New evidence for the relationship between government size and economic growth in Iran: an application for a three-regime Non-linear threshold regression model

abstract: Over the past two centuries, the role of government and the composition of government spending have changed in most countries of the world. Theoretically, there is no consensus among economists about the size of the state and the degree of state interference in the economy. On the one hand, one can observe classics and market's proponents who believe in a small government and little interference in the economy, and on the other hand socialists and Keynes believe in a lot of interference and a large government. This study, using data during 1967-2016 and Using nonlinear unit root test of caner and Hansen 2001, the nonlinear Co-integration of Enders and Siklos 2001 and Hansen's threshold regression model and based on the study of Annaman (2004), aimed to estimate the range of government optimal size for the Iranian economy by estimating a three-regime Non-linear threshold regression model. The results of this study show that the optimal size of the government in terms of current government expenditure (the ratio of current government expenditure to GDP) in the Iranian economy ranges from 10.12 to 18.25% and the optimal size of the government in terms of capital government expenditure (the ratio of capital government expenditure to GDP) in the Iranian economy is around 9.05 to 4.32%. Also, the degree of integration and co-integration of the variables follows a nonlinear threshold autoregressive process. As such that In the first place, some of the economic variables in Iran, according to the situation, have different degree of integration. Also Shrinking shocks, in comparison with increasing shocks create further deviations in the long run trend of the economic growth.