VoxEU Column Environment

Imperfect climate policy unlikely to increase domestic emissions

By promising to reduce fossil fuel demand in the future, some claim that climate policies will induce supply side responses today; firms will pump out emissions now before demand restrictions tighten. However, this column argues that the ‘green paradox’ is a red herring. Evidence from US coal prices suggests that, in industrialised countries, there is little danger of an increase in domestic emissions in response to imperfect climate policies.

Current implemented policies aimed at reducing greenhouse gas (GHG) emissions are far from perfect and leave owners of stocks of fossil fuels ample scope to increase current (as opposed to future) extraction. Some economists and policymakers fear that emission reduction policies may thereby induce an increase rather than a decrease in CO2 emissions. Should we be worried about this ‘green paradox’?

Current policy

A quick glance at the current state of climate change mitigation policies shows that such policies are nothing like the economists’ prescription of imposing a worldwide price on greenhouse gas emissions that reflects marginal damages caused by these emissions.

  • The development of emission reduction policy usually takes many years before implementation.

Note the time lags between the announcement and implementation of the Kyoto Protocol (ten years) and the EU Emissions Trading Scheme (five years, while European policymakers discussed a carbon tax already in the early 1990s), leaving firms free to emit what they want in the interim.

  • Most policies are aimed at subsidising carbon-free energy sources, either directly or through technology subsidies, rather than at pricing CO2 emissions.

Moreover, few countries have implemented emission-reduction policies in the first place, which leaves the majority of emitters potentially free to benefit from lower fuel prices which result from lower demand in abating countries.

Supply-side responses to demand-side policies

According to Hans-Werner Sinn, such imperfect policies affect the time path of profits of owners of stocks of fossil fuels. As the announcement of a future price on emissions or the increased likelihood of the penetration of carbon-free energy sources deteriorates the prospect of selling fossil fuels in the future, resource owners have an incentive to increase their current extraction, thereby increasing rather than decreasing current CO2 emissions (Sinn 2007, 2012). Sinn called this mechanism a ‘green paradox’ and his thought-provoking contributions did not only inspire other economists to find green paradoxes (cf. Van der Werf and Di Maria 2012 for an overview), but also provoked discussions among policymakers in several EU countries.

Learning from coal and sulphur dioxide policy

Should we be worried about a green paradox? Do resource owners increase their supply in response to imperfect policies, and does this induce an increase in emissions? In a recent paper (Di Maria et al. 2012) we use data on coal use in the US before and after the announcement of a cap on sulphur dioxide (SO2) emissions to answer these questions. In November 1990, the Bush administration signed into law the Acid Rain Program, part of the Clean Air Act Amendments. Under this programme, a group of previously unregulated coal-fired power plants (so-called Phase I plants) would be facing a collective cap on SO2-emissions from 1995 onwards. Clearly, this policy announcement deteriorated the future profitability of coal mine owners, and even more so for owners of stocks of high-sulphur coal, while the group of power plants under consideration still had time to emit before the policy would bite. The green paradox theory hence predicts a fall in the price of coal after the policy announcement (and even more so for high-sulphur coal), an increase in coal use in the interim of the policy (in response to lower coal prices), and substitution towards dirtier coal types (as their prices dropped even more).

We do indeed find a drop in the price of coal after the announcement of the Acid Rain Program. Not only did the average price drop by about 9% but the sulphur premium also increased with 40%, confirming that prices for high-sulphur coal were hit hardest. Were coal-fired utilities able to benefit from lower prices and expand the use of (dirty) coal?

Using a difference-in-difference analysis, with previously regulated power plants as a control group, we find that Phase I plants as a group did not increase their heat input. Focusing on specific sub-groups, we find that only utilities that were flexible on the coal market, and purchased a large share of their input on the spot market rather than through long-term contracts, increased their heat input.

Finally, we tested whether the sulphur content of purchased coal increased. Surprisingly, we find that the sulphur intensity of coal decreased rather than increased. This effect seems to occur especially at utilities in states that required prior approval of compliance plans for the upcoming environmental regulation.

Demand-side restrictions may mute changes in emissions

These findings suggest that, despite the drop in the price of coal, utilities were unable to increase their emissions due to restrictions on the demand side of the market for coal. To what extent may such restrictions mute ‘green paradoxical’ responses to imperfect climate policies currently implemented in industrialised countries?

The imperfect policies that may lead to a green paradox are largely aimed at the burning of coal and natural gas for electricity production, and the burning of petroleum products for transport. There are several reasons why the demand side of the markets for these fossil fuels may be unable to respond to lower resource prices. Not only may coal-fired utilities be locked in long-term contracts on the coal market, but the same may also be the case for gas-fired utilities. In addition, utilities will only increase their input levels when they can sell more of their output. However, not only are estimated price elasticities for electricity quite small, but in addition these electricity technologies compete with nuclear and renewable energy sources. For coal and gas to replace nuclear energy, their prices would need to fall by a large amount. Hence, the demand for coal and natural gas by power plants are not very price sensitive.

Even if utilities were sufficiently willing to increase their output, they may be facing other restrictions. First, there are capacity restrictions. Utilisation rates of existing capacity are high, which leaves little room for output expansion with given capacity. Expanding capacity, on the other hand, is time consuming. In case of announced policies, the period between announcement and implementation may simply be too short to profitably invest in new capacity. Both capacity expansions and increased utilisation rates may however be facing restrictions from existing (local) environmental and planning regulations.

Such restrictions on the demand for fuels are not only relevant for coal and gas, but also for oil. Although flexibility on the spot market for oil may be no problem for refineries, demand elasticities for gasoline are low, capacity constraints for refineries are present and there is a significant amount of local or national regulation of emissions by automobiles (at least in industrialised countries).

International carbon leakage

Our additional findings, alongside Di Maria et al. (2012), give reason to believe that there is little danger of an increase in domestic emissions in response to imperfect climate policies, when implemented in industrialised countries. More danger may come from international carbon leakage. When industrialised countries reduce their demand for fossil fuels, international prices may fall, inducing an increase in demand from other countries. In addition, domestic policies may increase the price of energy-intensive goods produced in industrialised countries, which may increase the demand for such goods produced in countries without emission reduction policies. Both channels may induce an increase in CO2 emissions in the latter group of countries, thereby (partly) offsetting the emission reductions of countries with demand-reducing policies. It seems that the restrictions mentioned above apply less to developing countries than to industrialised countries – especially where existing environmental regulation and local planning consent policies may be absent.

There is much scope for improvement in current climate policies of industrialised countries (Helm 2012), especially in preventing emissions leakage, but imperfect climate policies are not likely to induce a green paradox from domestic emissions.


Di Maria, C, I Lange, and Van der E Werf (2012), “Should we be worried about the green paradox? Announcement effects of the Acid Rain Program”, CESifo Working Paper, 3829.

Helm, D. (2012), The carbon crunch: how we're getting climate change wrong  and how to fix it, New Haven, Yale University Press.

Sinn, H W (2007), “Greenhouse gases: Demand control policies, supply and the time path of carbon prices”, VoxEU.org, 31 October.

Sinn, H W (2008), The green paradox: a supply-side approach to global warming, Cambridge, MA, MIT Press.

Van der Werf, E, and Di Maria, C (2012), “Imperfect environmental policy and polluting emissions: the Green Paradox and beyond”, International Review of Environmental and Resource Economics, 6(2), 153–194.

420 Reads