Having discussed the effects of conflict on trade in my previous post, what can we say about the implications of international trade for conflict?
The optimistic case for commerce reducing the likelihood of war can be traced back to Montesquieu ('peace is a natural effect of trade'), Kant ('The spirit of commerce, which is incompatible with war, sooner or later gains the upper hand in every state') and JS Mill ('It is commerce which is rapidly rendering war obsolete'). A bit more recently, as Michael mentioned in his initial post, there is Norman Angell, who gave a great diagnosis of the material futility of war in the modern age, but was famously unlucky when he moved from analysis to prediction.
Subsequently, this thesis has been updated and refined by John Mueller and more recently still by 'Norman Angell with nukes', Thomas Barnett. One of the famous modern statements of the commercial peace comes from Solomon Polachek, who argued that mutual economic interdependence would make conflict more costly, and hence increase the chances of peace.
Away from theory, and the establishment and growth of the EU is arguably a concrete manifestation of these kinds of ideas. But is there much other empirical support for the idea of the commercial peace?
The chart below plots globalisation (measured as the ratio of world trade to GDP) and the occurrence of conflict (measured as the number of country pairs which in a given year are in a military conflict, divided by the number of existing country pairs) over the period 1870 to 2001. It suggests that there is no simple relationship between the two. For example, the first era of globalisation (1870-1914) is marked by both growing openness and rising military conflict. And the sharp rise in trade integration since 1970 has been associated with a relatively stable pattern of conflict.
This shouldn't be a complete surprise. Any sophisticated version of the theory would have to assert that trade discourages conflict, all else equal. But that last qualification is important since in reality, all else is never equal. This means that it's helpful to use statistical techniques to try to isolate the influence of trade on conflict. Read More
Political scientists rather than economists seem to have made most of the early running on this. In a large literature on the subject, they have tended to focus on the relationship between bilateral trade interdependence and military conflict. So, for example, one important early contribution from John Oneal, Bruce Russet and co-authors uses this approach to find empirical support for the hypothesis of a commercial peace.
On the other side of the debate, Katherine Barbieri's work has suggested that things are not quite so straightforward: instead of a negative relationship between bilateral trade interdependence and military conflict, Barbieri finds a positive one. Oneal and Russet have found that this result is sensitive to the way trade interdependence is measured.
Taken overall, surveys of this literature tend to find that the majority of studies support the idea of a negative relationship between bilateral trade and conflict, a result supported by some recent work.
Economists look to be relative newcomers to this debate. One interesting and relatively recent contribution comes from Philippe Martin, Thierry Mayer and Mathias Thoenig (summarised here). Their work confirms the proposition that a given pair of countries with more bilateral trade has a lower probability of bilateral war.
But they also report a more surprising finding: countries more open to global trade have a higher probability of war. They explain this result by arguing that, while the probability of war is lower for countries that trade more bilaterally because of the opportunity costs associated with lost trade, greater multilateral openness has the opposite effect because, by reducing the level of bilateral dependence, it reduces the opportunity cost of any given conflict.
They argue that multilateral trade openness therefore affects the nature of war: it increases the probability of local wars but deters global conflicts, a prediction which they argue fits well with the stylised fact that military conflicts have become more localised over time:
The same authors have built on these results to suggest that free trade agreements are good for peace.
Another recent paper, this time from Jong-Wha Lee and Ju Hyun Pyun (summarised here), also finds that bilateral trade interdependence is associated with a lower probability of conflict. And, in contrast to the surprising findings of Martin, Mayer and Thoenig, they find that global interdependence is good for peace. Lee and Pyun also find that the peace-promoting effects of trade vary depending on the degree of geographical proximity: greater bilateral interdependence seems to produce larger peace-promoting effects for neighbouring countries, while greater global trade openness has a relatively greater positive effect on peace for distant countries.
Summing up, I would judge that, while the evidence for the Pax Mercatoria is not absolutely conclusive, nevertheless its still strongly supportive of the general idea that international integration is good for peace, all else equal.
One last thought. I have concentrated on trade flows, but globalisation, even narrowly defined, encompasses more than the market for goods. While there has been some work on the relationship between conflict and other variables (such as foreign direct investment), this seems to be an area which is still relatively unexplored.