Designing multimarket-contact hypothesis tests: Patent citations and multimarket contact in the product and innovation markets of the chemicals industry

This paper discusses the design of hypothesis tests about multimarket contact and uses a sample of chemicals firms to illustrate researchers' choices and to provide new evidence about the effects of multimarket contact in product and innovation markets. Multimarket contact has a large, stable and statistically significant association with a firm's citations of the patents of another firm. The effect is present even when controls are added for each paired firm's pertinent characteristics (including the effects of its product and innovation markets) that would be expected to affect citations or reflect the effects of the research underlying the citations. New probability measures of multimarket contact are introduced and discussed, and their performance is compared with simple count measures.