DemocracyThe God That Failed

by Hans-Hermann Hoppe

  IV

  Having reconstructed the myth of collective security—the myth of the state—and criticized it on theoretical and empirical grounds, I now must take on the task of constructing the positive case for private security and protection. In order to dispel the myth of collective security, it is not just sufficient to grasp the error involved in the idea of a protective state. It is just as important, if not more so, to gain a clear understanding of how the nonstatist security alternative would effectively work. Rothbard, building on the pathbreaking analysis of the French-Belgian economist Gustave de Molinari,9 has given us a sketch of the workings of a free-market system of protection and defense.10 As well, we are in debt to Morris and Linda Tannehill for their brilliant insights and analyses in this regard.11 Following their lead, I will proceed with my analysis and provide a more comprehensive view of the alternative—nonstatist—system of security production and its ability to handle attacks, not just by individuals or gangs but in particular also by states.

  Widespread agreement exists among liberal-libertarians such as Molinari, Rothbard, and the Tannehills as well as most other commentators on the matter that defense is a form of insurance, and defense expenditures represent a sort of insurance premium (price). Accordingly, as Rothbard and the Tannehills in particular would emphasize, within the framework of a complex modern economy based on worldwide division of labor, the most likely candidates to offer protection and defense services are insurance agencies. The better the protection of insured property, the lower are the damage claims and hence an insurer's costs. Thus, to provide efficient protection appears to be in every insurer's own financial interest. Indeed, although restricted and hampered by the state, even now insurance agencies provide wideranging services of protection and indemnification (compensation) to injured private parties. Insurance companies fulfill a second essential requirement. Obviously, anyone offering protection services must appear able to deliver on his promises in order to find clients. That is, he must possess the economic means—the manpower as well as the physical resources—necessary to accomplish the task of dealing with the dangers, actual or imagined, of the real world. On this count insurance agencies appear to be perfect candidates, too. They operate on a nationwide and even international scale, and they own large property holdings dispersed over wide territories and beyond single state boundaries. Accordingly, they have a manifest self-interest in effective protection, and are 'big' and economically powerful. Furthermore, all insurance companies are connected through a network of contractual agreements of mutual assistance and arbitration as well as a system of international reinsurance agencies, representing a combined economic power which dwarfs that of most existing governments.

  9Gustave de Molinari, The Production of Security (New York: Center for Libertarian Studies, 1977).

  10Rothbard, Power and Market, chap. 1; idem, For A New Liberty (New York: Collier, 1978), chaps. 12 and 14.

  11 Morris Tannehill and Linda Tannehill, The Market for Liberty (New York: Laissez Faire Books, 1984), esp. part 2.

  Let me further analyze and systematically clarify this suggestion: that protection and defense are 'insurance' and can be provided by insurance agencies. To reach this goal, two issues must be addressed. First off, it is not possible to insure oneself against every risk of life. I cannot insure myself against committing suicide, for instance, or against burning down my own house, becoming unemployed, not feeling like getting out of bed in the morning, or not suffering entrepreneurial losses, because in each case I have full or partial control over the likelihood of the respective outcome. Risks such as these must be assumed individually. No one but I can possibly deal with them. Hence, the first question must be what makes protection and defense an insurable rather than an uninsurable risk? After all, as we have just seen, this is not self-evident. In fact, does not everyone have considerable control over the likelihood of an attack on and invasion of his person and property? Do I not deliberately bring about an attack by assaulting or provoking someone else, for instance, and is not protection then an uninsurable risk, like suicide or unemployment, for which each person must assume sole responsibility?

  The answer is a qualified yes and no. Yes, insofar as no one can possibly offer unconditional protection, i.e., insurance against any invasion whatsoever. That is, unconditional protection can only be provided, if at all, by each individual on his own and for himself. But the answer is no, insofar as conditional protection is concerned. Only attacks and invasions that are provoked by the victim cannot be insured. Unprovoked and thus 'accidental' attacks can be insured against, however.12 That is, protection becomes an insurable good only if and insofar as an insurance agent contractually restricts the actions of the insured so as to exclude every possible 'provocation' on their part. Various insurance companies may differ with respect to the specific definition of provocation, but there can be no difference between insurers with regard to the principle that everyone must systematically exclude (prohibit) all provocative and aggressive action among its own clients.

  As elementary as this first insight into the essentially defensive—nonaggressive and nonprovocative—nature of protection-insurance may seem, it is of fundamental importance. For one, it implies that any known aggressor and provocateur would be unable to find an insurer, and hence, would be economically isolated, weak and vulnerable. On the other hand, it implies that anyone wanting more protection than that afforded by self-reliant self-defense could do so only if and insofar as he submitted himself to specified norms of nonaggressive, civilized conduct. Further, the greater the number of insured people—and in a modern exchange economy most people want more than just self-defense for their protection—the greater would be the economic pressure on the remaining uninsured to adopt the same or similar standards of nonaggressive social conduct. Moreover, as the result of competition between insurers for voluntarily paying clients, a tendency toward falling prices per insured property values would come about.

