climate system, but also due to the fact that climate policy also reduces SO2 emissions, reducing atmospheric aerosols that lead to a net cooling. In other words, impacts on agri­culture in 2050 are similar in the stringent policy case as in the reference run (see 5.3.3.4). Uncertainty does not come from different variants—but differences in climate sensitiv­ity. In the longer run, however, the temperature of the policy case will remain significantly below the reference case.

5.4.2 Trade policies and international market constraints

Support policies and border protection of wealthy OECD countries, valued at hundreds of billions of dollars each year, cause harm to agriculture in developing countries. Evaluating the overall effects of subsidies and trade protec-

tion policies of industrialized countries on developing coun­tries and poverty within developing countries is challenging. The evaluation must rest on counterfactual simulation of alternative policy variants leading to a diverse set of policies and application of a range of different models.

     As part of this report, GTEM was applied to two hy­pothetical variants representing two alternative global trade policy regimes. Variant 1 represents a global economy in which import tariffs (and tariff equivalents) on all goods are removed incrementally between 2010 and 2020 across the globe. Variant 2, on the contrary, represents a world in which trade barriers will escalate gradually between 2010 and 2020 such that by 2020 these barriers will be equivalent to twice the size of the existing tariff (and tariff equivalent) barriers across the board.