Some new results in testing for economies of scale in computing 1985 and 1988 data

Abstract This paper presents and evaluates a new specification of models used to test for the presence of economies of scale in computer processing power. An analysis of data for systems available in 1985 (which has been analyzed previously in the literature) is consistent with this specification. Analyses of a new set of data for systems available in 1988 are also performed using the same method. These results are also consistent with the specification, and in addition imply statistically significant diseconomies of scale. A discussion of some managerial implications of the results is also presented, as well as suggestions for future research.