Wednesday, 12 March 2008

Disproportionately useful theories #4: war-growth analysis

Economics is important in everyday life, but is it the most important determinant of major societal events?

In the case of civil war, some answers have recently appeared. There has been a burst of academic work examining whether economic growth and civil war are related. Maybe low economic growth causes civil war - people can't find jobs and take up guns to feed themselves, for example - or maybe civil war causes low economic growth, for obvious reasons. The links are fairly clear, but the precise measurement is important.

The measurement techniques are quite innovative, and owe their design partially to the impetus provided by the MRW growth model (one of the earlier disproportionately useful theories). We might see that a country has a collapsing economy and civil war in a year, but there has to be some way of determining how much they caused each other. War-growth analysis helps to provide it.

The approach is new - some date the recent developments to a 1998 paper - so it is too much to ask for it to be already heralding a new era of harmony and other niceness. Its uses are prospective, such as identifying which countries are most at risk of conflict and acting to prevent it. Even if prevention is too grand a goal, just knowledge of precise risks will help countries who are at risk of conflict and usually poor. Investors to conflict-prone countries who do not know exact risks can demand enormous returns on their money, because they demand compensation for high risk, which is assumed, plus compensation for the extra uncertainty that the lack of knowledge brings. War-growth analysis could remove the latter premium, and save countries money.

And the answer to the question on the importance of economic growth in causing war? Leading estimates say that it is very important. Comparison with other factors is difficult from many published papers because it is not standard to list the importance of country specific factors, unfortunately, but those factors would have to have a very large impact to trump economic growth's impact.

No comments: