EXPERIENCE WITH CARBON TAXES AND GREENHOUSE GAS EMISSIONS TRADING SYSTEMS.

Carbon taxes and emissions trading systems (ETSs) to limit emissions of greenhouse gases (GHGs) are increasingly common. At the end of 2015, 17 GHG ETSs were operational in 55 jurisdictions, and 18 jurisdictions collected at least one carbon tax. This paper assesses the performance of carbon taxes and ETSs with respect to environmental effectiveness (reduction of emissions regulated by the instrument), costeffectiveness (marginal abatement cost), economic efficiency, public finance, and administrative issues. Data on emissions subject to carbon taxes are rarely reported. We estimate the taxed emissions for 17 taxes in 12 jurisdictions from 1991 through the end of 2015. All 17 taxes have reduced emissions relative to business-as-usual. Six of the jurisdictions actually reduced emissions, although in at least three of those jurisdictions the reductions appear to be due to other policies. The small sizes of reduction in almost all 17 cases are partially due to the low tax rates; the modest and uncertain changes in tax rates over time; and the limited response of taxed sources, such as fossil fuels, to price changes. Actual emissions declined for at least six of 10 ETSs. Other policies and developments, such as the 2009 recession, contributed to the reductions, but estimates of the share of the reduction attributable to the instrument are rare. All of the ETSs have accumulated banks of surplus allowances and most have implemented measures to reduce these banks. On average, the marginal cost of compliance is substantially lower for ETSs than carbon taxes. ETS experience has been shared bilaterally and via dedicated institutions. As a result, most ETSs have increased the share of allowances auctioned; adopted declining emissions caps; specified future caps and floor prices several years into the future; shifted to benchmarking for free allowance allocations to emissions-intensive, trade-exposed (EITE) sources; reduced accessibility to foreign offset credits; and established market stability reserves. By contrast, there is little evidence of shared learning and virtually no change to the design of carbon taxes. We found no jurisdiction that routinely tracks the taxed emissions. Very few jurisdictions regularly assess the effectiveness of the tax in achieving emission reductions. Additionally, adjustments to the tax rate often are unpredictable after an introductory period of three to five years. Both instruments reduce emissions, but ETSs have performed better than carbon taxes on the principal criteria of environmental effectiveness and cost-effectiveness. Many jurisdictions have implemented both a carbon tax and a GHG ETS, and every jurisdiction that has adopted either instrument has also implemented other policies. More research is needed to improve the design of both instruments and their interaction with non-market-based carbon policies because the use of multiple instruments produces complex interactive and distributional effects. While economically inefficient, market-based policies should be supplemented by non-market-based policies to ensure sustained political support. [ABSTRACT FROM AUTHOR]