US Environmental Policy Attributes and their Effects on Innovation

Abstract

Environmental policy is expected to have a significant, positive effect on innovation. Policies can be implemented by direct regulation, or through market-based processes. Direct regulation policies have various attributes, that affect innovation in different ways. This study investigates the effect of directed environmental policy on innovation. I examine how the stringency, and the uncertainty of environmental policy affected rates of patent counts (as a metric of innovation) in the United States from 1990 to 2006. I used a classification that identified and extracted patent counts for 16 green sectors of invention. The data was retrieved from the National Bureau of Economic Research (NBER) database of patents filed at the United States Patent and Trademark Office. I test three hypotheses predicting how particular attributes of environmental policy influence the rate of innovation in different sectors. Recently published, indices of the environmental policy stringency (restrictiveness of regulation and strictness of enforcement) and uncertainty (volatility, unpredictability) are used to measure each attribute. Three specifications of a simple time-series model were used to determine the sign and magnitude of policy attributes on patenting activity over 17 years. The results show that both stringency uncertainty index have significant and persistent negative effects on almost all patenting activities. Renewable energy technologies suffer the largest from stringency, and especially uncertainty of environmental policy; while innovation in water management and pollution abatement patents have little responses from stringency and uncertainty. The effects of uncertainty are much greater than those of stringency across all innovation sectors except pollution abatement. These findings highlight the complexity of the policy design, suggesting that regulatory policies designed to restore and maintain environmental quality must be highly certain and not too stringent if they are to also stimulate innovation. Further quantitative analyses of long time series of national data are required.