Printer Friendly

Corporate governance, product market competition, and the wealth effect of R&D spending changes.

This article examines whether corporate governance and product market competition interact to affect the profitability of corporate research and development (R&D) investments. Firms announcing R&D spending changes experience positive and significant wealth effects, and these effects are mainly driven by good-governance firms. Investors appear to view announcements of R&D spending changes undertaken by firms with stronger shareholder rights as evidence of value creation. Moreover, the favorable wealth effects are stronger for good-governance firms in noncompetitive industries than in competitive industries, supporting the argument that good governance substitutes for product market competition.

Prior studies have analyzed the relation between research and development (R&D) investments and the firm's market value (Griliches, 1981; Jaffe, 1986; Hall, 1993; Chan, Lakonishok, and Sougiannis, 2001; Eberhart, Maxwell, and Siddique, 2004; Hall and Oriani, 2006; Oriani and Sobrero, 2008, among others), as well as stock returns following corporate R&D announcements (Chan, Martin, and Kensinger, 1990; Woolridge and Snow, 1990; Szewczyk, Tsetsekos, and Zantout, 1996; Sundaram, John, and John, 1996; Xu, Magnan, and Andre, 2007, among others). Typically, empirical results show that firms' R&D expenditures are positively related to current and future market values as well as future profitability. In addition, stock prices react positively to announcements of R&D increases.

Prior studies have also examined factors affecting the valuation effects of corporate R&D strategies. Empirical results show that R&D intensity, firm size, investment opportunities, debt ratio, market concentration, and institutional ownership have positive effects on the ability of R&D investment to create firm value (e.g., Chan et al., 1990; Doukas and Switzer, 1992; Chauvin and Hirschey, 1993; Sundaram et al., 1996; Szewczyk et al., 1996; Cannolly and Hirschey, 2005; Pindado, Queiroz, and Torre, 2010), whereas free cash flows and capital intensity have negative valuation effects on R&D (Szewczyk et al., 1996; Zantout, 1997). These findings suggest that there can be distributional consequences to announcements of R&D investments.

An unanswered question in the literature is whether corporate governance mechanisms affect investor valuations of R&D spending. In addition, the literature has not investigated whether governance mechanisms and product market competition interact to affect the efficiency of corporate R&D expenditures. (1) To the extent that R&D investments are subject to high levels of information asymmetry between managers and investors, R&D activities may give rise to agency concerns (Aboody and Lev, 2000; Ho, Tjahijapranata, and Yap, 2006; He and Wang, 2009). (2) Without proper monitoring and bonding, managers can and often do make suboptimal investment decisions to extract private benefits at the expense of shareholders. Therefore, well-designed governance mechanisms should channel managerial effort toward the efficient deployment of corporate R&D resources. Giroud and Mueller (2010, 2011) find that firms in noncompetitive industries benefit more from good governance than do firms in competitive industries. Hence, it follows that the extent to which the expropriation or rent-seeking potential of a firm's R&D resources is fully realized will depend on both the effectiveness of corporate governance and the degree of product market competition in deterring unproductive resource deployment by managers and in providing managers with the proper incentives to maximize current shareholder wealth.

We argue that the wealth effect associated with announcements of corporate R&D spending changes is likely to be more favorable for well-governed firms in noncompetitive industries than in competitive industries. Unlike previous studies on announcements of R&D expenditure increases, we focus on firms announcing changes in R&D spending, which include both increases and decreases in R&D expenditures. (3) Hence, our investigations enable a comparison between these two subsamples and provide a better understanding about the nature of corporate investment policy in R&D changes. Furthermore, because little research examines the valuation of R&D reductions (Chan, Lin, and Wang, 2015), exploring the announcement effects of R&D changes expands the literature on the valuation of R&D spending. Following the prior literature (e.g., Gompers, Ishii, and Metrick, 2003; Bebchuk and Cohen, 2005; Core, Guay, and Rusticus, 2006; Bebchuk, Cohen, and Ferrell, 2009; Wang and Xie, 2009), we use four indices of antitakeover provisions to proxy for the quality of corporate governance. (4) Product market competition is measured using the Herf indahl index (HI), and the wealth effect is measured in terms of abnormal stock returns using the market-adjusted returns model and the trading volume around the announcements of R&D expenditure changes.

