In my last two posts, I explained Rawls’s idea of a property-owning democracy (POD) as elaborated by Thad Williamson and Martin O’Neill in much of their recent work, including this Boston Review piece. In this post, I argue that property-owning democracy is unworkable. I shall make two arguments to this effect:
(1) POD is unworkable at the level of ideal theory because, even under idealized conditions, it generates bad incentives and faces severe information problems.
(2) POD is unworkable at the level of nonideal theory due to the problems in (1) and due to self-interested behavior by officials and citizens not holding political office.
When I say that POD is unworkable, I mean that it plainly fails to realize several aims of any plausible regime type. For instance, if POD is supposed to maintain market efficiency but significantly and systematically falls short, it is unworkable. Similarly, if POD is supposed to realize the difference principle by maximizing the position of the least well-off, and plainly fails to do so, then it is unworkable. In sum, unworkability means that POD fails to meet its aims in an especially significant fashion.
I. Rawls on Ideal and Nonideal Theory
To understand Rawls’s nonideal theory, we should first review the terms of ideal theory. The most significant assumption in ideal theory is strict compliance in that persons are assumed to comply with the dictates of just institutions in virtue of having a normally effective sense of justice, that is, they are motivated to be just even if justice is not in their short-term interest (Rawls assumes that people will see just action as in their long-term interest under certain conditions).
Strict compliance does not assume that all principles motivate compliance to the same degree. Different principles of justice will generate different degrees of support and compliance. Instead, strict compliance only assumes that we can imagine a society’s institutions getting going, but some principles of justice will still violate the “strains of commitment” by exceeding “the capacity of human nature.” (TJ, 176).
Now, nonideal theory relaxes this assumption, but in different ways on different topics. For instance, nonideal theory covers civic disobedience, where people are compelled by the state to comply with unjust laws but must disobey for moral reasons. We can set this case aside. Rawls’s nonideal theory also covers international ideal theory, which we can set aside. A third, less explicitly defended, form of nonideal theory covers cases where individuals fail to comply with the duties associated with justice as fairness, such as not supporting just institutions or obeying the law. It is this third form of nonideal theory I shall focus on.
II. More on Ideal and Nonideal Theory (For Specialists)
It’s probably helpful to reproduce John Simmons’s six-part division of Rawlsian nonideal theory to give us a better grip on what sort of ideal/nonideal distinction we’re after.
1. Basic Structures: a. deliberate noncompliance (institutional injustice): civil disobedience, etc., b. unfortunate noncompliance (poverty, culture): the general conception.
2. Individuals: a. deliberate noncompliance (wrongs, crimes): punishment, reparation., b. unfortunate noncompliance (insanity, immaturity): paternalism.
3. Nations: a. deliberative noncompliance (outlaw states): just war, intervention., b. unfortunate noncompliance (burdened societies): international assistance.
The stuff after the colon is how to respond to the conditions described prior to it. For instance, you response to institutional injustice via civil disobediance.
In this post, I’m concerned with 1 and 2, that is, when institutions and individuals deliberately or unfortunately fail to comply with the principles of justice. I classify public choice type problems – self-interested behavior by politicians – in 1a. I will include rent-seeking by lobbying groups in 2a.
I don’t include the strains of commitment in IIb because noncompliance in that case is part of ideal theory. Remember that for Rawls a desideratum for the validity of a principle of justice is whether it is compatible with our personal good (that’s what Part III of TJ is all about). I will also not include information problems as part of nonideal theory, as these are already present in ideal theory.
And note that the problems in nonideal theory hold on top of problems present in ideal theory. So most of the problems in ideal theory apply to nonideal theory as well. For this reason, I begin with my criticism of ideal theory and then move to nonideal theory.
III. POD is Unworkable in Ideal Theory
Rawls believes that ideal regime types should exhibit a high degree of economic efficiency and, while not a requirement of economic justice, growth should be available to citizens should they decide to pursue it (understand efficiency at least in terms of the production of non-social primary goods). Thus, if POD undermines economic efficiency and successful growth and production, then a Rawlsian must worry.
My criticisms assume strict compliance. Officials will comply with the dictates of their office, as determined by the principles of justice. Second, citizens will comply with the law without trying to willfully transgress it. We’ll relax these assumptions later.
Recall that we are assessing the implementation of POD’s distinguishing features, namely (iv) and (v) in O’Neill and Williamson’s description:
(iv) A right of individuals to a share of society’s productive capital and/or wealth.
(v) A collective right to sufficient productive capital to sustain viable democratic communities at the local level.
Implementing these conditions requires a host of bureaucratic institutions (allocation, transfer, stabilization and distribution bureaucracies) that distribute capital in terms of real estate, cash and stock.
I identify two general classes of economic failures, i.e., failures to produce efficient outcomes:
Incentive problems – people lack an incentive to produce efficient outcomes.
Information problems – people lack the information to generate efficient outcomes.
O’Neill and Williamson’s version of POD will face both incentive and information problems. It is simply remarkable that they seem not to acknowledge either as a possibility.
