VoxEU Column Frontiers of economic research Politics and economics

The diplomacy of arms: Conflict as a negotiation instrument

Are conflicts worth it? This column argues that they can be. While wars are extremely damaging, they can be in the interests of one party if they help reveal the true balance of power and thereby change the stakes in eventual negotiations. This explains why small countries take on superpowers with no chance of winning and why unions go on strike against laws already passed.

Why do people fight? Why do countries, firms, unions, and individuals engage in costly confrontations? One line of argument states that humans enjoy fighting per se, because it is in a fundamental part of our violent and dark nature. A more appealing explanation is that people fight to get something out of it.

Sociologists like Gilles Lipovetsky have argued that people in pre-industrial societies fought wars or took revenge on others as part of the socialisation process. Being violent meant that you belonged to the group, that you accepted its norms, usually at the expense of those outside of the group such as other villages or nations or perhaps a fellow member of society who offended your honour – hence duels.

Conflict as a negotiation instrument

Economists, on the other hand, tend to believe that people fight to attain more material objectives. Nations fight to obtain territories, firms fight price wars to get customers and unions strike in order to obtain better wages. Powerful as it is, this view leaves a big open question. Conflicts are extremely costly. People die, resources are consumed, and staggering amounts of time and money are wasted. Why is conflict so pervasive then? Here I will discuss one explanation to this puzzle. I argue that people fight in order to alter the beliefs of their opponents about the existing balance of power between them. In doing so, they obtain a better deal in the negotiations that follow. In this sense, they use conflict as a negotiation instrument.

This is a simple yet powerful idea that applies to a wide range of contexts, from industrial action against corporations and interstate wars over natural resources to legal disputes over compensations or couple arguments about where to spend the next summer vacation. Still, it is not until very recently that this explanation has been seriously and systematically studied by scholars, especially economists.

Yet the underlying idea is nothing new, it has been around for the last two centuries. Nineteenth-century Prussian military theorist Carl von Clausewitz famously asserted that “War is the continuation of political activity by other means” (von Clausewitz 1832). He was the first person to notice that war and peace are not so distinct phenomena and that conflict can be indeed a tool of diplomacy. Almost seventy years later, the great German sociologist Georg Simmel observed that power in conflict is so difficult to measure that the only really accurate way to assess it is through conflict itself (Simmel 1904). Again seventy years later, it was the iconoclast military historian Geoffrey Blainey who in his now classic book The Causes of War (1973) added another piece to this puzzle. His main claim was that overconfidence is the factor that most often has precipitated war throughout history. It is optimism about their own military capabilities that makes kings and statesmen go to war. When conflict takes place, it reveals information about the true balance of power, it becomes the “stinging ice of reality” that forces parties to make more realistic estimates of their actual relative strength. And, rather paradoxically, lead to peace.

Can fighting be a good thing?

This is a provocative idea. Can conflict be “good”? The answer is “It may”. When there exists imperfect information about key elements of the negotiation environment – such as the amount of arms, evidence on a legal case, or the actual profitability of a firm – and parties are sufficiently optimistic about their prospects in case of conflict – an all-out war, a trial, a strike – a peaceful agreement becomes impossible. Optimism makes their demands incompatible. In those circumstances, the existence of limited confrontations – military skirmishes, discovery procedures, holdouts – can help parties to avoid a more costly confrontation. The outcome of such limited conflicts can help them to better assess their strengths, reduce optimism, and be conducive to an agreement. As a matter of fact, around two thirds of interstate wars end up in a negotiated settlement rather than in the collapse of one of the sides. In that sense, conflict and peace are not opposites. They are rather inextricably linked.

On the other hand, limited conflicts may become “too frequent”. They can delay agreement and create unnecessary losses. The reason is simple. If you know that by fighting a skirmish you may convince your opponent that you are stronger than he previously thought, you may want to take the gamble. If you win, you will reduce his demands and obtain an advantage in the negotiation table. This was the case of the Verdun battle in World War I, an offensive designed specifically by the Germans to alter French estimates about the balance of power. A more recent example took place in Spain last June, when unions, unhappy with the labour market reform just proposed by the government, staged a strike in the public sector as a way to show their strength in case of an eventual general strike. The general strikes across Spain next week (29 September), at a time when the socialist government is undertaking unpopular budget cuts, may reveal a further shift in power still.