In a sample of firms announcing changes in R&D spending from 1988 to 2014, we find that the average announcement-period abnormal return for announcing firms is positive and significant. Good-governance firms mainly exhibit a positive effect. Hence, R&D changes undertaken by firms with stronger shareholder rights appear to increase investor confidence that the R&D activities are indeed value creating. Furthermore, good-governance firms in noncompetitive industries earn more favorable R&D announcement stock returns than do good-governance firms in competitive industries. Finally, we find a significant increase in trading volume turnover around the R&D announcement days for good-governance firms in noncompetitive industries. For both R&D-increasing and R&D-decreasing subsamples, we also find a similar pattern of higher R&D announcement returns for good-governance firms in noncompetitive industries than in competitive industries. These findings hold for all of the corporate governance indices and are robust to alternative test implementations and controlling for endogeneity concerns.

Our study generally shows that the positive wealth effect of R&D expenditure announcements as presented in prior studies is strengthened after controlling for the type of corporate governance mechanism at the firm. In particular, a simultaneous consideration of governance mechanisms and product market competition is crucial for stock price reaction to announcements of R&D spending changes. A substantial portion of the market value gains for announcing firms arises from investors being aware of the benefits from good governance in noncompetitive industries, where a lack of competitive pressure fails to enforce discipline on managers. Investors appear to attribute the interaction between governance mechanisms and the degree of competition as a significant factor in a firm's ability to profit from its R&D activities. These findings provide support for the argument that the comparative gains of good-governance firms over weak-governance firms in generating abnormal returns from R&D expenditure announcements are conditional on the degree of product market competition. As a result, we expand the agency-based view prior studies provide for why the protection of shareholder rights is related to shareholder wealth (e.g., Gompers et al., 2003; Masulis, Wang, and Xie, 2007) and contribute to the literature that documents a link between corporate governance and product market competition (e.g., Aggarwal and Samwick, 1999; Cremers, Nair, and Peyer, 2008; Giroud and Mueller, 2010, 2011).

In keeping with the resource-based view, our findings also have implications for managerial decisions on resource allocation to R&D activities in that the firm-specific nature of such activities is particularly important for helping a firm sustain its competitive advantage (Chan et al., 2001; Ho et al., 2006). By showing that the market contextualizes the quality of corporate governance as well as the degree of market competition upon announcements of corporate plans to change R&D expenditures, our study provides new insights into how the interaction between governance and competition affects the extent to which a firm's innovative resources can create shareholder value.

The remainder of the article proceeds as follows. Section I describes the data and methodology. Section II presents the empirical results. Finally, Section III summarizes the main conclusions and implications.

I. Data and Methodology

A. Sample Selection

We search for the initial announcements of R&D expenditure changes from the Dow Jones Factiva database, which provides news-service articles and selected stories from the Dow Jones News Wire and the Wall Street Journal database. Because an announcement is often reported by several news media, we make sure that we have the date and source of the first public announcement of the event. The Dow Jones News Service, PR Newswire or Business Wire, is typically the first to report it. We select the words and phrases and their synonyms commonly used to describe R&D spending changes as keywords for a database search routine. (5) Our sample consists of firms that voluntarily announce plans for the forthcoming fiscal year to increase or decrease company-sponsored R&D expenditures over that of their previous fiscal year. The sample focuses on firms listed on the New York Stock Exchange (NYSE), American Stock Exchange (AMEX), and NASDAQ and covers the period from January 1988 to December 2014. We exclude American depositary receipts and noncommon stocks. For inclusion in the final sample, a firm must have the necessary data on stock price and financial statement items to estimate the wealth effect, along with data having positive values for total assets, total liabilities, stockholders' equity, and sales. We obtain such financial data from Compustat tapes and data on stock price and returns from the Center for Research in Security Prices (CRSP).

