^[1]

Will globalisation reduce the chances of war in the Asia-Pacific? The numbers say we can’t be sure.

Perhaps it has something to do with the fact that 2014 is the centenary of the outbreak of World War I, but when analysts ask whether globalisation has dampened the possibility of conflict between great powers, 1914 is usually the go-to example for both sides of the argument. Speakers at the recent ‘Asia Today – 1914 Redux?’ conference held at ANU certainly drew extensively upon the lessons from that time. One colleague noted that ‘people thought trade would bring peace in 1914 too’. Another preferred to note the differences rather than the similarities between 2014 and 1914 ^[2], pointing to the growth in global supply chains to argue that globalisation is now far more advanced than it was at the beginning of the 20th century.

There’s a problem with over-reliance on World War I analogies, however. True, World War I was a tremendous tragedy with enormous geopolitical consequences, but it’s also just one out of potentially thousands of data points. It might be a statistical anomaly. Perhaps trade usually reduces the prospects of war and World War I is just an exception. Most people can think of an example of some ninety year old who’s been smoking and drinking heavily since adolescence but enjoys rude good health. The statistical outlier doesn’t mean that smoking and drinking aren’t bad for you. That’s why it’s important to complement in-depth studies of World War I with larger scale statistical analysis to understand the broader picture.

John O’Neal and Bruce Russett’s work ^[3] is perhaps the best known in this regard—and Steven Pinker cites them approvingly in his book The Better Angels of Our Nature. Analysing trade and conflict data from the nineteenth to the twenty-first centuries, they found that trade flows do have a significant impact in reducing the chances of conflict, even when taking a variety of other factors into account. But their conclusions have in turn been questioned by other scholars. For one thing, their model failed to take three things into account. First, it’s quite possible that peace causes trade rather than the other way around—no company wants to start an export business to another country if it anticipates that business linkages will be cut off by war further down the line. Second, conflict behaviour exhibits what’s called ‘network effects’— if France and Germany are at peace, chances are Belgium and Germany will be too. And third, both the likelihood of conflict and the level of trade are influenced by the number of years a pair of countries has already been at peace—because prolonged periods of peace increase mutual trust. Take any of these factors into account, and studies have shown (here ^[4] and here ^[5]) that the apparent relationship between trade flows and peace disappears.

Perhaps, though, conceiving of globalisation solely in terms of trade flows is mistaken. Alternative indicators of globalisation include foreign direct investment, financial openness and the levels of government intervention in economic relations with the rest of the world. Data on those variables is less extensive than on trade flows, usually dating back only to the post World War II period. But some analysts, such as Patrick McDonald ^[6] and Erik Gartzke ^[7], have argued that a significant correlation can be found between them and a reduction in the probability of conflict. Those findings, newer than O’Neal and Russett’s, haven’t yet been subjected to the same intense scrutiny, so may in turn be qualified by future research.

What does all that mean for the policy-maker? The statistical evidence certainly doesn’t tell us that globalisation has made war in East Asia impossible. ‘Cromwell’s law’ counsels us that a logically conceivable event should never be assigned a probability of zero. The most we could conclude is that globalisation has made such an occurrence much less likely. There’s some hopeful numerical evidence that globalisation does indeed have that effect, but the evidence isn’t so compelling that we can substitute an economic engagement policy for a security policy. By all means, let’s continue to promote trade in the Asia-Pacific. But we should also continue to be prepared for scenarios which are unlikely but would be hugely damaging if they were to occur.

Charles Miller is a lecturer at ANU’s Strategic and Defence Studies Centre. Image courtesy of Flickr user nickestamp ^[8].