Graph of the week: Comparative defence spending revisited
7 Aug 2012|

Andrew’s recent chart demonstrated how Australia’s defence spending equals the combined budgets of the five biggest spending ASEAN countries. Charts like this do a great job of communicating the magnitudes of spending involved, and put the budget cuts in perspective. There is a danger, however, that our appetite for budget comparisons can lead us astray.

When analysing countries of different sizes with different economies, comparing budgets in absolute dollar terms is misleading. Where labour rates are lower, that component of the budget is reduced. Where a nation has a large population and a conscription policy or retains a large military reserve, the nation can devote resources to a defence capability that isn’t captured accurately by the defence budget.

One attempt to broaden the measure of resources dedicated to defence is the Bonn International Center for Conversion (BICC) Global Militarization Index (GMI). The GMI aims to objectively depict worldwide militarisation, and is calculated on the basis of quantifiable aspects such as the availability of heavy weapons, and the size of the military budget in comparison to the healthcare budget.

The GMI is especially effective at illustrating regional military power balances. For example, it can capture the effect where one nation may ratchet up their militarisation through increased conscription (at low cost) and a neighbouring nation (where that solution is not politically acceptable) responds with increased spending on defence capital.

For comparison with last week’s charts, I’ve charted the GMIs over the same period (click to enlarge).

Proportion of total GMI Southeast Asia

Proportion of total GMI North AsiaOne conclusion is perhaps unchanged, in that the defence posture of the countries has remained reasonably constant; however the view that Australia is investing much more intensively than its neighbours is tempered.

As instructive as this is, it is important to understand the limitations of indexes and other measures based on a per-capita or percentage of GDP basis. The GMI and other indices tell us about the figurative ‘size of the fight in the dog’, not the ‘size of the dog in the fight’. Timor Leste has a substantial defence budget in % GDP terms but it would be a stretch to call it a military heavyweight.

A middle ground might be to compare defence spending adjusted for Purchasing Power Parity (PPP). This adjustment is an economic technique used to account for the different cost of goods and services within a country (more here), and goes some way to balancing out distortions such as varying labour rates. It is interesting to note how the North Asian PPP distribution highlights an increasing share spent by China over the last six years, despite the GMIs for the region staying constant.

Proportion of total defence budget PPP Southeast AsiaProportion of total defence budget PPP North AsiaEven here we need to be careful when interpreting the results, as the process of correcting for the indigenous cost variation leads to an overestimation of the country’s ability to purchase imported high end military equipment from suppliers from advanced economies—an important consideration if what we are assessing is power projection capability. At the end of the day we can only draw limited conclusions from these sorts of comparisons. The fact that similar data can be presented in three very different sets of graphs demonstrates the care that needs to be taken in deciding what data to look at and what decisions can be reliably drawn from it.

It is also important to step back and realise that defence budgets and heavy weapon counts are very crude aggregates of the people, training, systems, and technologies that interact to provide defence capabilities. To put it another way, we wouldn’t analyse household grocery expenditure and expect to predict how well a family can throw a barbecue. Make sure you season your data with plenty of healthy scepticism.

Kevin Atkinson is a senior engineer in KBR’s Defence and Government Service team.