Our selection procedure results in a final sample of 723 announcements made by firms covered by the RiskMetrics database. Table I summarizes how we arrive at the final sample. Table II lists the sample distribution by Fama-French (1997) 48-industry classification in Panel A, by time profile in Panel B, and by R&D spending change type in Panel C. It shows that the majority of the announcing firms come from pharmaceutical products and electronic equipment industries, which constitute about 54% of the total sample, and the number of R&D spending change announcements increases in the early 2000s. Of the 753 R&D announcements, 570 announce increases in R&D spending for the forthcoming fiscal year compared to the previous year's spending levels, and the rest announce a decline in R&D spending. (6)

B. Computing the Wealth Effect of R&D Investment Announcements

To measure the shareholder wealth effect of R&D announcements, we use the standard event-study methodology to evaluate stock price responses to announcements of R&D spending changes. Day 0 is defined as the initial R&D announcement date. We use the CRSP value-weighted return as the market index and measure the abnormal returns using the market-adjusted returns model. Daily abnormal returns are cumulated to obtain the cumulative abnormal returns (CARs) from 30 days before to 30 days after the initial announcement date. We use a three-day period (-1, +1) to capture the price reaction to the R&D spending change announcement, an earlier period (-30, -2) to capture the anticipation of the information, and a later period (+2, +30) to capture early revisions to the initial reaction of the R&D change announcement. (7) As Barber and Lyon (1997), Fama (1998), and Lyon, Barber, and Tsai (1999) argue, the short-term event-study approach is less subject to critiques on a long-run event. The short-run approach assumes informational efficiency of the stock market; that is, the immediate stock price reaction to the R&D change announcement provides an unbiased estimate of the announcing firm's profitability from the shareholders' perspective.

Additionally, because trading volume is positively associated with information flow (Karpoff, 1987), we also examine the trading volume of stocks around the R&D announcement period. If announcing firms experience heavy trading around R&D announcement days, this should ensure that their stock prices fully reflect publicly available information. Hence, we analyze daily turnover (trading volume divided by shares outstanding) over three intervals: (-30, -2), (-1, +1), and (+2, +30). Notably, volume turnover is used rather than trading volume to prevent unusually high trading volumes in a few large stocks from disproportionately biasing the volume effect. Following Davidson III, Chhachhi, and Glascock (1996), the volume turnover data are converted into average daily volume turnover for each firm for each interval. The average daily volume turnover of announcing firms during each interval is calculated as a percentage of the average daily volume turnover for the entire period (-30, +30). We conduct pairwise comparisons among the three intervals for the mean average daily volume turnover using Tukey and Scheffe tests.

C. Measuring Corporate Governance

We focus on the universe of provisions that RiskMetrics monitors for institutional investors and researchers interested in corporate governance as a measure of the quality of firms' governance structure. Governance is measured using the G index of Gompers et al. (2003), which consists of 24 antitakeover and shareholder rights provisions. The G index is constructed by adding one point for each provision that enhances managerial power. Firms with higher G indices are viewed as having weaker shareholder rights, as it is more difficult and costly for shareholders to remove managers at these firms. Additionally, we use the BCF index of Bebchuk et al. (2009), which consists of 6 of the 24 provisions listed in Gompers et al. (2003). Bebchuk et al. (2009) show that their index has a stronger association with stock returns and firm value than does the G index. Because Bebchuk and Cohen (2005) focus on a staggered board as a key antitakeover provision, we refer to the staggered board indicator as an index as well. To test for robustness, we use the ATI index of Cremers and Nair (2005) as an alternative measure of governance quality. The ATI index focuses on only three key antitakeover provisions--the presence of staggered boards, of a preferred blank check ("poison pill"), and of restrictions on shareholder voting to call special meetings or act through written consent. We thus consider all four indices and separately examine their effects on R&D announcement returns.

D. Measuring Product Market Competition

Following prior research (e.g., Gillan, Hartzell, and Starks, 2003; Masulis et al., 2007; Giroud and Mueller, 2011), we use the HI to capture the competitive structure of an industry, calculated as the sum of squared market shares of all Compustat firms in each Fama- French (1997) industry:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where [s.sub.ijt] is the market share of firm i in industry j in year t. Market shares are computed from Compustat using firms' sales. When computing the HI, we use all available Compustat firms, including those with dual-class shares. We exclude firms for which sales are either missing or negative. The HI is commonly used in empirical research and is well grounded in theory (see Tirole, 1988). (8) Industries with lower HI possess more competitive product markets. We classify industries using the 48-industry classification scheme of Fama and French (1997).

