A major problem affecting public projects in many developing countries is that of inadequate standards of operation and maintenance (O and M). There are various reasons but the overriding one is the shortage of public funds and foreign exchange. Despite the efforts now being made to alleviate the problem it is likely to continue to be a major constraint which should be taken fully into account in the planning and analysis of public projects. In the planning of irrigation, water supply, power and other projects a choice often exists between technical alternatives with high capital costs and low recurrent costs and others with the opposite pattern of expenditure. Apart from technical factors the final selection is usually based on least‐cost economic analysis, using the conventional discounting methodology whereby all costs, both capital and recurrent, are discounted at what is taken to be the opportunity cost of capital (usually 8 to 15%). These relatively high discount rates favour low capital — high recurrent cost alternatives, even though in practice these are often not desirable, because of the O and M difficulties that will occur. In many developing countries capital funds for public projects are readily available, from foreign aid, whereas recurrent finance, which is normally provided by the government, with little foreign assistance, is much scarcer. It is argued that in such cases the use of a single discount rate for all types of expenditure is unrealistic. A better . approach would be to apply different interest rates for different types of expenditure, to take account of their differing opportunity costs. In the situation of capital abundance and recurrent funds scarcity, for example, capital costs would be valued differently from recurrent costs. This would be done by means of shadow pricing, the costs for the various types of expenditure being adjusted to take account of their relative opportunity cost, in the same way as adjustments are made for unskilled labour, foreign exchange and other factors in conventional economic analysis.