1. The Restitution Ratio Instability

In an earlier essay, A Fatal Instability in Anarcho-Capitalism?, I demonstrated the existence of a major difficulty for the implementation of anarcho-capitalism: the restitution ramp or restitution ratio instability. I shall thus attempt to summarise this problem before addressing the question of what such an unstable anarcho-capitalism could realistically be expected to evolve into.

We start with restitutional justice. In a just regime wrongs are put right; the offender pays for what he has done, and the victim receives full compensation for his loss. This is the basis of common law. Neither plaintiff nor defendant is discriminated against; penalties are set neither higher nor lower than necessary to make good the average loss (including costs). Retribution and restitution are equal and opposite.

Such a common-law regime underlies the free market. It defends voluntary trade: it punishes coercion. It is economically efficient. Without its justice the villain will plunder the weak, the scoundrel will prey on the simple, and the wastrel will live off the industrious.

In a just ultraminimal state the state court enforces the common law, even against such other courts as parties may choose to patronise; this keeps the other courts honest. But under anarcho-capitalism there is no state court. Law is determined on the market and will be just only if the market makes it so. Note that although just law produces an efficient market, the reverse is not necessarily true; just law is epistemologically prior to the free market, for without law the market is not coercion-free. Thus we can hope but cannot guarantee that anarcho-capitalism will be just; a free market is not inevitable. There are roughly two possibilities: either the anarcho-capitalist market tends towards the free market as a stable limit; or the anarcho-capitalist free market is unstable and tends to dissolve into something else. I argue that the latter is more likely.

A just court operates at a restitution ratio of unity (R=1). Anarcho-capitalist courts will not be able to operate at a lower ratio; for if they try they will lose business to courts that offer full restitution. Market forces drive them from the sub-just to the just regime.

Unfortunately, the same market forces now drive them to increase the restitution ratio further. Courts that offer plaintiffs super-restitution attract custom away from the conservative simple-restitution courts; other courts must follow suit or go out of business. Some courts increase the ratio still more, hoping to steal a march on the competition. A "restitution war" breaks out. But unlike a conventional price war, there's no direct constraint upon the restitution ratio, since neither the courts nor the customers have to pay the bill. Where does the money come from? It comes from convicted offenders, who now face correspondingly higher penalties for their deeds. But who cares about them?

Higher penalties act as a deterrent. Fewer crimes get committed. So now the courts are competing for a smaller cake; they must take a larger share or go under. The restitution ratio takes off. The effect is of a classic speculative ramp, like tulip mania, hyper-inflation or a stock market bubble. Unable to keep pace, more and more courts collapse. Soon only a single court is left, holding a monopoly of a shattered market.

If the last court fails also (as is probable) anarcho-capitalism dies with it, replaced by a pure anarchy which almost instantaneously dissolves into chaos: law of the jungle; dog-eat-dog. If the court survives, however, it becomes a de facto state; depending upon the circumstances it may become a minimal or ultraminimal state, or an oppressive one; whichever, the regime is no longer an anarcho-capitalist one.

This, in short, is the restitution ratio instability. In the previous essay I examined various ways in which an anarcho-capitalist society might attempt to avert or survive it. I found no credible escape route; the instability would appear to be fatal.

2. The Nature of the State

The restitution ratio instability is based upon the assumption that the diverse courts are in competition with each other. Where such competition is lacking there is no reason for the ramp to take place; but without such competition society could not properly be described as anarcho-capitalist.

It is important to distinguish here between genuine anarcho-capitalism and other deviations from the archetype of the unified sovereign nation state. Many libertarian writers fail to draw this distinction (I shall spare their blushes by not naming them); one of their, in my opinion, foolish catch phrases is polycentric law. We have polycentric law already, across the nation states of the world; but it is neither just nor anarcho-capitalist. Indeed, we have always had polycentric law; and it has never been either generally just or anywhere anarcho-capitalist. All that polycentric law means is that people are oppressed differently in different places — or occasionally well governed.

There are two popular theories concerning the origins of the state: the liberal notion of the social contract, exemplified in the (contract-breaking!) Declaration of Independence; and the libertarian notion of the state as a gang of thugs enslaving an earlier stateless society of inoffensive farmers. Both are false. The origin of the state is quite different: it is the natural authority of the alpha-male; the father figure; the head of the family. The recognition of authority then extends from the immediate family to kin, to tribe, to clan, to race, to nation. In this process, the relation of authority to blood becomes weak; non-relatives may then achieve power, whilst more diverse forms of society also become possible.

Because every human society has been based, one way or another, upon the biological imperative of the family, no society has been stateless. Even in the most primitive peoples we can distinguish the beginnings of the state; a kind of proto-state. For very brief periods territories have sometimes known anarchy, but only when society itself has been in abeyance. No society in history has ever been anarcho-capitalist. Wherever we find an ordered society we also find the state.