It is important to notice here that this use of conflict as a negotiation tool can take place even if demands were compatible and agreement was possible at the outset of the dispute. It may be in the interest of one of the parties to engage in a costly limited confrontation if it believes that by doing so it will extract better terms from the opponent than by negotiating. This may be good for the country, the union, or the partner that initiates the conflict, but it is worse for society as a whole. This is what ultimately explains the pervasiveness of conflict that puzzled economist John Hicks so much in the 1930s, that is, confrontation takes place whenever its returns as a negotiation instrument outweigh those of peaceful diplomacy.

David vs. Goliath: The uneven contenders paradox

Conflict as a negotiation instrument can explain why we often observe small parties fighting much more powerful ones. If capabilities were perfectly known and observable, clashes such as David against Goliath, North Vietnam against the US, or the Taliban against NATO should have never taken place. The explanation to this puzzle, that I call “the uneven contenders paradox” lies in the fact that when incomplete information is a serious issue, it may pay-off for small contenders to fight big ones, not to defeat them completely (it is hard to picture the Taliban aiming to march on Buckingham Palace) but to extract better terms from them. As a matter of fact, the imperial history of colonial powers is full of cases in which they were repelled by supposedly inferior and weak enemies, including the Third Anglo-Afghan war in 1919.

This was precisely the starting point of the empirical study I undertook in my recent article in the Economic Journal (Sánchez-Pagés 2009). Imperial and colonial wars were the perfect example of conflicts where incomplete information was a crucial issue. My hypothesis was that when conflict is used as a negotiating instrument it loses its power as fighting continues. This is because the more the two parties fight the more accurate the estimates of their relative strength become and the lower the extra concession they can get from an extra skirmish. Given that assumption, conflicts should be more likely to end the more they last (this is called “positive duration dependence” in duration analysis). By studying 94 colonial and imperial conflicts between 1817 and 1988, I show that the wars that ended in agreement, and that therefore had a stronger bargaining component, displayed this positive duration dependence whereas those that ended in the total collapse of one of the parties did not.

Isn’t there another way?

The reader may wonder at this point whether conflict as opposed to openness is the only way to avoid further conflict. Why do countries not just reveal their weaponry? Why do unions not just reveal their real resolve? The problem is that such revelations are not credible. Peaceful diplomatic negotiations are plagued with bluffs, double bluffs and uncertainties. The information transmitted on the table is “soft”, that is, is hardly verifiable and easily to manipulate – there are strong incentives to mislead your opponent. US, Iran or China, could perfectly exaggerate their demands based on exaggerated claims on their military capabilities. Unions can inflate their wage demands by misrepresenting their union membership and resolve. On the other hand, information obtained in the battlefield is “hard”. It is easily verifiable and hard to manipulate. While it is often extremely costly, depending on the circumstances it may be better to take the chance and fight before engaging in further peaceful talks.

Lewis Coser, a disciple of Simmel, proposed in the 1950s that countries should make public lists of their military personnel and weaponry as a way to avoid the risk of secrecy leading to a major war. Coser was almost surely too naive in his suggestion. It is often in the interest of each particular nation not to reveal such information. Yet his proposal went in the right direction. Transparency may not be in the interest of the audited country but it is in everybody's interest. That is why the work of arbitrators, international organisations of observers and nuclear inspectors is so incredibly valuable – but only if governments choose to believe their reports.


Blainey, Geoffrey (1973), The Causes of War, New York: Free Press, 1973.

Clausewitz, Claus von (1832), On War, Princeton University Press, 1976.

Sánchez-Pagés, Santiago (2009), “Conflict as a Part of the Bargaining Process”, Economic Journal, 119(539), 1189-1207.

Simmel, Georg (1904), "The Sociology of Conflict I", American Journal of Sociology, 9(4): 490-525.

3,989 Reads