The world is getting smaller. The Internet, cheaptransportation, the spread of free and open markets, and surgingeducation of the masses are steadily eroding the last vestiges ofeconomic autarchy. This increased integration presents afundamental practical challenge to the sovereignty of nations.Policies that are possible in an isolated island state can beimpossible in our new and mostly democratic world of nomadiccapitalists.

This metamorphosis has created heightened demand forinternational cooperation, a demand that has been the midwife tothe birth of organizations that are rapidly becoming a haphazardworld government. At the birth of the United States, AlexanderHamilton wondered whether men "are forever destined to depend fortheir political constitutions on accident and force." Today, it isnot reason, but accident and force that are carving the contours ofthe global political environment.

The feckless and corrupt actions of the United Nations in recentyears provide a case study in how harmful attempts at worldgovernment can be when they go wrong. It is necessary that citizensand leaders of the United States develop a theory of internationalcooperation that can provide a guide to future global associations.This theory must address several specific questions. What areas ofhuman endeavor are likely to require international cooperation inorder to enable efficient outcomes? What types of internationalorganizations should the United States join? When should the UnitedStates be willing to partially cede its sovereignty to suchbodies?

Two strands of economic research provide a natural startingpoint for this discussion. First, the public choice literature hasexamined extensively which services should be offered by localgovernments and which should be offered by national ones. Theextension to world governments is straightforward. Second,specialists in industrial organization have developed a quite broadunderstanding of the functioning of cartels. Coalitions ofcountries often face the same challenges that coalitions of firmsdo. A synthesis of these two literatures provides a valuable guideto the problem of world government.

How Big Should a Country Be?

In one of the great intellectual accomplishments of the 20thcentury, economists Kenneth Arrow and Gerard Debreu set out theconditions under which a free market could be expected to produceoptimal outcomes for individual consumers. Adam Smith's "invisiblehand" of the marketplace was shown--under some conditions--to be afirm and steady force that allocates resources optimally. Theydemonstrated that no reallocation of resources exists that makeseveryone better off than does the allocation that arises as anatural equilibrium of a competitive market.

In these Arrow-Debreu efficient markets, there is a very smallrole for government, which keeps the peace, enforces propertyrights and does little else. If all markets were Arrow-Debreuperfect markets, governments could be very small indeed. We mighthave a federal authority that enforced patents across states,established a minimal set of rules, and allowed state governmentsto provide most government services.

Perhaps the largest contribution of this work, however, was itsillumination of the situations under which larger governments andinternational unions might be necessary. If we identify theassumptions under which markets are perfectly efficient, then wealso have learned when they are not and when they can benefit fromgovernment action.

Markets can fail when citizens require goods that are public innature, or when the production of the private goods they desireproduces significant externalities. A public good is a good that isconsumed by all, without possible exclusion. A textbook example ispublic safety. There is no reason to believe that individuals, ontheir own, would voluntarily contribute enough money to a commonpool to build an army capable of providing real security. Eachindividual would be tempted to free ride, that is, to assume thatothers would provide enough security. Collectively, free-riding canlead to suboptimal provision of public goods. An externality is aslightly different beast. If a steel factory produces pollutionthat harms agricultural output, then that pollution is anexternality. The steel producer, on his own, might ignore thesocial harm from the pollution and produce far more than issocially optimal, and regulation of pollution by government mayimprove social welfare.

There are countless examples of public goods that we require andexternalities that we must monitor. These provide a rationale for alarger and more aggregate government than would otherwise emerge ina free society.

The specific individual characteristics of our desired publicgoods have a significant impact on the optimal level of geographicaggregation. There is little question, to take an early example,that the desire to create a federal force large enough to fight theBritish provided a strong impetus for the unification of the statesat the Founding. The public good of "safety" required coordinationbetween the individual states at the broadest possible level. Totake a more recent example, allowing the social safety net to varyacross states might set off a destructive race to the bottom.States that offer relatively more generous welfare benefits mightfind themselves inundated with migrant paupers. If society viewsthe provision of a safety net as important, policy coordinationbetween the states may be necessary.

