Using a time series data from 1990-2020, this study employs the augmented Solow human-capital-growth model to investigate the impact of government expenditure on economic deveopment outcomes in Ethiopia. Expenditure on public health and education were taken as proxy variables for human capital development in order to see their impact on economic development. The Augmented Dickey Fuller test is employed to test for stationarity and Johansen Cointegration technique is used to validate cointegration among variables as a sign of long run relationship. The error correction model is used to adjust for the short run error correction. Further tests of autocorrelation and residual normality distribution were done. The result of the ADF test has shown that all variables are non-stationary at level I (0) and stationary at I (1). There are two cointegrating equations implying convergence. The result of the error correction model show that the model is adjusting at a relatively stable rate of 61 % towards the long run equilibrium. The result of the short run causality tests show public expenditure on education, public expenditure on health have significant effect.