This study explores the welfare effects of patent protection in a Romer-type expanding variety model in which R&D and capital accumulation are both engines of growth. It shows that the comparison between the productivity of R&D and that of capital plays an important role in the welfare analysis. When the relative productivity of R&D compared to capital is high (low), social welfare takes an inverted-U shape for (is decreasing in) the strength of patent protection, and the welfare-maximizing degree of patent protection is no greater than (identical to) the growth-maximizing degree. Moreover, the model is calibrated to the US economy and the numerical results support these welfare implications.

How does patent policy affect innovation when patent licensing is crucial for firms? To address this question, the present study incorporates voluntary patent licensing between an innovator and followers, as discussed in the literature of industrial organization, into a dynamic general equilibrium model. Unlike previous studies, both the licensing fee and the number of licensees are endogenously determined by the innovator’s maximization and the free-entry condition. Using this model, we show that strong patent protection does not always enhance innovation, economic growth, and welfare. Furthermore, the extended analysis provides the policy implication that the effect of patent policy depends on how difficult further innovation is without patent licensing of the current leading technology.

This paper examines how strengthening patent protection affects welfare in a nonscale quality-ladder model, which was developed by Segerstrom [American Economic Review 88, 1290–1310] and generalized by Li [American Economic Review 93, 1009–1017]. In the Segerstrom–Li model, patent protection creates no distortion in static allocation among the production sectors. In order to examine the welfare effects of strengthening patent protection adequately, we incorporate a competitive outside good into the Segerstrom–Li model. In the general model, we derive the welfare-maximizing degree of patent protection analytically by utilizing a linear approximation of the transition path. The result shows that the welfare-maximizing degree of patent protection is weaker when the market share of the outside good is positive than when it is zero. In other words, disregarding the static distortion that patent protection creates leads to excessive patent protection.

We construct a growth model with status preference to explore the effects of patents on innovation and social welfare. We find a nonmonotonic effect of patent protection on innovation. Additionally, the growth-rate-maximizing degree of patent protection decreases when the strength of status preference is larger. The effect of patent protection on social welfare is ambiguous, depending on the strength of status preference. Moreover, wealth inequality widens as patent protection is reinforced. Finally, by using cross-section regression analysis, we document that a nonmonotonic relationship between patent protection and economic growth is statistically significant and that the growth-rate-maximizing degree of patent protection decreases with the strength of status preference.