A Test of Alfred Chandler's Theory of Corporate Control

Alfred Chandler, in Scale and Scope: The Dynamics of Industrial Capitalism (1990), suggests that the acquisition of targets is an alternative to direct investment in research and development (R&D). Chandler suggests that the failure of accounting to recognize investment in R&D as an asset may have made R&D less attractive. This study focuses on the relationship between investment in R&D and capital expenditures and a set of partitions based on Chandler's three technology types ("hightech," "stable-tech," and "low-tech") and three possible merger activity classes (acquirer next year, target next year, and neither acquirer nor target next year). Chi-square contingency tables are used to test the independence of merger class and technology type, a frequency test. Regression is used to test the relationship between R&D and sales and between capital expenditures and sales, with the sample partitioned by technology type and by merger class in a 3-by-3 research design. The sample is 23,146 firm years from 1974-1988 for 2,659 firms categorized into industry groups based on Chandler's criteria. The financial data are from COMPUSTAT data files. The frequency of being an acquirer is the same for high-tech and stable-tech firms (11.2 versus 11.5 percent of firm years) and higher for low-tech firms (13.9 percent of firm years). High-tech firms that are acquirers next year have 79% lower investment in R&D (.044 of sales versus .056 of sales) and 77% lower investment in capital expenditures (.071 of sales versus .092 of sales) than "high-tech" firms that are neither acquirers nor targets next year ("baseline" firms) as measured by the estimated slope coefficient in regression. "Stable-tech" acquirers are similar to "stable-tech" baseline firms in R&D investment (.016 of sales versus .016 of sales) and in capital expenditures investment (.072 of sales versus .080 of sales). "High-tech" targets have higher R&D investment (0.062 of sales versus 0.056 of sales) and lower capital expenditures (0.054 of sales versus 0.092 of sales) than "high-tech" baseline firms.