E. Test Design

We form two portfolios based on the governance measures. We assign announcing firms with a below-median G/BCF/ATI index or firms without a staggered board to the good-governance portfolio and announcing firms with an above-median G/BCF/ATI index or firms with a staggered board to the weak-governance portfolio. To analyze the interaction between corporate governance and competition, we divide both the good-governance and weak-governance portfolios into terciles by ranking firms according to their HIs and then sorting them into low-, middle-, and high-HI terciles. This yields 2 x 3 = 6 portfolios. To test our conjectures, we first examine the abnormal returns to the good-governance and weak-governance portfolios based on the entire sample of corporate R&D announcements over the three-day period (-1, +1). Next, we examine the announcement abnormal returns to the good-governance and weak-governance portfolios within each low-, middle-, or high-HI tercile.

To assess the cross-sectional variation in the stock price reaction associated with R&D change announcements, we need to control for variables shown in prior research to affect the market valuation of R&D investment (e.g., Chauvin and Hirschey, 1993; Szewczyk et al., 1996; Zantout, 1997; Chan et al., 2001; Cannolly and Hirschey, 2005; Hall and Oriani, 2006; Ho et al., 2006; Pindado et al., 2010). As mentioned earlier, these control variables include firm size, R&D intensity, debt ratio, Tobin's q, and free cash flow, all of which are measured at the fiscal year-end before the R&D announcement. We define firm size as the natural logarithm of the market value of common equity. R&D intensity is defined as R&D expenses divided by sales. The debt ratio is measured by total liabilities divided by total assets. We define Tobin's q as the ratio of an announcing firm's market value of assets over its book value of assets, where the market value of assets is computed as the book value of assets minus the book value of common equity plus the market value of common equity. Here free cash flow is defined as operating income before depreciation minus interest expense, taxes, preferred dividends, and common dividends, scaled by the book value of total assets. Moreover, we also control for the year fixed effects by incorporating year dummies and winsorize all variables at the bottom and top 1% levels to reduce the effects of a few extreme values. (9)

II. Empirical Results

A. Summary Statistics

Table III displays the mean and median firm characteristics of 723 R&D expenditure change announcements from 1988 to 2014. All selected variables are reported for the year preceding the R&D announcement. On average, industry concentration is 0.08. Market value averages $74,170 million. The average R&D intensity and debt ratio are, respectively, 0.22 and 0.51. The means and medians of four governance indices are similar to what Gompers et al. (2003), Bebchuk et al. (2009), Bebchuk and Cohen (2005), and Cremers and Nair (2005) obtain. The Pearson's correlation matrix shows that all four governance indices are highly correlated with each other.

B. CARs around R&D Spending Change Announcements and the Impact of Corporate Governance

Table IV exhibits how the information on corporate R&D spending changes gets disseminated to the market. Panel A shows that the mean (median) CAR over the announcement days (-1, + 1) for the whole sample reaches 0.36% (0.30%), significantly different from zero at the 1% level. The mean CARs over the pre- and postannouncement periods are 0.30% and 0.36%, respectively, and neither is significantly different from zero at the 10% level. These results indicate that on average the market responds positively to the information on corporate plans to change R&D expenditures and the price impact is captured during the three-day announcement period, with no discernible preannouncement information leakage or postannouncement lagged effect.

To examine whether the announcement effect is related to the nature of R&D spending change, we also analyze the return patterns for R&D-increasing and R&D-decreasing subsamples separately. As can be seen from Panels B and C in Table IV, the mean and median abnormal returns around the announcement period (-1, +1) for the R&D-increasing subsample are positive and significant (0.67% and 0.39%, respectively), compared to being negative and significant (-0.77% and -1%, respectively) for the R&D-decreasing subsample. The results suggest that ostensibly, the market views the announcement of corporate R&D increases (decreases) as a positive (negative) signal for future prospects of the announcing firm without considering the effectiveness of corporate governance mechanisms. Therefore, when the sample is taken as a whole, a significant positive effect appears to mask the significant effects with opposite signs that are found in these two subsamples.

To assess the influence of corporate governance structure, we depict mean CARs around R&D expenditure change announcements in Panels A to D of Figure 1 for all firms and for good-and weak-governance firms based on the G index, BCF index, staggered board, and ATI index, respectively. It is evident that for the whole R&D change sample and both R&D-increasing and R&D-decreasing subsamples, good-governance firms typically experience higher mean CARs over three intervals, that is, (-30, -2), (-1, +1), and (+2, +30), as compared to weak-governance firms. The picture that emerges from Figure 1 is that governance structure seems to matter for the valuation effects of R&D spending changes.

