Does information quality matter in corporate innovation? Evidence from the Korean market

ABSTRACT

Using extensive manually collected data on granted patents, this study examines the effects of the degree of a firm’s information asymmetry on corporate innovation in the Korean market, which is characterized by weak transparency and active firm innovation. Based on four measures of information asymmetry, we find that the quality of information about a firm has a positive influence on its innovation activities. In addition, this influence is more evident in firms with poor corporate governance practices and in Chaebol-affiliated firms. Overall, this study offers insights on the importance of information quality for firms planning investments in innovation, which is a long-term and highly uncertain commitment. An important policy implication is that regulatory authorities should promote the timely and reliable disclosure of information on firms.

KEYWORDS:

• Information asymmetry
• innovation
• corporate governance in emerging markets
• Korean Chaebol

JEL CLASSIFICATIONS:

• F21
• G32
• G34
• O31
• O32
• O33

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The literature shows that individual investors lack access to private information and the ability to process information owing to psychological biases. See, for example, Barber and Odean (2000), Choi and Sias (2012), Grinblatt and Keloharju (2000), and Nofsinger and Sias (1999).

2 The KIPRIS website sets a restriction on downloading by limiting the number of patents that each user can download at one time (500 per user). As our sample relies on hand-collected patent data, which are not readily accessible, our sample is limited to data prior to 2013. Given that the economic and corporate environments have changed since then, future studies may extend the sample period and examine the effect of a change in the information environment on innovation.

3 Our study does not necessarily exclude the possibility that poor accruals quality could also lead to a higher cost of capital, as in Francis et al. (2004, 2005). Francis et al. (2004) show that earnings accrual quality is significantly related to the cost of raising capital. Francis et al. (2005) provide further evidence that earnings accrual quality reflects managers’ opportunistic behavior, and thus that the related information risk increases the cost of capital.

4 In contrast, the U.S. stock market is a quote-driven market and displays only the bid and ask offers of designated market makers, dealers, or specialists. These market makers post the bid and ask prices that they are willing to accept at the time.

5 Based on the strong theoretical and empirical association between the BAS and the level of information asymmetry between investors, many studies in the large body of relevant literature use the BAS as a proxy for information asymmetry (Bagehot 1971; Brockman and Chung 2000; Lev 1988).

6 Acharya and Pedersen (2005), Amihud (2002), and Chordia et al. (2009) report that the AR predicts cross-sectional stock returns, suggesting that information asymmetry risk is priced.

7 To account for the long-term aspect of the innovation process incorporating information related to firms, we consider the relationship between the three-year lagged measures of the firm characteristics and the firm’s concurrent innovation, following He and Tian (2013).

8 Our measures of information quality are not materially different from those used by previous studies. Compared to the measures of Amihud and Mendelson (1986) and of Park (2018), which are based on the U.S. market, our BAS measure seems to be slightly higher on average given that the level of information asymmetry tends to be higher in emerging markets, such as Korea. We compute AR following Amihud (2002) using data from the U.S. market and find that the levels are largely comparable. Moreover, the level of our AR is similar to that of Park, Song, and Lee (2014), who also utilize Korean market data.

9 To account for the skewness of the innovation measure, the dependent variable in our regression model, we trim the outliers from the sample and find consistent results.

10 As a robustness test to address the potential endogeneity related to firms’ information quality, we consider an instrumental variables approach based on Appel, Gormley, and Keim (2016) and find that the results in Table 4 do not change. In addition, to test the potential relationship between forward information quality and current innovation, we use t+1 measures of DA, AQ, BAS, and AR as our independent variables and re-estimate the results in Table 4. We find that the coefficients of the leading measures of DA, AQ, BAS, and AR are not significant. In addition, the adjusted R-squared value is lower for this model, meaning that it does not have as much explanatory power for the lagged innovation variable. This additional robustness test suggests that sample firms with higher information asymmetry measures that are less innovative do not affect our main findings. The results of both robustness tests are available upon request.

11 We are grateful to the KCGS for providing access to their data.

12 The high value of the intercept may be attributable to the size effect because larger firms tend to implement better corporate governance practices. To control for the size effect, we include firm fixed effect dummies in all regression analyses. In addition, we divide the full sample into two subsamples based on firm size and re-estimate the main regressions for each subsample. We do not find materially different results for each subsample, unlike when we divide the sample based on the CGS in Table 5. Hence, we believe that the size effect does not materially alter our main finding on the relationship between corporate information quality and innovation. The subsample results are available upon request.