Pattern Discovery for climate and environmental policy indicators

Quantitative environmental policy indicators are useful for modeling the impact of environmental policy on the economy. They can be important tools for policy-makers, companies, investors, and researchers alike. Well-crafted environmental policies lead to cleaner environments whilst encouraging innovative behaviour to stimulate green growth and ‘win-wins’ for the economy and the environment. Such win-win policies are increasingly sought out by policymakers, evidenced in the growing number of green 'new deals' and 'net zero' carbon emissions pledges at a national level. But there is a gap between the needs for environmental policy data and the supply of reliable indicators and indexes. This disconnect has negative consequences for policy feedback as well as the inducement of potential innovators of environmental technologies. While there are now a wide range of indicators and indexes, these largely remain inadequate for various reasons. This is disappointing considering the immense progress that has been made in machine learning and pattern discovery methods—methods that are already fully deployed in other research disciplines. Such automated techniques can limit human biases which currently plague the environmental indicator's scholarship. Against this backdrop, the main objective of this paper is to highlight how researchers can carefully collect these data and augment the effectiveness of environmental policy indicators. This is an important research area that, apart from a handful of studies, is not sufficiently developed.