This state is not necessarily the modern nation state — this is where those writers I mentioned go wrong — but it has always involved some form of rule backed by the threat of force. Nor is sovereignty necessarily vested in a single central authority; more often than not it is divided amongst rival factions. Courts religious, secular and royal; commercial guilds; trades unions; local councils; houses of parliament; lords spiritual and temporal; the civil service; the King; the Pope; the Emperor: all these and more claim the right to exercise coercive power; and almost always they attempt to usurp more power than is rightfully theirs.

The broader the division of power, the greater the total force available, the more that force is likely to be exercised for coercive ends, and the less the rule of law is likely to be recognised. The naive notion that federation and the separation of powers reduces the propensity for state oppression is fallacious; plurarchy (by which I mean plural rule not rule by the many) may sometimes reduce the scope for oppression by each branch of government, considered severally, yet by multiplying the functions and divisions of the state it radically increases the potential for oppression overall. It is worse to be bullied by two gangs than by only one. Do not be deceived; devolution is not the route to anarcho-capitalism.

3. Why the World Order is not Anarcho-Capitalist

The world consists of competing jurisdictions. So doesn't it follow that the world order is anarcho-capitalist? And since the world order is (reasonably) stable, how is it possible to claim that anarcho-capitalism is not?

The key feature of anarcho-capitalism seems to be this. I am free to choose which court I will appeal to for protection. We needn't get hung up on the question of just how the private police and so forth are to be organised, since we can, without loss of generality, define as a court any particular protection package a client may choose, even if it involves separate firms for the various component functions, such as insurance, arbitration, enforcement, etc..

Suppose I sign up with Acme Insurance, while my next door neighbours pick Joe's Protection Racket and the Salvation Army Police respectively. If I fall out with Acme, I can change to another court without moving house. The courts are not territorial. This is the distinguishing characteristic of anarcho-capitalism. In a sovereign state things are very different; so long as you remain within the territory of the state, you are under the authority of the state court. In an ultraminimal state you are allowed to opt out of the state's protection; but you cannot opt out of its authority. That the mode of access to the state court may sometimes vary from person to person, or even from occupation to occupation, does not obviate this fact.

Territorial jurisdictions compete with each other only weakly. It is difficult and costly to migrate from one country to another, especially where the other country doesn't even use the same language. For practical purposes, most people are stuck with the legal system they're born into. Furthermore, the vast majority of court cases occur within a single jurisdiction; the far more problematical multi-jurisdictional cases are rare.

Even under anarcho-capitalism, if disputes only ever arose between clients of the same court, there would be few difficulties and little likelihood of instability. Unfortunately, if many courts attempt to cover the same territory, it is inevitable that the vast majority of disputes will then be inter-jurisdictional. The potential for conflict is thus enormously enhanced. The restitution ramp is only one example of the many problems that may then arise.

It is in this way that the world order of diverse nations deviates from our model of anarcho-capitalism. The world order is fundamentally territorial. Anarcho-capitalism is not. We should not therefore be astonished if anarcho-capitalism falls prey to instabilities that territorial states hardly notice. I should point out that even nomadic tribes are territorial: it's just that the territory moves with the tribe. I should also point out that although territories are almost invariably geographical in nature, this is not true of absolute logical necessity: territories are like ecological or economic niches, a way of minimising direct competition, or of restricting such competition to the (behavioural) boundaries. Anarcho-capitalist courts are required to compete head-on; and this is why anarcho-capitalism is unstable.

4. Territoriality within Anarcho-Capitalism

But are the anarcho-capitalist courts forced into direct competition? If they try and cover the same ground, yes! But such competition is unstable, and the market will shake it out.

Consider the small anarcho-capitalist town in which I (hypothetically) live; the protection business is currently shared among three courts (Acme, Joe's and the local branch of the Salvation Army Police). What will happen?

Let's simplify things a bit more by considering the two local courts only. Assume that they are in direct competition; ie, that they are attempting to supply the same market. Now suppose that Acme Insurance happens to have a larger share of the market than Joe's. Then a larger proportion of Acme's cases will be internal (disputes between two Acme clients) and a smaller proportion inter-jurisdictional. Other things being equal, Acme's average costs should thus be lower. So Acme can cut prices, drawing customers away from Joe's. This reduces Acme's average costs still further, while simultaneously increasing Joe's. Acme cuts prices again and again, each cut improving Acme's relative profitability by more than it costs. In the absence of any other reason for supporting Joe's over Acme, Joe's will suffer a hypertransfer of all its business to Acme. Even if we start with the courts at an equilibrium of market share, not necessarily fifty-fifty, appropriate to their comparative efficiency, friendliness and pricing, the position is unstable, like a pencil balanced on its point; any movement away from equilibrium is self-reinforcing.

