If some world regions unilaterally constrain emission intensive activities, the associated changes in relative prices and shifts in trade patterns may increase emissions elsewhere, partially offsetting the primary emission reductions. This is due to three types of effects. First, in energy markets, tighter emission policies lead to a reduction of demand for fossil fuels and thus a price drop in world energy prices, which in turn leads to an increase of energy inputs and carbon emissions in other regions. Second, tighter environmental policies lead to a relocation of energy intensive industries. They move to less regulated countries, and emissions tend to be increased there. Third, there is a Heckscher-Ohlin effect: with tighter policies, domestic supply of energy-intensive goods is reduced; this leads to an increase in their prices, and to an increase in their supply by foreign producers, who increase their energy use.
The simulation results, which were received with a stylised model, show that the leakage effect induced by the energy market is substantial, whereas the other two effects are rather small. The small impact of these other effects can be explained by the (empirically) low production elasticity of energy. Therefore, it is advisable to combine the taxation of domestic energy use with a subsidy on energy imports. This will lead to an increase of international energy prices and, thus, tend to reduce foreign energy demand. An instrument equivalent to the trade intervention is a tax on domestic production of fossil fuels. This would also raise world energy prices and mitigate leakage. Note that this policy recommendation to tax energy use and domestic supply of fossil fuels is in obvious contradiction to most national tax policies that subsidise domestic production of fossil fuels.
On final-goods markets, the optimal instruments in the presence of leakage are import tariffs or export subsidies. Both instruments lower the world market price and reduce foreign incentives to produce energy-intensive commodities. It was shown that optimal tariffs on final goods markets are rather small. The reason for this result is the small size of leakage through final goods markets. The basic results carry over to the case of the non-competitive goods market, where leakage effects are very small, too.
With respect to international environmental agreements like the Kyoto Protocol, it can be said that a stabilisation of such an agreement through trade restrictions is unlikely. However, as the above results make clear, import subsidies or export taxes should be in the self-interest of countries implementing greenhouse-gas reducing policies, and these instruments would probably not be interpreted as sanctions by non-compliant countries.