(FORTUNE Magazine) – Macartan Humphreys has a gig teaching game theory at Columbia University, but you're just as likely to find him in a dingy drinking hole in Sierra Leone, or Colombia, or any number of other war-ravaged nations, scribbling notes while listening to rebel leaders and financiers tell their tales. "What I do is often the nearest thing to market research in these places," says the 32-year-old Irishman, who studies the impact of war on local economies. One of his theories: In war zones marketplace collusion among factions often reduces violence against civilians and boosts economic returns. "In peacetime enforcing competition is good for economies," he says, "but in wartime collaboration often works better."

Like a growing number of political scientists and economists who work at the intersection of war and economics, Humphreys has focused lately on Iraq and Afghanistan. Those conflicts are in many ways unique, he says: "They have elements of colonial occupations, civil wars, and failed states."

Iraq's once-restricted retail sector has exploded with competition since the fall of Saddam Hussein, says Humphreys, and consumers are benefiting from lower prices. Yet with chaos worsening in some cities, the new free-market climate may not last. "Violence is already reregulating markets," he says. He points to Islamists in Basra, who have reportedly defied the coalition's authority and have begun regulating the sale of liquor by assassinating anyone who tries to sell it.

In peacetime, says Humphreys, a capitalist state is supposed to have a monopoly on the use of force while ensuring that businesses are kept from having monopolies in markets. "In wartime the system often gets mixed up," he says. "Whoever controls violence in a particular area can establish monopolies."

In Sierra Leone, for example, where Humphreys has surveyed hundreds of ex-combatants, rebels in the mid-1990s seized control of a key diamond-mining area. The government brought in Executive Outcomes, a now defunct South African security firm run by soldiers of fortune, with its own fleet of combat helicopters. The government, too impoverished to afford the company's hefty rates, instead worked out an arrangement whereby a Canadian mining company related to the mercenaries was promised monopoly rights to the richest part of the field for 25 years. The mercenaries quickly drove out the rebels.

U.S. military interventions also tend to foster monopolies, both deliberate and inadvertent, says Humphreys. The Bush administration has tilted markets in Iraq by allowing only coalition partners to bid on prime contracts. It has also kept Iraq's insurance market closed, thereby benefiting American firms, which have coverage via the U.S. government. Yet even without those actions, the inherent risk of doing business in a war zone tends to kill economic competition. Low-wage South Korean construction firms used to do well in Iraq, for example, but have largely withdrawn because they lack the ability to protect their workers from attacks. Some U.S. contractors are also starting to pull out because of safety concerns.

But as long as U.S.-led forces maintain dominance, the companies that operate in cartels with close relations to the military can expect to have a tremendous economic advantage. The reason, says Humphreys, is not just that in war governments can't prevent monopolies, but that they actually prefer them. Governments don't want bidders running around battlefields, perhaps perusing secret plans that may later find their way into enemy hands. Open bidding and the wish to minimize costs take second place to trust and getting the job done. Size matters too: The bigger the contractor, the better it may handle surge requirements crucial to both warfare and rapid reconstruction. That's the case in Iraq with the Halliburton subsidiary Kellogg Brown & Root, which brings to bear vast experience and capacity both in providing services on the battlefront and in extracting natural resources.

Is this the sort of out-of-control military-industrial complex that critics of capitalism have long warned of? With the open-ended war on terror, some have been dusting off copies of Vladimir Lenin's Imperialism: The Highest Stage of Capitalism, which foretells of corporations and capitalist armies conspiring to create monopolies around the globe. Lenin's references included the old British East India Company, to which the colonial government granted a monopoly in both trade and the use of force. Today there is evidence of "new pathways for private interests to affect U.S. foreign policy," says George Washington University's Deborah Avant. Fears that those interests are accountable to no one will be soon be tested, now that military contractors in Iraq have been accused of abusing prisoners in military custody. The contractors cannot be tried under U.S. military law, and according to Human Rights Watch legal advisor James Ross, U.S. federal law remains untested on the issue.

But from where Humphreys stands, there are very different concerns. Most of the world's violent conflicts are perpetrated not by military-industrial complexes but by warlords. In war zones like Iraq and Afghanistan, critics should worry not just that the monopolists will take control, but that no one will--that society will splinter into warring factions. Humphreys's research suggests that that is when business activity is apt to make matters worse. "In the presence of military dominance, private-sector interests often have a limited impact. In the absence of military control and in cases where attempts at political settlements fail, then we see private-sector activities aggravating factional divisions and violence." He cites evidence in Angola, Chad, Congo, and other countries of the sale of what UCLA political scientist Michael Ross calls "booty futures." They are deals in which companies finance warring factions in exchange for rights to the spoils of civil wars.

Humphreys, whom the United Nations Millennium Project has asked to suggest ways that economic activity in conflict zones might ease violence, is likely to recommend that public companies operating in unstable, unregulated regions be forced to report such payments or face delisting from stock markets. The object is to block them from partaking in such deals. He is also developing measures that would discourage private companies from profiting from conflicts. Humphreys is hopeful that such schemes, which are being promoted by George Soros and others, will gain support among U.S. corporations because they level the playing field. "Currently U.S. firms are competing against firms from countries that are less constrained by domestic pressures from human rights groups," he says. This summer he will head to Africa's Gulf of Guinea with Columbia University's Earth Institute to advise governments on ensuring that their new oil wealth does not touch off civil conflicts.

In the end, says Humphreys, few companies prefer war. While a handful may benefit from conflict and occupation, "for most, the optimal operating conditions are ones in which security is not an issue. Even if they obtain monopolies during conflict, it is still often in their interest to see conflicts end." In short, peace brings economic benefits. After the South African government negotiated a peace agreement in Congo, for instance, it also negotiated lucrative contracts for South African companies to mine the country's gold, copper, and cobalt. Likewise, U.S. efforts to broker peace in Sudan are expected to pave the way for the return of American oil companies.

The big puzzle in Iraq and elsewhere, Humphreys says, is how to restructure monopolistic war economies after wars end. "In postwar societies, reintroducing economic competition is always risky. It can lead to cutthroat politics, which can lead back to violence again."