摘要：This paper investigates whether there is a significant relationship between government borrowing and firms’ capital structure and liquidity in Iran as a developing economy. In addition, the most innovative aspect of our research is to inspect which source of government debt has the most impact on the firm’s financing and liquidity policies. We carry out our inquiry on non-financial corporations listed in Tehran Stock Exchange from 2006 to 2017. The econometric model utilized in this study is the Generalized Method of Moments (GMM). We consider some other proved firm-specific and country-level factors as control variables in our model development according to the previous researches. Moreover, in addition to the econometric model, we harness Sobol’ sensitivity analysis and Support Vector Regression (SVR) for selecting the most influential source of government borrowing to reveal whether econometric and machine learning methods have the same result in this case and to compare them. The econometric results evince that not only the total government borrowing and corporate liquidity have negative influences on leverage ratio as the determinant of capital structure, but also government debt to the bank sector and other lending institutes plays the most essential role. It also shows the affinity of capital structure and liquidity literature implicitly. Additionally, the SVR model indicates similar results to the econometric model.
