Global climate change is widely recognized as the most severe threat facing humankind. Combating the climate threat depends critically on the transition from unsustainable high-carbon economic growth to sustainable low-carbon development. This dissertation presents three scientific essays that endeavor to provide methodological innovations for evaluating and monitoring the low-carbon transition. The first essay focuses on the evaluation of the marginal abatement costs for greenhouse gases, which is fundamental to effective climate policy. The motivation of this essay stems from the observation that most empirical studies using frontier estimation methods grossly over-estimate the marginal abatement costs due to three sources of upward bias from ignoring input-side abatement options, inefficiency, and noisy data. To address the overestimation issue, we clarify the conceptual distinction between the shadow price and marginal abatement cost and develop a novel data-driven estimation approach based on convex quantile regression. Compared to the traditional approaches, convex quantile regression is more robust to random noise, the choice of the direction vector, and heteroscedasticity. An empirical application to U.S. electric power plants provides empirical evidence and demonstrates the advantages of the proposed approach. The second essay is motivated by a surprising paucity of well-developed studies investigating environmental productivity growth in consumption (a necessary condition for a smooth transition toward low-carbon consumption). This essay seeks to develop an environmental total-factor productivity index specific to the use of consumer durable goods. We first analyze the particular features of consumer durables (taking the passenger car as an example) compared to conventional production units, based on which we elaborate how to model the production activity during the use phase of consumer durables. Then, we present an overview of the existing approaches to measuring environmental productivity change and describe how they can be applied in the current context. Finally, we use a unique data set that covers all registered passenger cars in Finland to illustrate the interpretation of the proposed index. The third essay extends the second by further examining the driving factors behind decreasing CO2 emissions of new passenger cars. This essay develops a novel decomposition method to break down the change in the average CO2 emissions of new passenger cars into underlying factors such as technology, efficiency, and policy measures. Our decomposition draws insights from the traditional index decomposition analysis and frontier-based decomposition of productivity growth and satisfies several desirable properties. An empirical application to the Finnish register data sheds light on why and how the CO2 emissions of new cars decreased from year 2002 to 2014.
|Publication status||Published - 2019|
|MoE publication type||G5 Doctoral dissertation (article)|
- climate change, climate policy, data-driven analytics, environmental performance, frontier estimation, low-carbon consumption behavior, nonparametric regression, production theory, sustainable transportation