Note that I do not need to appeal to empirical studies here, as we’re working at the level of ideal theory. Rawlsians confine themselves to arguments from economic theory, so I will do likewise.
The main aim of POD is to distribute and redistribute capital goods so as to have them in a great many hands and run by a great many people. PODs use coercion to bring about these outcomes in order to prevent a few people from accumulating too many capital goods. So all exchanges that would leave some with a relatively greater amount of capital than others are subject to coercive blockages. This does not mean that a POD must coercively block such exchanges, rather that it will exhibit a general tendency to do so when capital accumulation violates (iv) and (v).
Further, PODs will provide people with the fruits of this coercion (largely through the reception of holdings) in accord with conditions (iv) and (v).
So, we’re covering problems that arise at the level of ideal theory in the systematic, bureaucratic distribution of capital goods and capital stock, along with the normal effects of such distribution on the market process, among them deliberate alterations in prices and practices.
a. Incentive Problems for an Ideal POD
PODs will create a wide variety of incentive problems. First, a POD prevents the accumulation of capital by dispersing it when it becomes too concentrated, or at least creates conditions under which capital is harder to accumulate. Note that it does so through the use of force. As we saw in the first place, while Williamson claims (without argument) that a POD will be in equilibrium, such that coercion will not need to be used very often, coercive threats are always present if people get out of line.
Consequently, the first incentive problem for an ideal-POD is that (iv) and (v) reduces the incentive to create and accumulate capital. First, everyone in a POD knows that capital accumulation has legal limits, as dictated by the massive administrative POD bureaucracy. As a result, effective capital accumulators will only seek to accumulate capital up to the point where doing so exceeds marginal costs and runs up against the legal maximum. In the name of the Different Principle, the Fair Equality of Opportunity Principle and the Political Liberty Proviso, the POD effectively tells capital producers to stop producing capital that they would produce if they could own and operate it.
What’s worse, the positive externality created by the accumulation of capital, namely the signaling generated by the presence of a few successful capital accumulators in a particular area of the economy, is significantly reduced. Prices are relative informational signals – they tell entrepreneurs how to redirect their resources to increase profits. If these signals are significantly blunted by a POD bureaucracy’s distribution of capital, there will be less incentive to create the information necessary to create the relevant signals. As a result, capital will not be reallocated to its most efficient uses.
Think about it. Whenever someone has too much capital, it will be taken from him, given to his workers, or invested in public services or capital-based funds for others. That is a powerful disincentive that rational Rawlsian members of a well-ordered society will respond to, for the law and justice does not require them to respond differently.
A natural reply is that those who receive the capital won’t just sit on it but will do something useful with it. But if the natural course of the market gives capital to those who (for whatever reason) use it more effectively, then a POD will take capital from those who best know how to create it and give it to those who do not. (Note here that I’m assuming there’s no massive market failure in the creation and distribution of capital – as far as I am aware none of the POD-people have claimed that there is.)
Another set of disincentives arises from the recipients of the various funds provided for by the state. Few surprises here: to the extent that the state provides citizens with unconditional funds, say through savings, checks, funds and services, to that extent those citizens can avoid working to acquire them. Thus, the productivity that would be generated by citizens who would otherwise have to work for these goods will be lost. A POD might provide a twenty-five year old with all of the following: free public education through graduate school, a $100,000 savings account awarded at adulthood, public housing, a permanent stock holding in the company that offers him his first job and a guaranteed, unconditional minimum income. Surely, surely, these people will be less productive workers if they receive all of these goodies for free and are taught to believe they are owed them as a matter of justice.
We can construct other disincentive arguments, but you can now see them popping up everywhere. When the POD blocks voluntary exchanges in the name of satisfying the DP, FEO or PLP, they block Pareto improvements that can redound to the benefit of all. Furthermore, PODs provide free resources to people who would otherwise have to produce to acquire them.
An obvious rejoinder: if Rawlsian actors face a disincentive, the state can simply make a law requiring them to do otherwise. And since we assume strict compliance, they will comply with the law. But not so fast. On Rawls’s view, strict compliance is not unconditional compliance. If laws significantly set back one’s good, say by requiring a person to produce capital for absolutely no material benefit, then such laws will undermine stability for the right reasons (or “inherent stability” in TJ). So while Rawlsian actors are public-spirited, they will quit working if the benefits of working are too low.
b. Information Problems for an Ideal POD
Remember that PODs have an extensive bureaucratic apparatus to administer the relevant services and impose the requisite regulations. These bureaucracies are charged with extraordinary tasks, such as making specific numerical judgments about appropriate price-levels, wage-levels, stock distributions, and the like. And they must do so without the ordinary information available to market actors, namely that provided by a free and functioning market. While Rawlsians like to claim that they advocate a free-market to allow for the distribution of goods and services, what they describe is a state that constantly intervenes in the market. After all, what are all of these incredibly large and powerful bureaucracies for if not to interfere with the market process?
Consequently, intervention distorts market signals. To see this, let’s contrast a POD with a market capitalist society that merely redistributes wealth via a negative income tax. No regulations, no breaking up industries, no price monitoring, etc. In that case, markets will suffer from few distortions given that there is (a) one single redirecting force (the negative income tax) and (b) the force is predictable so firms can adjust their production schedules around it.