  At the same time, a system of competing insurers would have a twofold impact on the development of law and thus contribute further to reduce conflict. On the one hand, the system would allow for systematically increased variability and flexibility of law. Rather than imposing a uniform set of standards onto everyone (as under statist conditions), insurance agencies could and would compete against each other not just via price but in particular also through product differentiation and development. Insurers could and would differ and distinguish themselves with respect to the behavioral code imposed on and expected of their clients, with respect to rules of evidence and procedure, and/or with respect to the sort and assignment of awards and punishments. There could and would exist side by side, for instance, Catholic insurers applying Canon law, Jewish insurers applying Mosaic law, Muslims applying Islamic law, and Non-believers applying Secular law of one variant or another, all of them sustained by and vying for a voluntarily paying clientele. Consumers could and would choose, and sometimes change, the law applied to them and their property. That is, no one would be forced to live under "foreign" law; and hence, a prominent source of conflict would be eliminated.

  12On the "logic" of insurance see Ludwig von Mises, Human Action: A Treatise on Economics (Chicago: Regnery, 1966), chap. 6; Murray N. Rothbard, Man, Economy, and State, 2 vols. (Auburn, Ala.: Ludwig von Mises Institute, 1993), pp. 498ff.; HansHermann Hoppe, "On Certainty and Uncertainty, Or: How Rational Can Our Expectations Be?" Review of Austrian Economics 10, no. 1 (1997); also Richard von Mises, Probability, Statistics and Truth (New York: Dover, 1957); Frank H. Knight, Risk, Uncertainty, and Profit (Chicago: University of Chicago Press, 1971).

  On the other hand, a system of insurers offering competing law codes would promote a tendency toward the unification of law. The "domestic"—Catholic, Jewish, Roman, Germanic, etc.—law would apply and be binding only on the persons and properties of the insured, the insurer, and all others insured by the same insurer under the same law. Canon law, for instance, would apply only to prof
essed Catholics and deal solely with intra-Catholic conflict and conflict resolution. Yet it would also be possible for a Catholic to interact, come into conflict with, and wish to be protected from the subscribers of other law codes, e.g., a Muslim. From this no difficulty would arise so long as Catholic and Islamic law reached the same or a similar conclusion regarding the case and contenders at hand. But if competing law codes arrive at distinctly different conclusions (as they would in at least some cases by virtue of the fact that they represent different law codes) a problem would arise. The insured would want to be protected against the contingency of intergroup conflict, too, but "domestic" (intragroup) law would be of no avail in this regard. In fact, at a minimum two distinct "domestic" law codes would be involved, and they would come to different conclusions. In such a situation it could not be expected that one insurer and the subscribers of his law code, say the Catholics, would simply subordinate their judgment to that of another insurer and his law, say that of the Muslims, or vice versa. Rather, each insurer—Catholic and Muslim alike—would have to contribute to the development of intergroup law, i.e., law applicable in cases of disagreement among competing insurers and law codes. And because the intergroup law provisions that an insurer offered to its clients could appear credible to them, and hence a good, only if and insofar as the same provisions were also accepted by other insurers (and the more of them, the better), competition would promote the development and refinement of a body of law that incorporated the widest—intergroup, cross-cultural, etc.—legal-moral consensus and agreement and thus represented the greatest common denominator among various competing law codes.13

  More specifically, because competing insurers and law codes could and would disagree regarding the merit of at least some of the cases brought jointly before them, every insurer would be compelled to submit itself and its clients in these cases from the outset to arbitration by an independent third party. This third party would not just be independent of the two disagreeing parties, however. It would at the same time be the unanimous choice of both parties. And as objects of unanimous choice, arbitrators then would represent or even personify "consensus" and "agreeability." They would be agreed upon because of their commonly perceived ability of finding and formulating mutually agreeable, i.e., "fair," solutions in cases of intergroup disagreement. Moreover, if an arbitrator failed in this task and arrived at conclusions that were perceived as "unfair" or "biased" by either one of the insurers and/or their clients, this person would not likely be chosen again as an arbitrator in the future.

  Consequently, protection and security contracts would come into existence as the first fundamental result of competition between insurers for a voluntarily paying clientele. Insurers (unlike states) would offer their clients contracts with well-specified property and product descriptions and clearly defined and delineated duties and obligations. Likewise, the relationship between insurers and arbitrators would be defined and governed by contract. Each party to a contract, for the duration or until fulfillment of the contract, would be bound by its terms and conditions; and every change in the terms or conditions of a contract would require the unanimous consent of all parties concerned. That is, under competition (unlike under statist conditions), no "legislation" would or could exist. No insurer could get away (as a state can) with "promising" its clients "protection" without letting them know how or at what price, and insisting that it could, if it so desired, unilaterally change the terms and conditions of the protector-client relationship. Insurance-clients would demand something significantly "better," and insurers would comply and supply contracts and constant law, instead of promises and shifting and changing legislation. Furthermore, as a result of the continual cooperation of various insurers and arbitrators a tendency toward the unification of property and contract law and the harmonization of the rules of procedure, evidence and conflict resolution (including such questions as liability, tort, compensation, and punishment) would be set in motion. On account of buying protectioninsurance, everyone would become tied into a global competitive enterprise of striving to reduce conflict and enhance security. Moreover, every single conflict and damage claim, regardless where and by or against whom, would fall into the jurisdiction of one or more specific insurance agencies and would be handled either by an individual insurer's "domestic" law or by the "international" law provisions and procedures agreed upon in advance by a group of insurers, thus assuring (ex ante) complete and perfect legal stability and certainty.