Unreported tests for differences between good- and weak-governance firms indicate that R&D changes made by good-governance firms generate significantly higher announcement returns than those made by weak-governance firms. These results show that the positive (negative) announcement returns for the whole sample and R&D-increasing (-decreasing) subsample are mainly driven by good- (weak-) governance firms.

C. Regression Results of Corporate Governance on R&D Announcement Abnormal Returns

Although the negative relations between governance indices and R&D announcement returns that we observe in the previous section are consistent with our conjecture, they do not allow us to draw reliable inferences on the effectiveness of corporate governance mechanisms, as the univariate analyses do not take into account the correlations between governance indices and other determinants of R&D announcement returns. Therefore, before we can draw any conclusions from these results, we need to control for all of the important variables shown in prior research to affect R&D announcement returns.

Table V reports the cross-sectional regression results. In all regressions, we report robust t-statistics that adjust for clustering at the Fama-French (1997) 48-industry level. (10) The dependent variable is the three-day CARs around each R&D change announcement. The key explanatory variables are the four governance indices introduced earlier. Because they are highly correlated with each other, we separately examine their effects on R&D announcement returns. We find that all four governance indices have negative and significant effects on announcement CARs. It thus suggests that even after controlling for various firm-specific characteristics, shareholders of good-governance firms realize higher announcement returns than those of weak-governance firms. This result supports our conjecture that on average managers at firms with weaker shareholder rights make poorer R&D investments; namely, the market views R&D investments by firms with stronger shareholder rights as a relatively more favorable signal of positive net present value projects. The evidence indicates that the findings in Figure 1 are not due to the omission of firm characteristics included in earlier studies of R&D announcement returns. For our control variables, both the magnitude and statistical significance of the parameter estimates are fairly stable across the four model specifications.

[FIGURE 1 OMITTED]

D. The Impact of Product Market Competition on the Association between Corporate Governance and R&D Announcement Abnormal Returns

To explore the influence of product market competition, we next plot mean R&D announcement returns for good- and weak-governance firms in competitive, middle-competitive, and noncompetitive industries, as shown in Figure 2. The competitive subgroup includes firms ranking in the lowest HI tercile. The middle-competitive subgroup includes firms ranking in the medium HI tercile. The noncompetitive subgroup includes firms ranking in the highest HI tercile. Among these subgroups, good- (weak-) governance firms in noncompetitive industries apparently realize the most (least) favorable announcement returns for the whole sample and R&D-increasing sub-sample. Additionally, the favorable announcement effect is stronger for good-governance firms in noncompetitive industries than in competitive industries. For the R&D-decreasing subsample, weak- (good-) governance firms in noncompetitive industries realize the most (least) unfavorable announcement returns for the R&D-decreasing subsample, and the unfavorable announcement effect is stronger for weak-governance firms in noncompetitive industries than in competitive industries. (11) Obviously, Figure 2 presents a synoptic view of the mean R&D announcement returns in relation to the interaction between governance and market competition. In other words, the positive effects of good governance on R&D announcement returns are relatively stronger in noncompetitive industries than in competitive industries.

[FIGURE 2 OMITTED]

To further ascertain the role of product market competition in explaining the association between corporate governance and market responses around R&D change announcements, we interact all four governance indices with a dummy of market competition, noncompetitive, which is equal to one for firms ranking in the highest HI tercile and zero otherwise. Based on the findings of Giroud and Mueller (2010, 2011) that firms in noncompetitive industries benefit more from good governance than do firms in competitive industries, we expect that the wealth effect associated with announcements of corporate R&D spending changes is more favorable for good-governance firms in noncompetitive industries than in competitive industries.

As expected the regression analyses in Table VI show that the coefficients on the four non-interacted governance indices are insignificant, whereas the coefficients on the interaction term in all four models are negative and significant, indicating that the negative relations between the governance indices and R&D announcement abnormal returns are pronounced only in noncompetitive industries. The evidence is consistent with the results in Figure 2 and supports the argument by Giroud and Mueller (2010, 2011) that governance and competition are substitutes. In sum, these results demonstrate that good governance mechanisms in noncompetitive industries channel managerial effort toward the efficient deployment of R&D resources and affect the extent to which a firm's innovative resource configuration can reach its full potential for value creation.

