The Political Economy of FDI 603 TABLE 3.Robustness of democracy and FDI (cross-section) Variables Model 5 Model 6 Model 7 Model 8 Model 9 MARKET SIZE 0.243 0.246 0.185 0.260 0.219 (1.445) (1.521) (1.162) (1.514) (1.344) DEVELOPMENT LEVEL -0.271 -0.173 0.160 -0.135 0.033 (-0.764) (-0.493) (0.517) -0.389) (0.117) GROWTH -0.361** -0.338** -0.277** -0.307** -0.293** (-3.561) -3.329) (-3.205) (-3.296) (-3.149) TRADE 0.033*** 0.034*** 0.033*** 0.034*** 0.033*** (11.363) (11.389) (10.886) (11.053) (11.139) NATURAL RESOURCES 5.861*** 6.130*** 6.025*** 6.255*** 6.137*** (3.352) (3.382) (3.171) (3.208) (3.100) GOVERNMENT CONSUMPTION -0.040* -0.042 -0.257 -0.038 -0.036 (-1.134) (-1.167 (-0.734) -1.043) (-0.916) BUDGET DEFICIT -0.114** -0.111* -0.112* -0.120** -0.115** (-2.523) (-2.413) (-2.430) (-2.493) (-2.329) DEMOCRACY 0.076*** 0.068*** 0.084*** 0.080*** 0.080*** (3.536) (2.922) (3.669) (3.488) (3.454) GOVERNMENT REPUTATION 0.198 (1.552) EXPROPRIATION 0.165 (1.210) CORRUPTION -0.159 (-1.288) RULE OF LAW 0.106 (0.836) BUREAUCRATIC QUALITY -0.017 (-0.128) FDI INFLOWS CONTROLS -1.816** -1.918** -1.840*** -1.813*** 一1.841** (-3.943) (-3.643) (-3.504) (-3.583) (-3.579) 69 69 69 69 69 0.76 0.76 0.76 0.75 0.75 Note:All regressions are ordinary least squares (OLS)cross-sectional regressions using net FDI inflows as a percent- age of GDP averaged from 1990-98 as the dependent variable. **p<01,*p<05,*p<,10. independent variables are lagged one year,and I have included a lagged depen- dent variable.36 The time-series cross-sectional econometric equation is: NET FDI INFLOWS,=FDI-1+B(INDEPENDENT VARIABLES-1)+ The empirical results presented in Table 4 support many of the findings from the cross-section regressions earlier.Not surprisingly,in these fixed-effects mod- 56.I have also tested this result without a lagged dependent variable,but with ARI correlations. The results are essentially unchanged.independent variables are lagged one year, and I have included a lagged depen￾dent variable+ 56 The time-series cross-sectional econometric equation is: NET FDI INFLOWSt 5 FDIt21 1 bi~INDEPENDENT VARIABLESt21! 1 «i The empirical results presented in Table 4 support many of the findings from the cross-section regressions earlier+ Not surprisingly, in these fixed-effects mod- 56+ I have also tested this result without a lagged dependent variable, but with AR1 correlations+ The results are essentially unchanged+ TABLE 3. Robustness of democracy and FDI (cross-section) Variables Model 5 Model 6 Model 7 Model 8 Model 9 market size 0+243 0+246 0+185 0+260 0+219 ~1+445! ~1+521! ~1+162! ~1+514! ~1+344! development level 20+271 20+173 0+160 20+135 0+033 ~20+764! ~20+493! ~0+517! ~20+389! ~0+117! growth 20+361*** 20+338*** 20+277*** 20+307*** 20+293*** ~23+561! ~23+329! ~23+205! ~23+296! ~23+149! trade 0+033*** 0+034*** 0+033*** 0+034*** 0+033*** ~11+363! ~11+389! ~10+886! ~11+053! ~11+139! natural resources 5+861*** 6+130*** 6+025*** 6+255*** 6+137*** ~3+352! ~3+382! ~3+171! ~3+208! ~3+100! government consumption 20+040** 20+042 20+257 20+038 20+036 ~21+134! ~21+167! ~20+734! ~21+043! ~20+916! budget deficit 20+114** 20+111** 20+112** 20+120** 20+115** ~22+523! ~22+413! ~22+430! ~22+493! ~22+329! democracy 0+076*** 0+068*** 0+084*** 0+080*** 0+080*** ~3+536! ~2+922! ~3+669! ~3+488! ~3+454! government reputation 0+198 ~1+552! expropriation 0+165 ~1+210! corruption 20+159 ~21+288! rule of law 0+106 ~0+836! bureaucratic quality 20+017 ~20+128! fdi inflows controls 21+816*** 21+918*** 21+840*** 21+813*** 21+841*** ~23+943! ~23+643! ~23+504! ~23+583! ~23+579! N 69 69 69 69 69 R2 0+76 0+76 0+76 0+75 0+75 Note: All regressions are ordinary least squares ~OLS! cross-sectional regressions using net FDI inflows as a percent￾age of GDP averaged from 1990–98 as the dependent variable+ ***p , +01, **p , +05, *p , +10+ The Political Economy of FDI 603