Unless they can avoid direct competition (for example, by specialising in different kinds of dispute), either Acme or Joe's will cease trading. Whereas if they specialise sufficiently for both to survive together, neither one alone will qualify as a court, since people will have to take out policies with both of them in order to be fully covered. Either way, the result is an effective monopoly of justice locally.

What of the third player, the local branch of the SAP? If it is merely a franchise, then it is on much the same footing as Acme and Joe's. But if it is an integral part of the parent organisation, drawing its funds from that source, it will not be not subject to immediate hypertransfer against local firms, because most of its market is safe from their predation; it is not in direct competition. However, it will still be subject to hypertransfer vis-à-vis other organisations whose catchment areas largely overlap; therefore we can expect this wider market to end up under a monopoly too.

It is also possible for such an organisation to suffer hypertransfer from the universe of local firms. Let me explain. Suppose the organisation gets some bad publicity and as a result temporarily loses market share; every branch will be affected similarly, and every competitor will get a similar boost. Thus the whole organisation loses profitability, while every local competitor gains. The result is hypertransfer away from the organisation (which folds) to the dominant (local) firm in each locality. Hypertransfer in the other direction is also possible.

Thus we are likely to end up with one of two outcomes. Either a single court monopolises the entire territory; or the dominant courts divide the territory into monopolistic domains. Intermediate and more complex arrangements along the same lines are also possible; such as where the most powerful court controls most of the territory, smaller but still powerful courts corner other large chunks, and the rest comprises the domains of the local courts. The point to note is that in each locality there obtains a monopoly; competition persists only along the domain boundaries.

I have described this process specifically in terms of the efficiency gain from internalising disputes. I could also have described it in terms of the results of a restitution war, or of other destabilising influences. The details don't really matter. The key point is that there is a strong tendency for the anarcho-capitalist courts to become territorial. Indeed it is hard to see how this could ever be avoided; only by becoming territorially dominant do they avoid direct competition; and only thus do they escape radical instability.

5. Territorial Courts are States

A court having a natural monopoly (or near monopoly) of business within a region is a de facto state. There is no one to whom you can appeal against its judgements — or to be more precise, no one who is likely to be willing and able to defend you against its judgements. If you don't like it, all you can do is leave its territory — the region wherein it is effectively sovereign. You can no longer change protector without moving house; so the distinguishing characteristic of anarcho-capitalism is absent.

If a single court gains the monopoly of the entire anarcho-capitalist region this is merely a failed attempt at anarcho-capitalism resulting in the re-emergence of the erstwhile state. The more interesting case is where the region becomes divided into local domains. Here we find anarcho-capitalism dissolving into city states.

It is important to realise that with the territorialisation of the courts, even though many courts continue to exist across the region, and even though there may be no single sovereign state, anarcho-capitalism is dead. Each domain defines a mini-state; within each domain, the dominant court is king.

How will these mini-states behave? What will their political structure be? This depends upon the details of their emergence and the internal arrangements of the courts in question. A co-operatively owned court might create a commune, a church court a theocracy, a privately owned court a monarchy, a joint-stock court a republic. The dominant court might be controlled by businessmen, or lawyers, or policemen, or soldiers, or gangsters — or solid citizens. The court might be monolithic, or composite; its powers might be unitary, or divided. It might be pragmatic, or ideological. It might be honest, or corrupt. It might be in it for the money, or for the power, or for the furtherance of justice.

Because these courts have risen to power locally, the outcome will also depend upon local factors. This means that different outcomes can be expected in different places. Thus the political structures of the resulting mini-states will be diverse. Some will be totalitarian, others ultraminimal; most will fall somewhere in between. This is not unlike a small-scale version of the world order.

There are several significant differences. In the first place, the boundaries between domains will not be well-defined; these are in essence not nation states but city states. Competition between courts will persist in the border zones, which will mostly consist of farmland, open countryside, small villages and market towns — but seldom if ever cities large enough to have their own dominant court. In the second place, the existence of a common language means that other domains will not be considered foreign; it will be comparatively easy to migrate from one to another.

6. Freedom in the City States

It cannot be guaranteed that the city states will be free. Some will be — but many will not. Nevertheless, there is reason to hope that freedom will become the norm. The reason lies in the residual competition between the courts or cities, a form of competitive constraint missing or greatly weakened in the larger world order.

Most people would rather live in a free society than in an unfree one — at least unless they could be sure of being the rulers not the ruled. So people will tend to migrate to the free cities. Remember that they don't have to face a new culture or a new language; they're not aliens; the open cities will welcome them. Consequently, oppressive regimes will lose much of their population; but free cities will grow.

It goes further. Free-market capitalism is economically efficient. Socialism is not. Those cities that defend property rights, the rule of law and the free market — thereby encouraging high levels of saving and capital investment — will be rewarded with rampant economic growth. Minimal and ultraminimal city states will grow fastest of all. Socialist states will stagnate. Population flight will become a flood.

