Research

The economic cost of carbon pricing depends on the ability and incentives of firms to switch towards cleaner fuels. Yet, many fundamental economic forces that drive firms' decisions to use different fuels are unobserved, causing significant uncertainty over the effectiveness of carbon policies. In this paper, I propose a new dynamic production model with multidimensional unobserved heterogeneity that underly technology differences and captures how firms' fuel choices respond to price changes. These differences cause heterogeneity in abatement costs, which generates heterogeneous responses to carbon pricing. Leveraging minimal assumptions about optimal input choice and the technology frontier, I quantify the model from a detailed panel of Indian steel establishments. Based on these estimates, implementing a carbon tax equivalent to 2,000 INR/ton (25 USD/ton) of carbon dioxide equivalent leads to a 70% reduction in emissions. But only 18% of this reduction comes from fuel-switching within existing firms. I find that the larger reductions come from reallocation of output across firms (58%) and costly reduction in aggregate output (24%). Substantial heterogeneity in the fuel efficiency of existing furnaces coupled with the limited geographical reach of natural gas pipelines towards high-emission firms explains the prevalence of output reallocation relative to fuel switching.

Presented at the 2023 EEA-ESEM Congress (Barcelona), the 2023 European Association for Research in Industrial Economics (EARIE, Rome), the 57th Annual Conference of the Canadian Economics Association (Winnipeg), UM-UWO-MSU Labo(u)r Day (2023, East Lansing),  The 21st Annual International Industrial Organization Conference (IIOC 2023, Washington DC), Young Economist Symposium (2022, Yale University), the 56th Annual Canadian Economics Association Meetings (2022, Ottawa), and the 17th CIREQ PhD Students' Conference (2022, Montreal) 

This paper studies how firms reorganize their production when faced with asymmetric carbon pricing. While firms may compete with each other across geographical regions, some regions may have carbon pricing policies while others do not. This can lead to carbon leakage, where emissions shift from regulated to unregulated regions. I build a model of imperfect competition with multiple fuels as energy inputs, which allows for region-specific carbon taxes. I estimate the structural model with publicly available Canadian plant-level data on a wide range of pollutants emitted in the air to quantify the effect of the British-Columbia (BC) and Quebec carbon taxes that were implemented in 2008 and 2007, respectively. I find strong evidence of carbon leakage in other Canadian provinces, which mitigated 45 % of emissions reduction efforts in BC and Quebec. I find that Canadian plants do not find it profitable to switch between fuels. As a result, regulated firms become less competitive, and much output is reallocated towards unregulated firms. A uniform carbon tax in all provinces, which was introduced in 2018, fully mitigates carbon leakage within Canada and reduces aggregate emissions by 21 % relative to a 2.15 % reduction with the asymmetric carbon tax.

This research uses the method developed by Huang and Salmon (2004) to estimate marketherding dynamics in a CAPM framework, exploiting time variation in betas and cross-sectional dispersion of individual assets. Their model is adapted to the cryptocurrencymarket in order to shed some lights on the recent surge in prices and volatility. Using high frequency data at 5 minutes intervals, I find significant evidence of increasing market herding over the high volatility period between August 2017 and March 2018.