Strategic Trade Policy
Winners and losers

Much recent research has examined the possible merits of interventionist trade policies in oligopolistic markets, in which export subsidies may lead to `profit shifting' towards domestic firms at the expense of foreign rivals. Since this argument stands in complete contrast to traditional trade theory based on competitive markets, it has been subjected to extensive scrutiny. In Discussion Paper No. 560, Research Fellow J Peter Neary addresses two largely neglected aspects of the issue. He first determines whether government should subsidize domestic firms that would be highly profitable on their own or those that are relatively less competitive, and he then assesses whether such subsidies can justified when the budgetary costs of financing them are taken into account.

For a simple model in which a domestic and a foreign firm produce identical products and sell them to a third market whose demand is described by a linear function, the optimal policy is an positive export subsidy, provided that raising revenue is costless, whether the foreign government retaliates or not. Even a mild asymmetry between social and private costs will render an export subsidy undesirable, however, while if the marginal cost to the home government of raising one unit of tax revenue rises above 1.33 the optimal policy is an export tax. Further, even if raising revenue is sufficiently `cheap' for a subsidy to remain optimal, its value should <MI>increase<D> with the cost competitiveness of the domestic firm: if governments are to subsidize at all, they should help `winners', not `losers'. This purely static result contrasts sharply with the `infant-industry' argument, since it relies instead on domestic firms' ability to `capture' foreign profits. Governments should therefore target assistance on firms that enjoy a `comparative advantage' in profit shifting, i.e. those that are highly cost-competitive from the outset.

Neary then investigates the robustness of these results in more general models. While the simplicity of the linear results does not survive generalization, their general thrust does: the threshold value at which the optimal policy switches from a subsidy to a tax can no longer be calculated exactly; but it is as likely to lie below 1.33 as above, and it cannot exceed 2, which still lies within the empirically plausible range. Governments should still subsidize the more competitive domestic firms disproportionately, especially if the foreign government also assists its own firms. Neary finally considers these issues in a framework where firms compete on price rather than quantity and governments provide levels of subsidy related to the prices charged by the firms, which they anticipate in negotiating export contracts. Again the optimality of export subsidies appears subject to the same qualifications as in the other cases of cost asymmetry considered above.

Cost Asymmetries in International Subsidy Games: Should Governments Help Winners or Losers?
J Peter Neary

Discussion Paper No. 560, July 1991 (IT)