The economic base model is frequently used by planners and forest economists to quantify the effect of timber and other resource management programs on local economies. The objective of the study was to test the assumption, usually employed in economic impact analysis, that basic activity (activity in industries producing for export) affects but is not affected by derivative activity (activity in industries not producing for export). Flathead County, Montana, where the economic base is dominated by the timber and other natural resource industries, was chosen for the case study. The tests were based on the concept of Granger causality: one variable "causes" another if the second can be better predicted by using past information on the first than by not using that information. Results challenged the assumption that changes in the basic sector were independent of changes in the derivative sector. Specifically, a bidirectional (feedback) relationship between basic and derivative sectors was observed in three of four tests using data on wages and salaries. A fourth test supported the unidirectional assumption from basic to derivative; two tests using monthly employment data failed to reveal any relationship between basic and derivative. Planners and forest economists are cautioned that static multipliers will present an inaccurate view of economic impacts when the basic sector is affected by the derivative sector. Forest Sci. 31:717-725.