Realism In Statistical Demand Forecasting: The Econometric Challenge

It is four years since the seminal study of demand for rapid excavation by the National Academies of Sciences (NAS) and Engineering outlined this technology's opportunities, requirements and significance to the challenges of urbanization and natural resource conservation. The demand study was quite simple. Opportunities were valued at market while requirements to produce new technologies were valued at cost. The significance of new technology in economic terms was measured by the net discounted gains to users over time. The Academy view in 1968 was that these gains, even most modestly estimated, appeared large enough to justify the bill for research proposed by that Committee (1). In 1970 the question of demands became international in what is recognized as the first international experiment in technological policy analysis, terminating in the Advisory Conference on Tunneling of the Organization for Economic Cooperation and Development (OECD)(2). A monumental demand survey formed the basis for its economic assessment. The twenty-two OECD nations also made an impressive attempt to measure the magnitude of existing research and development efforts. In their recommendations periodical statistical demand forecasting and collection of demand data (Recommendation 1) was given central importance for public planning (Recommendation 2), the financing of public construction (Recommendation 3) and the encouragement of technological advance (Recommendation 4). Their program for international cooperation led to OECD's establishment of a Directorate for Environmental Affairs, and influenced the organization of the World Environmental Conference meetings now held in Stockholm under United Nations auspices. The OECD Conference was the first example of "an attempt to apply the concept of technological policy analysis to a 'civilian' technology having potentially