Three essays on old and new technologies

The essays in this dissertation examine boundaries between new technologies and the old technologies they replace. The first essay examines the question of why technologies often experience a period of extraordinary improvement or a "last gasp" before being superseded by new technologies. The standard explanation for this last gasp is that adherents of threatened technologies try harder to improve their technology after the threat from a new technology materializes. Although this phenomenon should inform the study of technological transitions in general and firms' technology policy decisions in particular, the "trying harder" explanation has not been tested against alternative explanations. I develop and test two new explanations for last gasps. I find that the automobile carburetor's last gasp was caused by the selection effect and by the spillover of electronic control technology from Electronic Fuel Injection to carburetors. Significantly, however, only those carburetor firms that also manufactured EFI were able to capture the benefits of these spillovers. In the second essay, I propose and empirically test two hypotheses related to the question of whether firm absorptive capacity affects transfers from next-generation technologies to current-generation-technology products. I find that experience gained working with future technologies, as measured by patenting activity, increases the efficiency with which future-technology-derived components are transferred to current-generation-technology products. I also find that this increases the likelihood that this transfer will occur. I find preliminary evidence suggesting that those inventive activities must be sufficiently closely related to the current-generation-technology products for them to impact absorption of the next-generation-technology component. In the third essay, I focus on the use of Information Technology (IT) to improve Physician performance. Physicians, like other professionals, make mistakes. Catching and correcting those mistakes before they result in harm to patients is the goal of real-time physician performance feedback systems. We examine a medical IT program implemented as a randomized trial by a large health care insurer in California. We find that patients whose physicians receive the IT feedback are more likely to have their treatment come into compliance with clinical practice guidelines.