Government debt has many characteristics and thus cannot be fully captured by one indicator. There are several different ways of defining government debt, and each definition can lead to different interpretations of a government’s financial situation. As a result, international comparisons of government debt statistics must be conducted with care. This working paper will summarise some of the major differences in defining and measuring government debt and, based on available data, will demonstrate the impact of these differences when comparing the level of government debt as a percentage of GDP across various OECD countries. This paper will also look at the challenges in incorporating government-sponsored pension schemes in government debt statistics and the implications for making international comparisons of government debt. It will then attempt to compare government debt statistics by adjusting for some of these challenges. It will also present a complementary approach for analysing and comparing government debt, using projections of old age dependency and economic growth in order to provide additional context. The key findings of this paper are that: (i) international comparisons of government debt should exclude pension liabilities until more data are available from more countries; (ii) comparisons of government debt should be conducted separately for implicit and explicit liabilities as well as for funded and unfunded pension liabilities; (iii) further cooperation is required between the national accounts, actuary and public sector / business accounting communities to enable methodological consistency in the estimation of pension liabilities; and (iv) comparisons of government debt should not rely on one indicator but instead utilise a wide array of statistics in order to provide a more relevant and complete picture of government finances both within and across countries.