Voodoo forecasting: Technical, political and ethical issues regarding the projection of local population growth

Abstract

The case of an energy production community in Colorado is used to illustrate a) the great need for reliable subnational population forecasts, especially in communities expecting rapid population growth, and b) why such projections, as currently performed, cannot be reliable. Explanations for failure in forecasting are found in the methods themselves, the unavailability and unreliability of key data, politics, and the intrusion of the designers' and users' values into the forecasting models. As one solution it is proposed that forecasting be reconceptualized from a technical to an ethical issue and that it be based upon an active rather than a passive planning/forecasting philosophy. As such, the community and industry would set a desired, realistic goal, and forecasting models would be run backwards to determine the economic-demographic paths by which this goal could be achieved.

Population forecasting and local economic planning: The limits on community control over uncertainty

“…A proposal by Moen (1984) to run models backwards is consistent with the use of forecasts as a teaching tool. She suggests that: "The community and industry would set a desired, realistic goal, and forecasting models would be run backwards to determine the economic-demographic paths by which this goal could be achieved."…”

“…Examples of this kind of model are the OBERS model of the Bureau of Economic Analysis (U.S. Bureau of Economic Analysis, 1980;U.S. Department of Commerce, 1981), the models of the National Planning Association, (e.g., Lee and Hong, 1974) and the model described by Moen (1984).…”

“…As Moen (1984) points out, local governments can benefit from inflated forecasts in order to obtain funds for infrastructure, and industry also has several motives to inflate forecasts, among them the encouragement of local planning and development of housing, services, and infrastructure which will help the industry attract a stable, productive work force. Therefore, the forecaster may come under considerable pressure to generate optimistic forecasts, which, in time, will be erroneous.…”

“…If the forecaster is not constrained to producing a single forecast, the range of plausible forecasts gives the community an understanding of the uncertainty it faces. Moen (1984) gives an example of a community deriving "small comfort" from the forecast that its future population will be 456,000 or 1,248,000 in twenty years. If its future is indeed so uncertain, the role of the forecast should be to produce thought and action, not comfort.…”

Between a rock and a hard place: The evaluation of demographic forecasts

“…'Normative' forecasting, although infrequent in demographic applications, is another approach to reducing stress among the producers of forecasts (Moen 1984). This active approach to forecasting involves first deciding what future outcomes are desirable and then designing policies and actions to achieve them.…”