted an important book. . . . I have no basic quarrel with an author who, before he proceeds to that problem, acknowledges that the task of selecting specific systems or distributions of desired things as just must be "abandoned as mistaken in principle, and it is, in any case, not capable of a definite answer.  Rather, the principles of justice define the crucial constraints which institutions and joint activities must satisfy if persons engaging in them are to have no complaints against them.  If these constraints are satisfied, the resulting distribution, whatever it is, may be accepted as just (or at least not unjust)." This is more or less what I have been trying to argue in this chapter. (L, L, L, p. 100, quoting Rawls, "Constitutional Liberty and the Concept of Justice," Nomos IV: Justice (New York, 1963), p. 102)

It's worth noting that Hayek's preferred system of pure procedural justice, while it differed from Rawls' in rejecting constraints on the top outcomes, did, unlike Nozick's system, insist on constraining the worst outcomes for individuals.  He supported state action to abolish poverty in the sense of deprivation relative to objective needs (as opposed to relative to what others have) (L, L, L, vol. 2, p. 139).

I want to stress again that I'm not arguing for Rawls' system.  It isn't, in fact, my preferred system.  What I've argued for is the following:

1. Hayek was right to insist that the rules of distributive justice for a free society must take a purely procedural form.

2. Free market exchanges among private property owners play an indispensable and central role in any system of pure procedural justice that aims to secure and increase freedom for all.

3. A system of pure procedural justice in a system of private property and free exchange is consistent with rules that constrain outcomes at the top, at the bottom, and in the middle of distributions, and that implements those constrains by means of redistributive mechanisms.