«Will border carbon adjustments work? Niven Winchester*,†,‡, Sergey Paltsev* and John Reilly* Abstract The potential for greenhouse gas (GHG) ...»
BCAs influence welfare in three ways. First, as is well known, tariffs create production and consumption inefficiencies. Second, tariffs improve market access for coalition exporters at the expense of non-coalition firms. Third, tariffs generate terms of trade effects, which are considerable when commodities are differentiated by country of origin (Brown, 1987). In the U.S.-ALL simulation, U.S. welfare decreases due to efficiency losses. BCAs also induce a substitution in U.S. purchases towards goods shipped from coalition regions, resulting in a welfare improvement in the coalition and a welfare loss in the non-coalition. Welfare losses are largest in Mexico, Higher Income East Asia and China, all of which export large quantities of goods to the U.S. There is also a large fall in welfare in the Middle East, which is driven by a further decline in the price of Crude oil.
When the coalition imposes tariffs on all products, CLT-ALL, market access and terms of trade gains outweigh efficiency losses in the coalition so welfare for the coalition increases (from
-0.92 in the CAT scenario to -0.79). Welfare in the non-coalition falls (from -0.23% in the CAT scenario to –1.41%) and, as in U.S.-ALL, the largest losers are Mexico, Higher Income East Asia, China and the Middle East. It is also interesting to compare welfare changes for the CLTALL and U.S.-ALL scenarios. First, U.S. welfare improves when the coalition imposes tariffs relative to when only the U.S. levies tariffs due to market access effects. Second, the numbers reveal that nearly all of the decrease in welfare in Mexico and about two-thirds of that for China is brought about by U.S. BCAs.
Welfare impacts for the U.S.-MNF scenario are qualitatively similar to those for U.S.-ALL, but are smaller in magnitude. That is, BCAs reduce welfare in the U.S. and non-coalition regions and increase welfare in other coalition regions, but by relatively small amounts. In the CLTMNF scenario, coalition manufacturing tariffs raise welfare in some coalition regions and reduce welfare in non-coalition regions, as in CLT-ALL, and the largest losers are Mexico, Higher Income East Asia, China, the FSU, and the Middle East. The results also indicate that BCAs have little impact on the U.S. CO2-e price.
Although BCAs reduce leakage by up to 60%, the numbers mask small changes in global emissions. As displayed in Table 5, the coalition’s contribution to global emission is 22% and the non-coalition’s is 78% in the CAT scenario, so leakage calculations are sensitive to small proportional changes in non-coalition emissions. As a result, the 60% leakage reduction in CLTALL – the largest tariff-induced leakage reduction – corresponds to a 0.8% fall in non-coalition emissions and only a 0.6% decrease in global emissions. When combined with the welfare changes reported in Table 3, the leakage calculations indicate that reducing emissions via BCAs is nearly six times as costly as using direct controls. In the CAT scenario, global emissions fall by 4.7 billion tons and global welfare declines by 0.7%, resulting in a welfare reduction of 0.15% per billion tons of abatement. The corresponding number for CLT-ALL is 0.88%.
5. ALTERNATIVE LEAKAGE CONTROLSAt least two other policy measures can be used to address leakage concerns. First, Burniaux et al. (2008), Fisher and Fox (2009) and McKibbin and Wilcoxen (2009) claim that the most important source of leakage arises from reductions in the global oil price induced by coalition GHG restrictions, which raise oil-intensive consumption in regions with no GHG controls.
