Skip to main content
. 2023 Feb 13;49(2):740–781. doi: 10.1007/s10961-023-09995-9

Table 5.

The link between policy-induced R&D and firm performance, in foreign-owned subsidiaries and domestic firms

Variable Foreign-owned subsidiaries Domestic firms
Turnover (log) GVA (log) Exports (log) Turnover (log) GVA (log) Exports (log)
Privately-funded R&D 0.00806** (0.00322) 0.00751*** (0.00259) 0.00734** (0.00335) 0.152*** (0.0349) 0.129*** (0.0346) 0.159*** (0.0388)
Policy-induced: R&D tax credit 0.00668** (0.00315) 0.00599** (0.00323) 0.00607* (0.00322) 0.159*** (0.0306) 0.136*** (0.0312) 0.132*** (0.0361)
Policy-induced: R&D grant 0.00775** (0.00337) 0.00618** (0.00293) 0.00703** (0.00352) 0.146*** (0.0355) 0.119*** (0.0352) 0.172*** (0.0397)
Firm size: small 0.113 (0.188) − 0.0186 (0.170) 0.145 (0.190) 0.381*** (0.0256) 0.412*** (0.0297) 0.346*** (0.0439)
Firm size: medium 0.564*** (0.199) 0.343* (0.185) 0.500** (0.199) 0.751*** (0.0360) 0.775*** (0.0426) 0.663*** (0.0640)
Firm size: large 0.937*** (0.208) 0.671*** (0.199) 0.878*** (0.217) 1.128*** (0.0752) 1.184*** (0.0889) 0.893*** (0.143)
Materials 0.198 (0.169) − 0.912*** (0.182) 0.145 (0.158) 0.0252 (0.0409) − 0.787*** (0.0787) 0.0579*** (0.0224)
Unit labour costs − 0.0454*** (0.0133) − 0.0534*** (0.0166) − 0.0540*** (0.0168) − 0.000532 (0.000561) 0.00211 (0.00175) − 0.000351 (0.000979)
Training 0.0697*** (0.0174) 0.0831*** (0.0201) 0.0582*** (0.0181) 0.0517*** (0.00585) 0.0541*** (0.00736) 0.0589*** (0.0102)
Regional R&D 0.0136 (0.0418) 0.06244 (0.0538) 0.02296 (0.04668) 0.01592 (0.01937) − 0.0013 (0.0261) − 0.0210 (0.0406)
Constant 16.29*** (0.8034) 15.083*** (0.9937) 15.749*** (0.8720) 14.122*** (0.3575) 14.0183*** (0.48204) 13.3334*** (0.7455)
Firm fixed effects Yes Yes Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes Yes Yes
Observations 3,600 3,548 3,439 14,168 13,864 12,387
R2 0.235 0.191 0.199 0.238 0.181 0.167

*p < 0.1; **p < 0.05; ***p < 0.01. Robust standard errors in parentheses. The policy-induced R&D variables correspond to the α term defined in Eq. (2). The number of observations changes in each model due to missing values in the dependent variables. For ease of interpretation, the variables ‘Privately-funded R&D’, ‘Policy-induced: R&D tax credit (α)’, and ‘Policy-induced: R&D grant (α)’ have been scaled by 1 million