E. Volume Turnover around R&D Spending Change Announcements

The information value associated with announcements of changes in R&D spending is further examined using the volume turnover tests depicted in Table VII. The first column in Panel A lists the average daily volume turnover for each interval, and the second column displays the adjusted average daily volume turnover for each interval calculated as the percentage of the average daily volume turnover during the entire period (-30, +30). The results from both computation procedures for volume turnover are nominally and statistically consistent. The overall results demonstrate that the volume turnover increases during the R&D announcement days (-1, +1) as compared to the volume turnover during the pre- and postannouncement periods. For the whole sample, as in Panel A.l, the adjusted average daily volume turnover during the announcement days (-1, +1) is 1.14 times the adjusted average daily volume turnover for the entire period and exceeds that during all the other intervals. This rise is consistent with the argument that released information on corporate plans to change R&D expenditures creates investor trading desire and is associated with increased trading activities. For both R&D-increasing and R&D-decreasing subsamples, as in Panels A.2 and A.3, we find a similar pattern of substantial increases in volume turnover around the R&D announcement days.

Panels B and C in Table VII list the regression results of the effects of corporate governance and product market competition on R&D announcement volume turnover. It is evident from Panel B that all four governance indices have negative and significant effects on announcement volume turnover, reflecting that R&D change announcements undertaken by firms with stronger shareholder rights attract more trading activities. Likewise, the estimation in Panel C shows that the interaction term is negative and highly significant (0.001 level) in all four models. This result strongly supports the argument that good-governance firms in noncompetitive industries are in a better position to reap substantially greater benefits from R&D investments. These findings correspond to the return patterns described in the previous section, thus providing additional evidence regarding shareholder wealth associated with corporate R&D change announcements.

F. Robustness Tests of the Endogeneity Bias

1. Changes in Competition and Changes in Corporate Governance

Endogeneity might be less of a problem in this context because the primary dependent variable in our tests is a short-term market-based measure (i.e., returns). Nevertheless, it is possible that changes in competition and changes in corporate governance that took place over our sample period are driving the empirical results. (12) Deregulation represents a shock that radically changes the nature of competition because the regulatory barriers are removed through deregulation (Winston, 1998, 2012; Ovtchinnikov, 2010). Therefore, deregulation is expected to increase market competition. To investigate the possibility that our results could reflect in part the effects of exogenously increased competition due to the deregulatory changes on shareholder wealth around R&D change announcements, we first focus on a subsample of firms with deregulatory changes in their industries.

Over our sample period, 16 of 723 R&D change announcements experience deregulatory changes, including one in utilities (affected by the Energy Policy Act in 1992), six in telecommunications (two are affected by the proposed rules on price caps in 1988 and four are affected by the Telecommunications Act in 1996), and nine in petroleum and natural gas industries (one is affected by the Natural Gas Wellhead Decontrol Act in 1989 and eight are affected by the Federal Energy Regulatory Commission Order 636 in 1992). (13) It is not feasible to perform our full regression analysis using this small subsample. We therefore limit our sample to announcing firms that experience no deregulatory changes. For these firms, the mean and median absolute changes in the HI from the preannouncement year to the postannouncement year are 0.006 and 0, and neither of these changes is significant. We reestimate the R&D announcement return regressions using this subsample. We find that our full-sample results continue to hold in this subsample; that is, all governance indices continue to have negative and significant effects on R&D announcement returns only in noncompetitive industries (unreported but available upon request).

Although the governance indices are relatively stable at the firm level as noted by Gompers et al. (2003) and Masulis et al. (2007), it remains possible that announcing firms experience changes in governance structures during their R&D announcement period. If sample firms experience changes in governance mechanisms as a direct result of the R&D changes, our test design, which uses preannouncement governance indices to explain R&D announcement abnormal returns, could be problematic. To address this issue, we look at the staggered board and examine the absolute change in this provision from the preannouncement year to the postannouncement year. The reason for looking at the staggered board provision is that most of the important antitakeover amendments, especially staggered boards, are adopted in the 1980s (Bebchuk and Cohen, 2005). Of the 723 R&D announcements in our sample, 702 have RiskMetrics coverage of the announcing firm in both the preannouncement year and the postannouncement year. For these firms, the mean (median) absolute change in the staggered board provision from the preannouncement year to the postannouncement year is an insignificant 0.025 (0). To separate the effects of market competition from the effects of corporate governance, we focus on announcing firms that do not experience any changes in the staggered board provision but experience changes in the HI from the preannouncement year to the postannouncement year. These firms are divided into competition-increase and competition-decrease subgroups to examine the differential effects of competition changes on R&D announcement returns.