Accordingly, in a another scenario (OILTAX), we impose an endogenous tax on Crude oil production, applied uniformly across regions, so that the tax-inclusive Crude oil price is the same as in BAU. Such a tax is an unlikely outcome from climate negotiations but could represent oil producers exercising monopoly power to stabilize the world oil price. Second, direct controls (cap-and-trade system and/or energy efficiency mandates in non-coalition regions) can also reduce leakage. Here one might view the threat of BCAs as a way for the climate coalition to coerce other nations to restrict GHG emissions.10 To investigate this alternative, instead of BCAs, we impose cap-and-trade programs in non-coalition regions in addition to carbon Alternatively, unilateral emissions reductions by the coalition may create self-interested emission reductions by the non-coalition (Copeland and Taylor, 2005).
restrictions in the CAT scenario. The emissions cap for each non-coalition region in this scenarios is set at emissions observed in the CLT-ALL scenario. In another simulation, we use cap-and-trade systems to determine non-coalition carbon prices that eliminate leakage by returning emissions in each non-coalition region to emissions in BAU.
In our OILTAX simulation, a global tax on Crude oil production of 11.3% is required to equate the tax inclusive Crude oil price in the CAT scenario to that for BAU. As indicated in Table 3, aggregate welfare for the coalition when there is an oil tax is lower than in all of the BCA scenarios. This is because the terms of trade for the coalition (which is a net-importer of oil) improves as a result of BCAs but declines when there is an oil tax. The opposite is true for the non-coalition (which is a net-exporter of oil). Interestingly, the OILTAX leakage rate (4.3%) is higher than the CTL-ALL leakage rate (3.8%). That is, trade measures appear to be as effective at reducing leakage as oil price controls. This finding concurs with Felder and Rutherford (1993) but refutes recent conjectures that, “in practice, the most important source of mechanism through which leakage could occur would be world oil markets, not trade in manufactured goods” (McKibbin and Wilcoxen, 2009, p. 3).
When we apply cap-and-trade policies in the non-coalition, emissions are reduced to levels in the CTL-ALL scenario by CO2-e prices less than $0.01 per ton in the non-coalition (except Mexico, where the CO2-e price is $0.29) and welfare changes are very small.11 Indeed, proportional welfare changes are only distinguishable from CAT values at very high levels of precision, so we do not report welfare changes when cap-and trade programs are imposed on the non-coalition. Furthermore, eliminating leakage by non-coalition cap-and trade policies requires CO2-e prices less than $0.01 per ton in all non-coalition regions except in Mexico ($0.48) and the FSU ($0.02), which also have minor welfare effects. It is likely that near-term non-coalition capand-trade policies are infeasible, either because the non-coalition refuses to bind emissions and/or because tiny carbon prices render such systems impractical, so efficiency improvements may be a more feasible way of reducing emissions. In this connection, BCAs reduce Chinese emissions by 3.7 million tons in the CLT-ALL scenario so, assuming replacing a standard light bulb with a compact fluorescent light (CFL) bulb saves 100 kilograms of CO2-e emissions per year, Chinese leakage could be offset by the same amount if one in ten of China’s 360 million households installing a single CFL bulb.
6. CONCLUSIONS We evaluated the potential for BCAs to reduce leakage using an economy-wide model focusing on 2025. We found that BCAs reduce leakage by around 30% when imposed by the U.S. and about 60% when levied by all coalition countries. However, as the non-coalition accounts for more than three-quarters of global emissions, large proportional leakage changes Small non-coalition carbon prices reflect low initial emissions reduction costs in this region. As shown by Carbone et al. (2009), cheap abatement opportunities in developing countries provide scope for international trade in emissions permits even in the absence of a global cooperative agreement.
mask small changes in emissions – when leakage fell by 60%, non-coalition emissions fell by 0.8% and global emissions declined by only 0.6%.
Although BCAs have small emission impacts, they have pronounced welfare effects. When the coalition imposed BCAs on all products, the change in coalition welfare improved from in the CAT scenario to -0.53%, but the change in non-coalition welfare deteriorated from
-0.23% to -1.41%. The net result was a worsening in world welfare from -0.70% in the CAT scenario to -0.98% (for an almost negligible reduction in global emissions). As an alternative to BCAs, we considered pricing non-coalition GHG emissions so that, in each region, emission levels equaled those observed when BCAs are employed. CO2-e prices that achieve this objective were around one-tenth of one cent in nearly all regions and had negligible welfare effects. Capand-trade programs with such small carbon prices may not be viable, so the adoption of modest energy efficiency improvements in the non-coalition may be a more practical solution.
