Corruption, size of government, and economic growth: evidence from global data

A big size government fosters corruption, which can lead to inefficiencies and resource costs that impede economic progress. In this chapter, it is argued that much of the previous studies have focused only on detecting the linear effects of corruption on growth. This study, therefore adopts the Threshold Autoregression (TAR) approach by using an annual panel data of 100 countries during 1990-2012 to evaluate any existence of a non-linear relationship. This study presents evidence that suggests the existence of a hump shaped (nonlinear) relationship between corruption and long-run economic growth. When the government size is small (11.518%), corruption positively affects economic growth. Whereas, when the government final consumption expenditure (% of GDP) is larger than 19.027%, corruption negatively affects economic growth. Furthermore, the result indicates that a non-linear relationship of the 'Armey curve' exists in our panel of countries. Thus, a government should investigate whether government size is over-expanding or not when designing its public finance policy