Assumptions about the GM technology and seed industry structure

Before analyzing several outcomes from GMVs, let us specify assumptions about their benefits and cost structure. We assume that each traditional variety can be modified, but it may also be replaced by a "generic" GMV. The exact outcome depends on costs, constraints, and decision making about seed supply.

For simplicity, we assume that GMVs improve seed productivity. This corresponds to yield-increasing GMVs (e.g., Bt cotton in India or South Africa; see Qaim and Zilberman, 2003).4 Let Mp'"[A) denote the marginal productivity (Mp) of land planted with GMVj. We will assume that (a) GMVs have higher marginal productivity than the traditional variety, Mp™(A) >Mp°j[A) and (b) the marginal productivity gain declines with A, d{Mp'j(A)-Mp0j(A))/dA<0

This assumption that better-quality lands gain more from genetic modification is done for convenience, and most results hold for broader circumstances. It corresponds to situations, say, where a certain percentage of the crop is lost to pests. GMVs reduce pest damage and thus provide higher gains to locations with higher potential output.

If traditional varieties are replaced by a generic GMV, thenMpgj(A)=cffj(A)/dA is the marginal productivity of land in location

A. We assume that the Mp of the unmodified version of the generic variety is less than that of variety j. Because modification increases the marginal productivity of the target variety, it is assumed that the marginal benefit of the generic GMV, while less than that of the GMVj , is greater than the traditional variety. Hence,

3 Farmers either keep a portion of seeds or receive seeds at no cost from the government. Pricing seeds will add complexity but not change the results.

4 The model can be adjusted to also allow cost reductions and environmental advantages. The gain in details is costly in terms of complexity and does not lead to additional insights.

The introduction of the GM technology is associated with several cost categories. The first is the fixed cost to introduce the GMVs at the crop level. It consists of research, testing, registration, and regulatory compliance costs to introduce, say, Bt cotton in South Africa or Bt sweet potato in Kenya. We assume that this cost has been incurred and consider two other costs.

(a) Fj = fixed cost to modify variety j. This cost includes both the technical cost to insert genetic materials into a specific variety and the cost of IPR transactions and biosafety regulation. Countries with a more advanced breeding sector are likely to have lower modification costs. The IPR component depends on specific circumstances. If a private company already controls a local variety, it will not face any IPR cost. On the other hand, it may have significant cost if a competing company controls a specific variety.

(b) Vj = per acre costs of GMVs of variety j. We assume that these variable costs are constant. They include the physical variable cost of producing GMV seeds and the IPR and marketing costs that a manufacturer has to pay in order to be able to sell the seeds. Let Vg be the per acre cost of the generic seed variety when it is used. It is assumed to be smaller than V-. It is imported from a low-cost production center and does not require payment to owners of the rights for local seeds. The difference decreases as the seed sector producing GMVj becomes more advanced and when the seed producer does not face IPR costs for the use of the local variety.

The outcomes with GMVs depend on the cost to introduce them as well as the structure of the seed industry and the constraints it faces. We consider two patterns. The first pattern applies to countries where public research institutions develop GMVs, which are sold to farmers by a competitive seed sector. In the second pattern, GMVs are introduced and marketed by a monopolist. This scenario is appropriate for the developed world, where major companies such as Monsanto control the GMVs available in the market. In the case of a monopoly, the seed industry may face IPRs and other constraints that impede its capacity to modify traditional local varieties. Recognizing these constraints, we derive outcomes for four stylized scenarios presented below.

0 0

Post a comment