Domains will discover that they must either liberalise or lose most of what remains of their populace; even free republics will not be immune, so long as they fall short of economic laissez faire; the freest economies set the pace.

This doesn't mean that every city will become free; there will be enough masochists and ideologues left over to supply a fair number of political, democratic, welfare, egalitarian, socialist, communist, communitarian, fascist, syndicalist, religious and aristocratic mini-states — which will no doubt fight each other and cause trouble for the free states too.

Nevertheless, it seems reasonable to expect that most of the city states will be free — variations on the theme of the minimal or ultraminimal or limited or night-watchman state. Their courts will be just, enforcing the common law. They will provide cost-free restitutional justice. They will adhere to the rule of law.

Or mostly they will. Because the domains are less than nations there may be some room for contractual manoeuvre — citizens freely accepting certain constraints that would violate their rights if imposed unilaterally. But the departures from uniform justice will necessarily be slight, though more restrictive non-sovereign voluntary communities may exist within the courts' domains.

Might this scenario prove too optimistic? Perhaps. It hangs on the question of whether enough of the dominant courts will be just, or their people genuinely freedom-loving; the precedents are not encouraging. Moreover, any new laissez faire state will find itself at a short-term disadvantage, because it is a threat to every other form of government and vested interest, who may therefore combine to destroy or corrupt it. Nevertheless, circumstances are more propitious here than in the historical examples of the break-up of an empire, the balkanisation of a nation state, or the colonisation of a new land: by hypothesis, these city states are those naturally arising out of anarcho-capitalism (itself set up in imagination by a wave of the experimenter's magic wand); it is hard not to see at least some of them succeeding in the creation of just and robust common law regimes.

There is thus reason to believe that the marginal competition from other domains will not only drive the city states towards freedom, but will help to stabilise that freedom against the natural counter-tendency of government power to expand coercively and the danger of popular support for oppressive policies and ideologies.

7. Anarcho-Capitalism in No-man's-land?

So anarcho-capitalism has dissolved into city states. But what of the border regions? Will they be consolidated into the now-sovereign domains, or will they remain apart? Perhaps a bit of both. No doubt some of the city states will negotiate fixed boundaries between them, just like today's nation states; but there are also advantages in leaving areas of no-man's-land between them as buffers. Free states in particular should have no objection in principle to the continuing existence of unincorporated territory.

In these interstices between the sovereign city states, no court has a monopoly of justice; inhabitants may choose from among the courts offering their services there. Most of these courts will be based in the sovereign territories of the city states; indeed, they will mostly derive from the state courts themselves; but small local courts may also exist, as well as small-scale mutual defence associations among traders or farmers, and some of the inhabitants may be pure independents.

Is this stable? If so, why is it not liable to the instabilities previously noted? Several points come to mind. First, competition between the major courts is weak; operations in the borderlands are peripheral to their main business within their own territories and cannot seriously threaten their survival, because inter-jurisdictional disputes comprise only a small part of the total caseload; they can afford to take a fairly relaxed attitude to profitability in the provinces. Second, it is in the interests of neighbouring states to maintain a rough balance of market share; if any state court inadvertently gained a monopoly or position of dominance in the borderlands it would yield comparatively little extra profit, while exposing the state to the danger of direct confrontation with its rivals. Furthermore, since courts will not wish to appear to espouse double standards — this would undermine their position — the ways they can compete beyond their territorial boundaries will be limited; blatant predation, such as offering super-restitution, will be ruled out.

So is this anarcho-capitalism? It certainly doesn't match the usual utopian picture, but it has a lot in common with it. We may say that it is a system like anarcho-capitalism, but one stabilised from without. It relies upon a patchwork of free city states to keep it in being. It could only ever be available to a small part of the population; try to extend it and it becomes radically unstable. We might call it exogenous anarcho-capitalism, to distinguish it from endogenous anarcho-capitalism or anarcho-capitalism proper.

Yet if it is not true anarcho-capitalism it should surely satisfy everyone except the extreme anti-state fanatics — who surely need to learn that their hatred of the state is not universal. It is a poor liberty that will not allow people to live in the kind of society they prefer — whether monarchy, republic or anarchy — but demands the downfall of every state across the world. Indeed, it is no liberty — it is ideology.

To learn that the freedom of some to live life without the state may itself be predicated upon the very existence of diverse states populated by many other free persons may be galling to the radical anarchist. Tough. In a free world, people must be free to live in ways you do not like, even in ways that seem to you to restrict their freedom; for freedom must be freely chosen; and where freedom is freely chosen it can also be freely rejected. The greatest freedom exists where no option is finally denied, where every route is sometimes travelled, and where all possible journeys are eventually undertaken.

© Paul Birch, 9th Jan. 1998.