Corporate innovation strategy and disclosure policy

We take the view that public disclosure made by the firm to capital market investors is one venue through which competitors learn about the firm’s operation and R&D activities. This view is supported by prior studies (e.g., Li 2010) showing that competition from existing rivals decreases the quantity of firm’s disclosure to the capital markets, as proxied by management forecasts on earnings and capital expenditures.

Although exploration and exploitation are two distinct types of innovation strategy, prior studies also suggest that companies rarely make exclusive choice between them. March (1991) suggests that maintaining an appropriate balance between exploration and exploitation is critical for firm survival and prosperity. Therefore, instead of using an indicator variable to partition between exploratory and exploitative innovation strategy, we examine the intensity of exploration versus exploration.

The idea is that a firm’s workforce with long tenure and little mobility may hinder exploratory innovation. This is because stagnant workforce may fail to refresh itself in a timely manner, can no longer keep current with technological developments, and grow unable to offer new ideas into corporate activities (including R&D activities). Prior management literature notes that long tenure is often associated with rigidity and a commitment to established policies and practices that potentially kill the entrepreneurial spirit and hinder novel creation (Marcus and Goodman 1986; Tushman and O’Reilly 1997). March and March (1977) find that executives with short tenure contribute fresh insights and are more willing to take risks that deviate from industry norms. Jia (2017) study board tenure and find that firms with a higher portion of outside directors enjoying extended tenure have significantly lower exploratory innovation intensity. So we expect InventorMobility to be positively (negatively) related to exploitation (exploration) intensity, but is unlikely to directly affect management forecast behavior.

Available at http://​www.​google.​com/​googlebooks/​uspto.​html. There are a number of other studies that used this data source, including among others, Chien (2011), Weatherall and Webster (2014), Jia et al. (2016), and Tian and Ye (2017).

We use patent application year instead of patent grant year because prior studies (such as Griliches et al. 1987) have shown that the former is superior in capturing the actual time of innovation.

Jia and Tian (2016) show that the patent application process on average takes 2 years. So as an alternative measure, we replace the key variable of interest by exploration intensity in year t + 2, results are qualitatively the same as those reported in Table 2.

Available at http://​dvn.​iq.​harvard.​edu/​dvn/​dv/​patent. See Lai et al. (2013) for details about this database.

As a robustness check, we also used a 3-year window and a 5-year window. Results remain qualitatively unchanged.

Some firms nay not report R&D expenses because they do not investment in R&D. To ensure our findings are not sensitive to this issue, we refined the R&D dataset by removing observations that have missing R&D in year t and do not have any patents in the next 3 years (year t + 1, t + 2, and t + 3). Although this approach may not perfectly identify firms that truly have no R&D in year t and therefore didn’t report it (appear as missing in Compustat), nevertheless this approach is one reasonable strategy to identify such firms. The underlying idea is that R&D investment made in a year is expected to generate some outcome in the next 3 years.

A number of prior studies have used institutional ownership to capture different types of investors. For example, Nofsinger and Sias (1999) use institutional ownership to partition shareholders into institutional and individual investors and examine herding and feedback trading of these two types of investors. Yan and Zhang (2009) use institutional ownership to examine the relation between institutions’ investment horizons and their informational roles in the stock market. Aghion et al. (2013) utilize institutional ownership to examine the role of institutional investors on corporate innovation performance.

Barron et al. (2002) suggest an alternative explanation to the results reported in Table 9. They hypothesize that analysts’ earnings forecasts for firms with a higher composition of intangibles contain higher proportions of private (or idiosyncratic) information relative to common information. Such effect reflects the disagreement arising from analysts placing greater reliance on their own idiosyncratic knowledge and skill relative to the common information they infer from sources such as current earnings. Their empirical results reveal that analyst consensus is negatively related to the degree to which a firm is comprised of intangibles, indicating that forecasts of earnings for high-intangible firms contain a higher proportion of private information. We find that exploratory firms’ earnings forecasts are less informative and such firms are less likely to disclose innovation-related information, so there is less “common information” available to and shared by analysts (Footnote 1 of Barron et al. 2002). As a result, we find a positive relationship between exploration intensity and analyst forecast dispersion (and error). This also reflects a poorer overall information environment for exploratory firms.