But as we add POD-bureaucracies, multiple forces begin to interact, consistently readjusting and redirecting the market process. As these forces increase in number and degree, their distortionary effects multiply considerably. All else equal, increases in regulations and controls from an increasing number of government bodies make adjustment difficult to predict and even hard to understand in the first place. As a result, the informational function of prices will be significantly distorted.
Note that this is why Hayek was fine with a social minimum but strongly resisted a state that would regulate the economy through administration-made law. Markets are information systems and extensive regulatory states impoverish the amount of information available. And this holds even in the absence of government ownership and operation of firms. So the information problems arise even in the absence of central planning. And now notice that PODs permit and sometimes encourage government ownership and operation!
These information-based arguments illustration a general concern about the Rawlsian attempt to separate the “allocative” and “distributive” function of prices. Rawlsians want to get the benefits of efficient market allocations without producing unattractive distributive outcomes, such as excessive income inequality. Income inequality itself is a market signal, as those who make money in a particular field will grow relatively wealthier than others, signaling to others that there is money to be made in a particular field. POD-Rawlsians believe in a government that regulates in just the right ways to ensure that goods and services go to the same (or related) places they would if the relevant regulations and controls didn’t exist. How is this to be done? How are the bureaucracies to know what to do? What prices to set? From what I have read, Rawlsians have no real answer to this question.
Of course, POD-advocates will probably respond by trying to make the POD-ideal more flexible. Perhaps POD, for instance, does not require the massive bureaucracies that Rawls endorsed. Perhaps instead we could restrict the state to pursuing the least distortionary means towards achieving goals (iv) and (v). But what would those be? Williamson and O’Neill’s POD requires a government that performs a lot of these functions, at least in terms of determining everyone’s stock ownership below a certain minimum, figuring out how to divide up ownership shares of a company to give over to its workers, and the like. To realize (iv) and (v), a society will need a powerful government that will be invariably constrained by informational problems even at the level of ideal theory.
Unlike incentive problems, by the way, we cannot simply demand that people comply with the law because informational problems make it difficult even for appropriately motivated people to behave. Strict compliance does not increase access to economic information.
IV. POD is Unworkable in Nonideal Theory
Nonideal theory relaxes strict compliance and so permits both ordinary citizens and officials to break the law and violate their duties for a variety of reasons and motivations, so they can distort the law-making process via diverse moral and self-interested considerations.
We begin nonideal theory with the problems faced in ideal theory, namely the incentive and information problems. Those problems arise even under strict compliance. But now all the problems documented by public choice economists can be brought to bear on our assessment of POD’s workability. Administrators can fail to do their jobs by altering prices and distributing holdings in ways that benefit their friends and political allies. Political leaders can order administrators to engage in popular but destructive policies that sacrifice the short-term for the long-term. Individuals can lobby politicians to provide them with particular benefits that aren’t worth the social cost, and so on.
Given the extraordinary power that Rawls, Williamson and O’Neill assign to federal bureaucracies, the incentives for regulatory capture are huge. A state powerful enough to realize (iv) and (v) is worth staffing with friendly regulators, as the state could potentially completely destroy an individual or corporation’s livelihood through regulation and redistribution.
For instance, suppose that Reba is a small businessperson who has accumulated just enough capital to come under the scrutiny of the bureaucratic apparatus of a POD. She is now subject to monitoring. Perhaps Reba’s workers have complained to administrators that they have too small a share of the business owner’s capital stock, or perhaps a competitor has tipped off a regulator that she has been making new capital stock purchases. In both cases, Reba has an incredibly powerful incentive to resist the regulator in any number of ways, perhaps by supporting appointments to the relevant local regulatory board.
Is it really so hard for non-libertarians to admit that the public choice problems will be debilitating for a state that concentrates this much power in the hands of government officials?
V. POD is Unworkable
As we can see, POD will face enormous challenges, especially because POD bureaucracies will create systematic incentive and informational problems at the level of ideal theory, compounded by public choice problems at the level of nonideal theory. Remember: I don’t have to prove my presumptions are empirically true in order to criticize POD. It’s an ideal regime type, created at a high degree of abstraction, as a standard of economic justice for present regimes. To show that POD is unworkable, I only have to identify problems formulated in standard economic models given that the Rawlsian members of a well-ordered society will comply with these models. Remember that Rawls assumes as much, given his use of welfare economics, market failure theory, and so on. My claim here is that if we appeal to the full range of economic modeling, POD has serious problems even given Rawls’s simplifying assumptions.
So I think it is safe to conclude that POD is unworkable.
In the next post, I will argue that property-owning democracy is unjust on Rawlsian terms.
UPDATE: O’Neill and Williamson, fortuitously for me, have just published a critical review of John Tomasi’s Free-Market Fairness, a book familiar to our readers. Among their claims is that Tomasi has an unrealistic view of how economies function. I hope that if you’ve reached the end of this post you will find their claim rather bold given their almost complete insensitivity to basic economic concerns about property-owning democracy.