The effect of industry leaders’ exploratory innovation on competitor performance

ABSTRACT

This paper examines the effect of an industry leader’s exploratory innovation, defined as the innovation embodying novel knowledge relative to the firm’s extant knowledge, on the performance of its direct competitors. We argue that an industry leader’s exploratory innovation can benefit its competitors, resulting in an average increase in competitors’ sales. The benefit can come from advantageous inter-industry structure, higher perceived status through association, and expanded knowledge pool. The extent of benefit, however, is conditional on the number of competitors in the industry as well as the level of competitors' financial slack and Return on Assets (ROA). Using data on the U.S. computer sector, we find support for our hypotheses. This study suggests that while an industry leader’s exploratory innovation is intended to further its own interests, it also affects the plight of its direct competitors.

KEYWORDS:

• Industry leader
• exploratory innovation
• firm growth
• external knowledge

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ We define an industry leader as a firm with the highest market share within an industry in a given year.

² We chose this time window for consistency as data on patents suffer from right censoring after 2004.

³ In their paper, Katila and Ahuja (2002) use a five-year moving window. Because we are not looking at the effect of exploration on the innovator firm, but on other firms in the same industry, we use a seven-year window. Nevertheless, estimating this variable using the same five-year window they use does not change our results in this paper.

⁴ The full list of the runner-up firm’s market share is as follows: 0.76% (SIC 3571); 0.68% (SIC 3572); 0.89% (SIC 3575); 0.87% (SIC 3577); 0.51% (SIC 3674); 0.36% (SIC 7372); and 0.64% (SIC 7373).

⁵ Given that our dependent variable concerns change in sales and not innovation, we do not have to restrict our sample to firms that routinely innovate. It means that for some observations in our sample we will not observe patenting activity. To perform this test, we replace the exploration value for non-patenting firms as zero.

⁶ The results are available upon request from the authors.