  13See on this Hans-Hermann Hoppe, Eigentum, Anarchie und Staat (Opladen: WestdeutscherVerlag, 1987), pp. 122-26.

  V

  Now a second question must be addressed. Even if the status of defensive protection as an insurable good is granted, distinctly different forms of insurance exist. Let us consider just two characteristic examples: insurance against natural disasters, such as earthquakes, floods, hurricanes, and insurance against industrial accidents or disasters, such as malfunctions, explosions, and defective products. The former can serve as an example of group or mutual insurance. Some territories are more prone to natural disasters than others; accordingly, the demand for and price of insurance will be higher in some areas than others. However, every location within certain territorial borders is regarded by the insurer as homogeneous with respect to the risk concerned. The insurer presumably knows the frequency and extent of the event in question for the region as a whole, but he knows nothing about the particular risk of any specific location within the territory. In this case, every insured person will pay the same premium per insured value, and the premiums collected in one time period will presumably be sufficient to cover all damage claims during the same time period (otherwise the insurance industry will incur losses). Thus, the particular individual risks are pooled and insured mutually.

  In contrast, industrial insurance can serve as an example of individual insurance. Unlike natural disasters, the insured risk is the outcome of human action, i.e., of production efforts. Every production process is under the control of an individual producer. No producer intends to fail or experience a disaster, and as we have seen only accidental—non-intended—disasters are insurable. Yet even if production is largely controlled and generally successful, every producer and production technology is subject to occasional mishaps and accidents beyond his control—a margin of error. However, since it is the outcome (intended or not) of individual production efforts and production techniques, this risk of industrial accidents is essentially different from one producer and production process to another. Accordingly, the risk of different producers and production technologies cannot be pooled, and every producer must be insured individually. In this case, the insurer will have to know the frequency of the questionable event over time, but he cannot know the likelihood of the event at any specific point in time, except that at all times the same producer and production technology are in operation. There is no presumption that the premiums collected during any given period will be sufficient to cover all damage claims arising during that period. Rather, the profit-making presumption is that all premiums collected over many time periods will be sufficient to cover all claims during the same multi-period time span. Consequently, in this case an insurer must hold capital reserves in order to fulfill its contractual obligation, and in calculating his premiums he must take the present value of these reserves into account.

  The second question is what kind of insurance can protect against aggression and invasion by other actors? Can it be provided as group insurance, as for natural disasters, or must it be offered in the form of individual insurance, as in the case of industrial accidents?

  Note that both forms of insurance represent only the two possible extremes of a continuum, and that the position of any particular risk on this continuum is not definitively fixed. Owing to scientific and technological advances in metereology, geology, or engineering, for instance, risks that were formerly regarded as homogeneous (allowing for mutual insurance) can become more and more dehomogenized. Noteworthy is
this tendency in the field of medical and health insurance. With the advances of genetics and genetic engineering—genetic fingerprinting—medical and health risks previously regarded as homogeneous (unspecific) with respect to large numbers of people have become increasingly more specific and heterogeneous.

  With this in mind, can anything specific be said about protection insurance in particular? I would think so. After all, while all insurance requires that the risk be accidental from the standpoint of the insurer and the insured, the accident of an aggressive invasion is distinctly different from that of natural or industrial disasters. Whereas natural disasters and industrial accidents are the outcome of natural forces and the operation of laws of nature, aggression is the outcome of human actions; and whereas nature is 'blind' and does not discriminate between individuals, whether at the same point in time or over time, an aggressor can discriminate and deliberately target specific victims and choose the timing of his attack.

  VI

  Let me first contrast defense-protection insurance with that against natural disasters. Frequently an analogy between the two is drawn, and it is instructive to examine if or to what extent it holds. The analogy is that just as every individual within certain geographical regions is threatened by the same risk of earthquakes, floods, or hurricanes, so does every inhabitant within the territory of the U.S. or Germany, for instance, face the same risk of being victimized by a foreign attack. Some superficial similarity—to which I shall come shortly—notwithstanding, it is easy to recognize two fundamental shortcomings in the analogy. For one, the borders of earthquake, flood, or hurricane regions are established according to objective physical criteria and hence can be referred to as 'natural.' In distinct contrast, political boundaries are 'artificial' boundaries. The borders of the U.S. changed throughout the entire nineteenth century, and Germany did not exist as such until 1871 and was composed of thirty-eight separate countries. Surely, no one would want to claim that this redrawing of the U.S. or German borders was the outcome of the discovery that the security risk of every American or German within the greater U.S. or Germany was, contrary to the previously held opposite belief, homogeneous (identical).