Hand-outs that can damage the world

For more than a decade now, environmental economists have drawn attention to a major cause of environmental degradation - subsidies. A subsidy can take many forms, but its most common manifestation appears as direct cash transfers from government to industry. Those transfers may be aimed at reducing costs of production or at making up the difference between the price secured in the market and some notion of a "guaranteed" price. The former are commonplace in the energy sector, the latter in the agricultural sector. The prima facie case for supposing that subsidies harm the environment is that, by reducing costs, or providing money for output that no one really wants, they encourage wasteful use of materials and energy, or simply encourage overproduction which in itself uses up materials and energy.

In practice, subsidies are immensely complex. Many are unquestionably damaging. An example might be the practice in major tropically forested countries of providing cash incentives for clearing forest land for agriculture or livestock. Subsidies to irrigation water encourage over-use of water and hence waterlogging and soil salinisation. But a subsidy to burning kerosene, say, may be environmentally beneficial because it reduces demand for fuel-wood and hence deforestation.

Deciding what is and what is not environmentally benign is hazardous. On balance, however, the consensus among economists is that the bulk of subsidies in the world - the perverse subsidies - harm the environment.

Norman Myers and Jennifer Kent have tried to put together what we know about subsidies. They go beyond the normal definition of subsidies to include "environmental externalities" - the money value of the damage done to the environment for which no compensation is paid. While the intention is sound, inclusion of externalities muddies the water considerably. Is the relevant external cost the damage associated with the subsidy only or the damage associated with the subsidy and the failure to price the resource correctly in the first place? Glossing over such issues reveals the authors' limited theoretical underpinnings to the numbers they quote.

They also fail to observe that measuring subsidies requires a benchmark and the benchmark they opt for is cost of production. But economists also discuss the benchmark of world prices: the price that could be obtained if a given product or resource was sold internationally. Suppose that fuel oil sells for Pounds 200 a tonne in a country but, if exported, it could secure Pounds 300 a tonne. The domestic market is effectively being subsidised at Pounds 100 per tonne. World prices at least provide a consistent benchmark. Unfortunately, the numbers quoted in this volume are mixtures of estimates based on very different benchmarks. Adding them up is therefore a doubtful exercise. Adding them up and classifying some as perverse and others as not perverse is even more doubtful. Further adding a mix of carefully estimated and downright eccentric estimates of "external costs" makes the whole thing unscientific.

But the message in this book is, of course, correct. From a policy standpoint, the very first thing to do in environmental policy is to pay attention to subsidies. Economists have been saying this for some time, and organisations such as the World Bank in Washington and Organisation for Economic Cooperation and Development in Paris have produced by far the most robust work in this area and subsidy reform is a central plank of their policy agenda. Myers and Kent are also right to draw attention to the subtle nature of many subsidies. Selling a concession to log a forest has a subsidy element if the relevant government fails to tax the profits at the highest rate it can, for example. Other subsidies have their origins in strange historical events. Aviation kerosene, for example, is not taxed because of an international agreement to that effect. Yet it is an incentive to under price air travel, thus increasing air pollution and noise nuisance.

The Myers-Kent volume is a brave attempt to put some order into the many papers on subsidies and we should be grateful for that. Sadly, it treats all of the literature as being of equal value and wiser counsel should have been obtained here. While cautioning the reader about deficiencies in the data, it nonetheless makes unsupported claims on the basis of that literature. Above all, some clearer theoretical underpinnings are required. None of this diminishes the importance of repeating a well-known message: if you want to start saving the environment, stop financing its destruction.

David Pearce is professor of environmental economics, University College London.