The Potential Benefits of Trade in Carbon Emissions

Gains from Trade in Carbon Emissions: A Conceptual Framework

This section provides a rough estimate of the potential gains of trade in just carbon emissions between the tropical South American countries and the “rest-of-the-world.”

Figure 1 illustrates this. The horizontal axis measures the remaining forest areas before any settlement, with F-bar being the land area originally covered with forest, or the maximum forest area, and the vertical axis measuring the forest land (rental) value. Schedule NN represents the net marginal value product of the alternative uses of the forest lands (agriculture, logging, etc.). Schedule D^W shows the marginal value product of the standing tropical forest for the world as a whole, and D^L shows the marginal value of the standing forest for the countries where the forest is located. The NN curve is upward sloping, reflecting the increasing opportunity cost that conserving a greater forest area involves for the country that owns the forest. The downward slope of schedules D^W and D^L reflects the increasing cost of deforestation for both the world and the landlord countries as the forest is depleted. The position of the demand schedule D^w is naturally dependent on the availability of forest in the rest-of-the-world. Thus, the analysis is conditional on a fixed level of forest outside Latin America.

FIGURE 1

Schedule D^L includes all the benefits that the standing forest brings to the landlord countries, including protection against slides, floods, and soil degradation and their valuation of biodiversity and carbon sequestration capacity. Schedule D^W is the sum of the D^L plus the benefits for the rest of the world that the standing forests provide (mostly associated with the avoidance of global climate changes and the value of the biological reserves). This schedule is drawn as a convex curve to reflect the fact that the marginal value of standing forests increases very fast as the forest stock declines.

The welfare of the world as a whole would be maximized at point C, where the stock of forest is F^W and its marginal value is P^W. By contrast, the welfare of the landlord countries that own the forest is maximized at point E, at a forest level of only F^L and a rental value of the forested land of only P^L. Deforestation takes time and investments, and thus the landlord countries may not yet be at the point of welfare maximization. Suppose that the current level of the forest remaining is between F^L and F^W, as is likely in most of the tropical South American countries where there is still a large area covered with forests. If the world does not compensate the forest “owners,” deforestation will continue at least to level F^L, with large losses to the world (indeed, it could go beyond F^L if the domestic externalities are not internalized). If the loss of natural forest is more or less irreversible (biodiversity losses, for example, are likely to be irreversible), the best that the world can do is to minimize further losses by stopping deforestation at level F^L₀.

The world can easily, in principle, compensate the landlord countries to stop further deforestation, since the world marginal gains are P_max and the landlord countries’ marginal opportunity cost is only P_min. The actual compensation at the margin will be anywhere between these two prices, depending on the negotiating capacity of the landlord countries’ vis-á-vis the rest of the world. Since the landlord countries also obtain benefits out of the standing forest (represented by Schedule D^L), the rest of the world would only have to pay BH as a minimum and AH as a maximum. Indeed, the minimum total compensation to stop deforestation at F₀ would be the area EBH and the maximum possible total compensation would be the area IEHA. The important thing is that even if the rest-of-the-world paid the marginal price P_max per ha for all the excess (potential) deforestation (F₀ - F^L), it would still have a net gain of IMA.

Forests at a level between the world optimum and the individual country optimum, such as point F^L₀ in Figure 1, probably depict the situation that roughly prevails in most of tropical South America. The situation in Central America and Mexico is perhaps better approximated by a point such as F^C_0, that is, where the current forest levels are below even the individual country optimum. This distinction is important because international cooperation in the form of technical assistance, policy advice, and concessionary projects to decrease deforestation (which are the most common forms of international involvement) is likely to be more effective and encounter true cooperation from the national governments in cases illustrated by F^C₀ than in those represented by F^L₀. In the latter case governments are not likely to cooperate and many international initiatives would be diluted.

In the absence of a comprehensive program of action and compensation to reduce or eliminate deforestation, sporadic internationally financed projects to protect specific sites are likely to be rather ineffective even if the actual projects are successful. The reason is that, if countries are not committed to reducing deforestation, all that these projects achieve is to shift it elsewhere. That is, a government may agree to receive aid to protect specific sites, but unless reducing overall deforestation is a true national priority, more deforestation will be allowed elsewhere. There is a degree of substitution or fungibility in the geographic location of deforestation.⁵

López (1996) used this framework to quantify the value to the world of conserving 650 million ha of tropical forests in South America. The most likely estimate of the net benefits of carbon sequestration (which assumes unit elastic world demand and supply curves) amounts to US$713 billion. This, of course, represents the stock value of the tropical forest in the region in terms of net present value, considering only its carbon-sequestration benefits. To make the analysis more consistent with the global warming negotiations, we consider next the annual flow of resources on the basis of historical rates of deforestation in the region.