responsibility, that is so vital for the efficient operation of modern integrated oil industries.
Rejoinder Global energy models: a comment The paper by E. L. Allen, J. A. Edmonds and R. E. Kuenne, ‘A comparative analysis of global energy models’,’ stressed the inherent inaccuracy of forecasts of future levels of energy demand and supply. It evaluated the margin of error discernible from a comparison of various different forecasts for the year 1985. This it found to be some 4-5 million bbl/day of oil for the non-communist world in that year. Since forecasts are not made for their own sake but in an attempt to descry the future and thereby derive pointers for current decision making that may help one to ‘get from here to there’, we should understand the practical conclusions to be drawn from the paper. In my view these are : (1) The currently irreducible margin of error in forecasting demand for just 5 years ahead is roughly equivalent to more or less the production of any one of a number of mediumsized oil countries, Venezuela, for example. In global terms it represents a swing of lo- 15% of the total call on OPEC oil foreseen for 1985 (Table 1 of the paper). It is clear therefore that such imprecise forecasting will not help anyone to fine-tune the supply/ demand balance or predict the future course of oil prices. However, it does indicate very clearly that the only realistic way to improve upon the marked instability of energy markets is by working to increase the margin of potential supply over demand, a task that involves the pursuit of both moderation of demand and enhancement of supply availability. (2) As to projections of future supply I would recommend that the authors of the paper avoid the trap of ‘plausible surmise’. Future levels of OPEC production will not only (perhaps not even mainly) be determined by policy aims, geological resource base, or demand projections. A major determining feature will always be the organizational and managerial capability of the various national companies charged with doing the
work of exploration, production development, and all the other tasks involved in getting oil to markets. One needs to evaluate this aspect most carefully, since today’s oil producers are developing countries that face massive problems of national development and modernization affecting the whole of their social and economic fabric. Developing countries promoting change of such amplitude in their national life commonly encounter severe difficulties, specifically in the organization and management of large public and private enterprises. This is because of unfamiliarity both with modern technology and also with the sophisticated modern methods needed to handle large bodies of specialized professional and technical (mainly whitecollar) employees, whose efforts need to be most carefully directed and coordinated if they are to be truly effective and productive. It should not escape the attention of energy analysts that modern systems of organization and management may not be easy to implant in the more traditional, paternalistic environments characteristic of many developing countries. The cultural (and political, and managerial) traditions that gave these countries their stability in the past, in future may impede their efforts to develop the practice of delegated management
(3) Of course we could make better forecasts if we knew more about the future. To pursue such a line of endeavour is unproductive because it is an illusion. The authors have done something more truly useful: they have attempted to quantify the ‘range of uncertainty’ surrounding our more or less mechanistic predictions. In so doing they teach us the uselessness of predictions for any attempts at fine-tuning, and therefore must needs make us examine more practicable broadbrush approaches to stabilizing energy markets. For political and industrial decision makers having to face up to these practical problems, neither the demand side nor the supply side will be easily or quickly resolved. The former, because of the ingrained habits of consumption derived from years of cheap energy. The latter, because of the learning process that most producing nations will have to undergo in developing and testing the managerial capabilities of their respective national oil companies, both for their own national purposes and also in attempting to serve the needs of an evidently energyhungry and inter-dependent world.
R. A. Irving East India Club London
lE. L. Allen, J. A. Edmonds and Fi. E. Kuenne. ‘A comoarative analvsis of alobal energy models’, knergy Econbmics, iol 3, No 1, January 1981, pp 2-l 3.
The authors reply It is always a pleasure to find that one’s scholarly endeavours have stimulated others’ thought and guided them to further conclusions. Mr Irving’s complimentary comments on our paper are most welcome, and we find ourselves in broad agreement with his conclusions - with some reservations. (1) Modern econometric modelling - mixed with a great deal more ‘intuitive’ input than most modellers readily admit to - is an extremely imprecise forecasting tool (but, then, what isn’t?). ‘Order-of-magnitude’
exactitude is more frequently a realistic desideratum than Mr Irving’s ‘fine-tuning’. But there is a danger in implicitly accepting the major function of modelling as projection of an unforeseeable future. Far more important, we feel, is the potential of models - even those with wide forecast error potentials - to yield insights into the complex structure of economic phenomena, to provide a framework of analysis for current policy problems, and to stimulate debate. (2) We somewhat stubbornly cling