Reading military spending data: what a database like SIPRI’s does and does not tell you

The Stockholm International Peace Research Institute (SIPRI) maintains a widely used database of military expenditure estimates covering many countries and years. It is a good example of a source that is valuable and easy to misread.

What it offers: comparable figures over long periods, usually expressed in constant prices so that inflation does not distort the trend, and as a share of national income, so that large and small economies can be compared.

What to keep in mind. First, definitions matter: what a country counts as military spending, such as pensions, paramilitary forces or research, differs. Second, some countries publish little, so figures are estimates. Third, spending is not capability: the same amount can buy very different things depending on prices, wages, procurement and corruption.

Limits: a higher number does not by itself mean a country is more dangerous, and a lower one does not mean it is safer.

A fair use of such data is to track one country's trend over time, and to compare like with like, rather than to rank countries on a single year's total.