Effects of patent policy on income and consumption inequality in a R&D growth model.

• Author: Chu, Angus C.
• Position: Research and development

1. Introduction

   Since the seminal work of Simon Kuznets (1955), the tradeoff between growth and inequality has been an important issue in economics. Given that economic growth is driven by technological progress, which in turn is influenced by innovation policies, this article analyzes the effects of patent policy on economic growth and income inequality within a research and development (R&D)--based endogenous-growth model. In the model, the effect of patent policy on income inequality is driven by the rate of return on assets. Therefore, even if patents do not represent a significant fraction of assets in reality, (1) the effect of patent policy on income inequality can still be significant in the presence of other capital incomes that depend on the real interest rate. Although the prevailing wisdom is that the rising income inequality in the United States is largely driven by an increase in the relative wage between skilled and unskilled workers, some studies, such as Atkinson (2000, 2003), suggest that inequality in capital income is also playing an increasingly important role. For example, Reed and Cancian (2001) show that capital income contributes to one-quarter of the increase in income inequality in the 1990s, but it accounts for less than one-tenth of the increase in the 1970s. The current study relates to this literature by providing a model that highlights the effects of capital income on the rising inequality in the United States.

   [FIGURE 1 OMITTED]

   The growth-theoretical framework is a canonical quality-ladder model with the additions of heterogeneity in households' wealth, variable patent breadth, and elastic labor supply. The model predicts that strengthening patent protection increases (a) economic growth by stimulating R&D investment and (b) income inequality by raising the return on assets. However, whether it also increases consumption inequality depends on the elasticity of intertemporal substitution. If this elasticity is less (greater) than unity, strengthening patent protection would increase (decrease) consumption inequality. Calibrating the model to aggregate data of the U.S. economy shows that strengthening patent protection leads to a larger increase in income inequality than consumption inequality. This divergence between income and consumption inequality is consistent with the empirical pattern in the United States.

   Krueger and Perri (2006) and Blundell, Pistaferri, and Preston (2008) provide empirical evidence to show that the sharp increase in income inequality in the United States since the 1980s has been accompanied by a much smaller increase in consumption inequality. For example, based on the Consumer Expenditure Survey, Krueger and Peril (2006) find that the variance of log of income (consumption) increases by over 20% (about 5%) from 1980 to 2004. During the same period, R&D investment as a share of gross domestic product (GDP) has increased (see Figure 1) while patent protection in the United States has strengthened. (2) Table 1 presents an index for the strength of patent protection in the United States from Park (2008).

   Given this empirical pattern, I calibrate the R&D-growth model to see whether it can replicate a similar divergence in income and consumption inequality as in the data. The model predicts that the coefficient of variation of income over the coefficient of variation of consumption increases from 1.55 in 1980 to 1.69 in 2004. This finding suggests that patent policy may provide a partial explanation for the recent trend of income and consumption inequality in the United States.

   The intuition of the results is as follows. Strengthening patent protection increases R&D as well as the equilibrium growth rate that drives up the rate of return on assets. This higher return on assets increases the income of asset-wealthy households relative to asset-poor households. Furthermore, the allowance of elastic labor supply creates an additional effect on income inequality through labor income, and this effect will be discussed in detail in the main part of the article. As for the ambiguous effect on consumption inequality, the higher growth rate also increases the fraction of assets for saving. Therefore, whether the relative consumption between asset-wealthy households and asset-poor households increases or decreases depends on the relative increase in the equilibrium growth rate and the real interest rate, which in turn is determined by the elasticity of intertemporal substitution.

   Literature Review

   Since the creation of the seminal Kuznets curve that hypothesizes an inverted U-shape effect of economic development on income inequality, economists have become interested in the empirical relationship between economic growth and income inequality. Early empirical studies, such as Alesina and Rodrik (1994), Persson and Tabellini (1994), and Perotti (1996), find a negative relationship, while the more recent studies, such as Li and Zou (1998) and Forbes (2000), find a positive relationship. Forbes (2000) argues that the different results are due to omitted-variable bias and measurement error in previous studies and suggests the use of panel estimation and improved data on inequality to overcome these problems. Barro (2000) considers a larger sample of countries than Forbes (2000) and finds a positive (negative) relationship between growth and inequality in developed (developing) countries. Assuming that economic growth in developed countries is driven by innovation, the theoretical result that stronger patent protection increases income inequality is consistent with a positive relationship between growth and inequality in developed countries.

   Early theoretical studies, such as Galor and Zeira (1993), Alesina and Rodrik (1994), and Persson and Tabellini (1994), tend to derive a negative relationship between growth and inequality. Garcia-Penalosa and Turnovsky (2006) argue that the theoretical relationship between growth and inequality should be ambiguous and depends on the underlying structural and policy changes. To explore this theoretical relationship, they incorporate heterogeneity in households' wealth into a canonical AK growth model with capital externality and elastic labor supply. The comparative static results show that a positive growth-inequality relationship is more likely to emerge. They also derive a law of motion for the distribution of assets and show that the distribution is stationary in the model. (3) The current study adopts a similar approach to show that the distribution of assets is also stationary in a canonical quality-ladder growth model. An interesting difference between the two models is that the AK model relies on elastic labor supply to generate an endogenous income distribution, while the quality-ladder model does not. Nonetheless, the consideration of elastic labor supply is still interesting because it creates an additional channel through which growth affects income inequality through labor income.

   Although the capital-accumulation-driven growth models are useful frameworks for analyzing many macroeconomic issues, they are not suitable for evaluating innovation policies. Therefore, the current study incorporates wealth heterogeneity and elastic labor supply into a R&D-driven endogenous-growth model to analyze the effects of patent policy on growth and inequality. Chou and Talmain (1996), Li (1998), Zweimuller (2000), and Foellmi and Zweimuller (2006) also consider wealth heterogeneity in R&D-growth models. However, they focus on the effects of wealth inequality on growth. The current article differs from these studies by considering the effects of patent policy on income and consumption inequality given wealth inequality that is independent of growth in the model.

   Bertola, Foellmi, and Zweimuller (2006, ch. 10) also consider an R&D-growth model in which wealth inequality is independent of growth because of homothetic preferences. Bertola et al. analyze the effect of firms' market power determined by the elasticity of substitution between products on the distribution of income between workers and entrepreneurs, and they derive an inverse relationship between the labor share of income and the market power of firms. The current study explores a different issue. It first derives a closed-form expression for the coefficient of variation of income/consumption and then shows that stronger patent protection has a positive effect on income inequality but an ambiguous effect on consumption inequality. The current study also differs from Bertola et al. in the following ways. First, they consider a variety-expanding model with inelastic labor supply, while the current study considers a quality-ladder model with elastic labor supply that creates an additional effect on income inequality. Second, the current study analyzes consumption inequality in addition to income inequality and shows that these two measures of inequality could in theory go in opposite directions. Finally, the current study also provides a quantitative analysis to illustrate the effects of patent breadth on income and consumption inequality.

   This article also relates to the issue on the underinvestment in R&D. There is an important empirical literature that finds the social return to R&D to be much higher than the private return. (4) Jones and Williams (1998, 2000) develop a R&D-growth model and use these empirical estimates to quantify that the socially optimal level of R&D is at least two to four times higher than the market level. Given this underinvestment in R&D, patent policy is a relevant instrument that can be used to correct for this market failure and increase growth. In the R&D-growth literature, Li (2001) and O'Donoghue and Zweimuller (2004) analyze the growth effects of patent breadth in a quality-ladder...