In order to provide for services that require a centralgovernment, nations have established federal authorities withpowers of taxation and enforcement. These powers can lead, overtime, to a significant evolution in the focus of the governmentaway from its original design. Elected officials may be tempted topursue actions, such as subsidizing farm production, that do notnecessarily focus on public goods or externalities.

As government's reach evolves, the problem of federalism becomesa tradeoff for those who might choose to join. On the one hand,allowing oneself to be absorbed into a larger whole can increasesecurity and provide access to valuable public goods that smallerentities might not be able to produce. On the other hand, joiningthe union forces one to harmonize undesirable policies withcitizens who may have significantly different tastes for governmentintrusion. This conflict between coordination and harmonization, asHarvard economist Alberto Alesina has recently demonstrated inseveral important studies, is the essence of the economic problemof modern government.

The problem was recognized long before Alesina, however. Indeed,a series of works by my colleagues Christopher DeMuth and MichaelGreve have documented the debate over these issues that occurred atthe birth of our nation. They have argued that our forefathers'preferred design of a federal system allowed for ample competitionbetween the individual states, and free movement of citizensbetween the states, for a good reason. Should a state pursuepolicies that are harmful to overall welfare, its citizens couldpick up and move to a neighbor.

The federal system set up a competition between states thatacted as a natural governor of the growth of harmful government.This political insight was later incorporated into the economiclexicon by economist Charles Tiebout, who demonstrated that "votingwith your feet" often leads to efficient provision ofgovernment.

As the United States turns its gaze outward to the worldcommunity, therefore, it must entertain the possibility thatmembership in an aggregated global governing body may be necessaryin order to ensure that global public goods are adequatelyprovided. But it must weigh the benefit of improved coordination inthe provision of these public goods with the potential costs ofharmonizing policies with nations that have far different socialpreferences. And the citizens of the United States should be waryof entering into compacts that limit the beneficial competitionthat preserves liberty and constrains the growth of inefficient andintrusive government. It is essential, then, that we identify whencompetition between states leads to better outcomes, and when itdoes the reverse.

Government or Cartel?

In the classic Arrow-Debreu world, products are produced bycountless identical and competitive firms. From the point of viewof consumers, the competition is beneficial, since it drives theprice of any product down to its cost. From the point of view ofthe firms, however, the competition is an annoyance, since itlimits profits. Accordingly, firms may recognize that anassociation between all of them might lead to much higher prices.This association is a "cartel", of which OPEC is the leadingpractical example. From the point of view of the firm, a cartel isa wonderful, profit-enhancing device. From the point of view of theconsumer, it is a terrible, price-raising conspiracy.

Economists have learned that cartels can be very unstableentities. If all OPEC countries, for example, agree to reduce theiroutput sharply and drive the price of oil up to $80 per barrel,then the enterprising country that cheats and produces more thanits quota can reap tremendous financial benefits. Accordingly,mechanisms for enforcing cartel discipline are important forensuring cartel survival.

In the early days of cartels, discipline was often imposed byagglomeration. The German chemical firm I.G. Farben was formed in1925 when the eight members of a chemical cartel decided it waseasier to merge. When agglomeration is impossible, as happenedafter anti-cartel legislation was passed in most Western nations, aprecondition for success has often been the existence of a clearlydominant member. For example, many observers believe OPEC hasremained successful because Saudi Arabia has such a large share ofproven world oil reserves that it can alter the world price of oilwith its own behavior when it so chooses.

For governments, international bodies serve the same role as thecartel. They allow countries to establish rules that harmonizegovernment actions and eliminate competition. From the point ofview of the citizens of these nations, however, a key questionarises: When governments form a cartel, should a citizen think ofhimself as a member of that cartel or as a consumer of the productproduced by it?

The answer depends on whether the competition between states inthe instance cited can be expected to lead to a better or worseoutcome for the individual. In terms of the provision of publicgoods and coordination of externalities--the key economicrationales for government--the question clearly depends on thescale of the public good or externality in question. Nations cancompete over the provision of public goods that are efficientlyproduced at the national level. If a particular country'sgovernment establishes a wonderful environment filled withefficient public infrastructure, then activity will migrate to thatcountry to the detriment of those with wasteful governments.