Although there is by now an extensive body of literature on problems of resource depletion, with few exceptions the literature has been concerned with depletion in closed economies. Very little has been written on an important policy question facing resourceendowed open economies, namely, the degree to which producers should be encouraged to process resources domestically as opposed to exporting them in raw form. Many such economies, Canada being the case discussed here, do in fact subsidize natural resource processing. The case for encouraging domestic processing rests, of course, on arguments relating to market failure, the most persuasive form of which in depletable industries concerns the tendency for private and social discount rates to differ. If, for example, the degree of mineral processing by integrated private resource companies were shown to be inversely related to the discount rate, this would constitute a prima facie case for subsidization. Whether this may happen, as argued in the remaining sections of the paper, depends critically on the malleability of capital in processing. In an open economy with completely malleable capital Dasgupta, et al. have recently shown that the degree of processing would be independent of the discount rate, implying, at least in a first-best world, that there would then be no grounds for encouraging domestic processing. The opposite putty-clay assumption is more interesting and, given the typically long lifetime of mineral capital equipment, possibly more realistic. In this case, because once ore-processing capacity is in place it is fixed, its productivity depends on the length of time it will be used, which in a depletable industry depends in turn on the discount rate. As Gordon (1966) first showed in the context of mineral extraction, the relationship of processing capacity with the discount rate is not onesided. This is because an increase in the rate of discount, while shifting extraction to the present and tending to increase the demand for processing capacity, also increases the opportunity cost of installing that capacity and reduces its supply. It is found by simulation, however, that with parameters representing the situation of the Canadian copper industry, the supply outweighs the demand effects, and, except possibly on short-term employment grounds, the present policy of subsidizing domestic processing appears suboptimal. The subsequent sections of the paper are organized as follows: In a model with perfectly malleable capital, we first demonstrate the Dasgupta result of independence of the level of processing from domestic preferences, a conclusion that follows from the standard free-trade condition that the domestic and foreign marginal rates of transformation are equal. The alternative fixed capital model is then developed, and we describe the resulting time paths of extraction, processing, and mineral exports and show the relationship of processing capacity with the discount rate. The third section summarized the processing incentives in Ontario, one of Canada's largest mineral exporting provinces, and simulates the model with parameters derived from the Canadian copper industry. Tentative conclusions are offered in the final section.