The new growth theory establishes, among other things, that government expenditure can manipulate economic growth of a country. This study attempts to explain whether government expenditure increases or decreases economic growth in the context of Sri Lanka. Results obtained applying an analytical framework based on time series and second degree polynomial regressions are generally consistent with previous findings: government expenditure and economic growth are positively correlated; excessive government expenditure is negatively correlated with economic growth; and an open economy promotes growth. In a separate section, the paper examines Armey's (1995) idea of a quadratic curve that explains the level of government expenditure in an economy and the corresponding level of economic growth. The findings confirm the possibility of constructing the Armey curve for Sri Lanka, and it estimates the optimal level of government expenditure to be approximately 27 per cent. This paper adds to the literature indicating that the Armey curve is a reality not only for developed economies, but also for developing economies.(author's abstract) / Series: SRE - Discussion Papers