UC Berkeley

This dissertation studies the economics of historical and modern innovation. The first chapter makes inroads into understanding how competition and incentives shape the creative process which lies at the heart of all technological progress. The creative act is a classic example of a black box in academic research: we can see the inputs and outputs, but we know little about what happens in between. This paper uses new tools for measuring the content of digital media to see how commercial graphic designers’ work evolves in winner-take-all competition. In this chapter, I show that competition both creates and destroys incentives for innovation: some competition is necessary to motivate high-performers to experiment with novel, untested ideas over tweaking tried-and-true approaches, but heavy competition will drive them out of the market.

In the second chapter, I study the effects of performance feedback on innovation in competitive settings. Feedback typically serves two functions: it informs agents of their relative performance, and it also helps them improve the quality of their product. The presence of these effects suggests a tradeoff between participation and improvement, as the revelation of asymmetries discourages effort. Using data from the same setting as chapter one, I first show that this tradeoff is real. I then develop a structural model of the setting -- the first of its kind in the literature -- and use the results to evaluate counterfactual feedback policies. The results suggest that feedback is on net a desirable mechanism for a principal seeking high-quality innovation.

In the third chapter, I use the farm tractor as a case study to demonstrate that technologies diffuse along two distinct margins: scale and scope. Although tractors are now used in nearly every field operation and with nearly all crops, early models were far more limited in their capabilities, and only in the late 1920s did the technology begin to generalize for broader use with row crops such as corn. Diffusion prior to 1930 was accordingly heavily concentrated in the Wheat Belt, while growth in diffusion from 1930-1940 was concentrated in the Corn Belt. Other historically important innovations in agriculture and manufacturing share similar histories of expanding scope. The key to understanding the pace and path of technology diffusion is thus not only in explaining the number of different users, but also in explaining the number of different uses.

A common theme across all three chapters is the focus on developing tools or strategies to study innovation that are less dependent on patent data than the extant literature, since the majority of innovation is not patented (and often not patentable), and doing so while advancing the empirical literature on innovation in new directions.