These findings suggest that non-coalition regions may wish to adopt emissions controls as part of a global agreement, providing such measures prevent the coalition from adopting BCAs.
China recently announced plans to reduce its 2020 GHG emissions to GDP ratio by 45% relative to 2005 through GHG efficiency improvements, so an agreement binding China to this goal (or a slightly more ambitious target) may be a viable alternative to BCAs. However, as leakage reductions achieved by modest non-coalition controls will still leave coalition producers at a cost disadvantage relative to imports from the non-coalition, it remains to be seen whether coalition politicians will be willing to strike out BCAs.
Regarding competiveness concerns, BCAs applied to all sectors will not necessarily increase energy intensive output. This is because, relative to other sectors, a high proportion of coalition energy intensive imports are sourced from other coalition regions and do not attract BCAs.
Consequently, the energy intensive tariff value-weighted across sources is lower than valueweighted tariffs for some other sectors. However, BCAs applied only to manufacturing raise domestic manufacturing output but do not fully offset the impact of domestic carbon restrictions.
If the U.S. acts unilaterally, BCAs are detrimental to U.S. exports and increase the cost of climate policy.
We also evaluated the conjecture that trade in goods is a minor leakage source compared to the oil-price channel. Contrary to conventional wisdom, we found that comprehensive coalition BCAs reduced leakage to a greater extent than measures that offset the decrease in the price of Crude oil caused by coalition emissions restrictions. Nevertheless, we do not recommend using BCAs to address leakage concerns. Instead, we conclude that although the political landscape in the U.S. and other Annex 1 nations may call for BCAs to control leakage and address competitiveness concerns, BCAs are imprecise instruments that, even when finely tuned to target embodied GHG emissions, cause much collateral damage.
Acknowledgements The Joint Program on the Science and Policy of Global Change is funded by the U.S.
Department of Energy, Office of Science under grants DE-FG02-94ER61937, DE-FG02ER61677, DE-FG02-08ER64597, and DE-FG02-06ER64320; the U.S. Environmental Protection Agency under grants XA-83344601-0, XA-83240101, XA-83042801-0, PI-83412601RD-83096001, and RD-83427901-0; the U.S. National Science Foundation under grants SESEFRI-0835414, ATM-0120468, BCS-0410344, ATM-0329759, and DMS-0426845;
the U.S. National Aeronautics and Space Administration under grants NNX07AI49G, NNX08AY59A, NNX06AC30A, NNX09AK26G, NNX08AL73G, NNX09AI26G, NNG04GJ80G, NNG04GP30G, and NNA06CN09A; the U.S. National Oceanic and Atmospheric Administration under grants DG1330-05-CN-1308, NA070AR4310050, and NA16GP2290; the U.S. Federal Aviation Administration under grant 06-C-NE-MIT; the Electric Power Research Institute under grant EP-P32616/C15124; and a consortium of 40 industrial and foundation sponsors (for the complete list see http://globalchange.mit.edu/sponsors/current.html).
7. REFERENCES Alexeeva-Talebi, V., N. Anger and A. Löschel, 2008: Alleviating Adverse Implication of EU Climate Policy on Competitiveness: The Case for Border Tax Adjustments or the Clean Development Mechanism? Centre for European Research (Zentrum für Eoropäische), Discussion paper No. 08-095.
Armington, P.S., 1969: A Theory of Demand for Products Distinguished by Place of Production.
IMF Staff Papers, 16: 159–76.
Babiker, M.H. and T.F. Rutherford, 2005: The Economic Effects of Border Measures in Subglobal Climate Agreement. The Energy Journal, 26(4): 99-125.