Government Expenditure and Private Sector Growth in Saudi Arabia: A Markov Switching Model Analysis

This paper investigates the relationship between government expenditure and non-oil private GDP in Saudi Arabia over the period 1970-2015, using a Markov Switching Autoregressive Model (MSAR)-which captures the dynamic pattern of time series and allows us to test the impact of government expenditure on GDP growth in finite unobserved states of the economy. The study results show that the growth effects of contemporaneous government consumption expenditure and government fixed capital formation expenditure are found to be negative only in the low (recessionary) state of the economy. This could be explained by the crowding out effect that exists only in recessionary periods. On the other hand, the disaggregated government expenditure model indicates that defence and security expenditure is estimated to have the expected negative and significant impact on non-oil private GDP growth, but only in the low state. In addition, out of the three highest government civil items of expenditure, human resources development is found to have a positive and significant growth effect in both states of the economy, while the growth effect of the other two types (health and economic development expenditure) is found to be insignificant in the low state. This insignificant growth effect of some types of government expenditure in the low state of non-oil private GDP could be explained by the procyclical nature of government expenditure because of the excessive dependency on oil revenues.