157 NTU Management Review Vol. 31 No. 2 Aug. 2021 Accordingly, the DID regression, which incorporates other control variables, may obtain more accurate results than the DID estimator. Thus, the results of this study showing that the Biopharmaceutical Act does not influence the innovation quality of the approved biopharmaceutical firms. In sum, the Biopharmaceutical Act encourages biopharmaceutical firms to expand innovation input but does not improve their innovation quality. 4.3.3 Subsample Analysis of Intra-industry The subsection considers two intra-industry subsample analyses. By grouping firms with similar characteristics, these analyses may help to further realize which groups may dominate the main results of the sample. First, we consider that the subsamples are divided by different operating items. The different operating items in the biopharmaceutical industry may have different effects on the Biopharmaceutical Act’s encouragement of innovation activities. In Taiwan, the biopharmaceutical industry is usually divided into four groups: pharmaceuticals, medical equipment, applied biopharmaceutical, and others. The approved biopharmaceutical firms in our data include 66 pharmaceutical firms, 15 medical equipment firms and 2 applied biopharmaceutical firms. We divide the approved biopharmaceutical firms into two subgroups: pharmaceutical and non-pharmaceutical firms, because the sample of medical equipment and applied biopharmaceutical firms was too small for the DID regression.27 Table 5 shows the DID regression results for the intra-industry analysis of pharmaceutical and non-pharmaceutical firms. In Panels A.1 and A.2, the dependent variable is R&D investment. In Panels B.1 and B.2, the dependent variable is LN (1+adjusted patent citation). The coefficients of the interaction term, After×Treatment, for two, three and four matched firms of Panel A.1 are significantly positive but those of Panel A.2 are not significant. These results indicate that relative to unapproved pharmaceutical firms, the Biopharmaceutical Act encourages the approved pharmaceutical firms to increase R&D investment. 27 For the PSM exercise, the control firms are matched using the same operating items as the treated firms. Thus, the control firms for the approved pharmaceutical (non-pharmaceutical) firms are the unapproved pharmaceutical (non-pharmaceutical) firms.
RkJQdWJsaXNoZXIy ODg3MDU=