Taking Stock

This chapter introduced the basic Schumpeterian model of economic growth to empha­size the importance of competition among firms both in the innovation process and in the product market.

The baseline model features process innovations leading to quality improvements. The description of economic growth that emerges from this model is, in many ways, more realistic than the expanding variety models. In particular, technological progress does not always cor­respond to new products or machines complementing existing ones, but involves the creation of higher-quality producers replacing incumbents. Arrow’s replacement effect, discussed in Chapter 12, implies that there is a strong incentive for new entrants to undertake research because the new, higher-quality products will replace the products of incumbents, leading to Schumpeterian creative destruction as the engine of economic growth. Even though the description of economic growth in this model is richer than the expanding varieties model, the mathematical structure turns out to be quite similar to the models in the previous chapters. In reduced form, the model again resembles an AK economy. The main difference is that now the growth rate of the economy, through the rate of replacement of old products, affects the value of innovation. Nevertheless, in the baseline version of the model, the effects of various policy interventions are the same as in the expanding product variety model of the previous chapter.

An important insight of Schumpeterian models is that growth comes with potential con­flict of interest. The process of creative destruction destroys the monopoly rents of previous incumbents. This raises the possibility that distortionary policies may arise endogenously as a way of protecting the rents of politically powerful incumbents.

Schumpeterian models also enable us to make greater contact with the industrial organi­zation of innovation. In particular, the process of creative destruction implies that the market structure may be evolving endogenously over time. Nevertheless, the baseline Schumpeterian models have a number of shortcomings, and addressing these is an interesting and important area for future research. An important discrepancy between the baseline models and the data is that, while the models predict all productivity growth to be driven by the process of creative destruction and entry, in the data much of productivity growth is driven by innova­tion by continuing firms and plants. Section 14.3 provided a first look at how the baseline models can be extended to account for these patterns and to provide a richer framework for the analysis of the industrial organization of innovation. A second important shortcoming of the baseline models is that they predict that markups are constant and there is always a single firm supplying the entire market. These implications can also be relaxed by con­sidering a richer framework, for example, by allowing cumulative or step-by-step innovation and competition between multiple firms that engage in innovation. Section 14.4 showed how the baseline model can be extended to incorporate step-by-step (cumulative) innovations and how such a model leads to endogenously evolving monopoly markups and market structure. Perhaps more interestingly, in models that incorporate different aspects of the industrial or­ganization of innovation, the effects of competition and patent protection on economic growth are potentially quite different from both the baseline model of Schumpeterian growth and from models of expanding varieties. This suggests that Schumpeterian models might pro­vide a useful framework for the analysis of a range of industrial policies, including anti-trust policies, licensing and intellectual property rights policies.

14.6.