Financing and Market Timing of R&D

This section reports the results from OLS re­gression analysis where the dependent variable is the ratio of R&D to total assets in year t with the country and time fixed effects (not reported to save space) by using entire sample in Table 4.

Next, we investigate the influences of equity financing and market timing on R&D. The coef­ficients of EqFin1 (EqFin 2) and AdjMtoB are positive and significant at 1% level in all models where equity financing variables are inserted alone (not reported) or together with mispricing vari­able into the models⁵. EqFin 1 and EqFin 2 have very similar coefficients. This evidence supports the hypotheses 2 and 3, and indicates that as the percent of equity financing and industry adjusted market to book ratio increase so does R&D. Equity

Table 4. Financing and market timing of R&D

Robust standard errors are in brackets. The symbols ***, **, and * denote statistical significance at 1%, 5%, and 10% levels, respectively.

financing seems well suited for R&D given its unpredictability, its lack of collateral value, and the fact that the asymmetric information problem is not easily resolved. High market valuation low­ers the cost of equity and, thus, may encourage future R&D.

The results show that Patent Rights Protection Index (PRI) has also a nonlinear effect on R&D as consistent with the argument of Allred and Park (2007). However, the signs of PRI and SqPRI are not the same direction as Allred and Park (2007) suggested. PRI has significant and positive and SqPRI has significant and negative effects on R&D. Allred and Park (2007) suggest that low level (high level) of patent rights should have negative (positive) effect for especially developed countries. Therefore, models 3 and 4 introduce interaction variables between PRI, SqPRI and the dummies for the US and Market-Based Countries Without the US (MWUS). The coefficients of interaction variables of MWUS*PRI and MWUS*SqPRI, which represent the effect patent rights protec­tion in market-based countries, are negative and positive, respectively. This evidence is consistent with the re sults of Allred and Park (2007). Interest­ingly, the coefficients of PRI and SqPRI have the same signs (significant at 1% level) with previous models, indicating that the effect of patent rights protection is opposite in bank-based countries⁶. One possible explanation for this evidence would be that strong patent protection reduces competi­tive pressures to innovate for firms in bank-based countries. For firms in market-based countries, since develop a new technology and protect it are very expensive, firms may not want to carry this high cost if rights protection is not so strong. Benefits of high risky investments exceed costs after a certain level of protection.

Control variables Size and HTD are statistically significant in all models. Larger firms involve less R&D investment and there is a major difference in the level of expenditures between firms in high tech industries and firms in other industries.