Of the 702 R&D announcements for which we are able to track the change in the staggered board from the preannouncement year to the postannouncement year, 275 (410) have announcing firms that experience no change in the staggered board provision but experience increases (decreases) in the HI. Still, we continue to find that the negative relations between the staggered board provision and R&D announcement returns are significant only in the competition-decrease subgroup (unreported but available upon request). This evidence again indicates that managers at good-governance firms facing lower competitive pressures make better R&D investments for their shareholders. Therefore, our earlier results do not appear to be driven by the changes in competition and changes in corporate governance that took place over our sample period.

2. Omitted-Variable Problem

In addition to the concerns described above, there are other monitoring and governance mechanisms that might attenuate agency conflicts between managers and shareholders. Ignoring these factors potentially creates an omitted-variable problem. For example, some unobservable firm traits could be responsible for both the level of shareholder rights protection in a firm and the profitability of its R&D investment. To address this omitted-variable issue, we include a series of firm and governance controls in our model to test the robustness of the results. This addition can be important because the negative correlations between governance indices and shareholder value could be spurious if alternative firm traits and corporate control mechanisms are not considered.

For the firm factors, we include firm age and preannouncement stock price run-up. As firm age captures differences in competitiveness and agency conflicts associated with history, it may influence the firm's market valuation. Age is measured by the natural logarithm of the number of years the firm is listed with a nonmissing stock price on Compustat. An appealing feature of this variable is that it is much less endogenous than most other firm variables (Hadlock and Pierce, 2010). In addition, given the evidence in Gompers et al. (2003) and related studies that firms with weaker shareholder rights realize worse stock return performance, we control for an announcing firm's stock price run-up before the R&D announcement to isolate the effect of shareholder rights protection from that of prior stock performance. We measure the announcing firm's preannouncement stock price run-up through the buy-and-hold abnormal return over the 200-day window (event days -230 to -31) by using the CRSP value-weighted market index as the benchmark.

For other corporate governance mechanisms, we control for institutional ownership, board size, and board independence. Because institutional investors are better informed and more capable monitors (Shleifer and Vishny, 1997; Grinstein and Michaely, 2005), institutional investors play an important corporate governance role. As a consequence, the presence of institutional investors makes external governance more effective and increases the marginal costs of engaging in empire building or other types of moral hazard activities (Lin, Ma, and Xuan, 2011). We use two proxies for institutional holdings: the percentage stock ownership by a firm's largest institutional blockholder, defined as an institutional investor with at least 5% equity ownership (Block), and the aggregate percentage stock ownership in a firm by public pension funds (PP). We include Block and PP jointly in our regressions because of their low cross-correlation (Cremers and Nair, 2005). Aside from institutional ownership, board size and board independence are important attributes shown in prior work to affect how effectively a board functions. For example, Yermack (1996) finds an inverse relation between board size and firm value. Weisbach (1988), Schellenger, Wood, and Tashkori (1989), and Brickley, Coles, and Terry (1994) indicate that firms with a majority of independent directors make major corporate decisions in the best interests of shareholders. We define board size as the number of directors on a board and measure board independence as the proportion of independent directors on the board. We obtain institutional ownership and board data from the CDA Spectrum and RiskMetrics databases, respectively. By requiring these data, we lose some observations leaving us with 466 announcements. (14)

Table VIII presents the regression estimates. We find that after introducing other firm characteristics and governance mechanisms, the marginal effects of governance indices on R&D announcement returns remain negative and significant in noncompetitive industries. The coefficients on the control variables in Table VIII are generally consistent with those in previous tables. Therefore, our main results do not appear to be driven by, or sensitive to, endogeneity bias caused by unobservable omitted variables.

G. Other Sensitivity Tests

To determine whether the tests could be biased by the methodology employed, we consider the sensitivity of the results to variations in testing implementation. First, following Masulis et al. (2007), we take a dummy variable approach to the G/BCF/ATI index, where R&D announcing firms with an above-median G/BCF/ATI index are equal to one and zero otherwise. We reestimate the R&D announcement return regressions in Table VI after replacing the G index, BCF index, and ATI index with these three dummy variables. Again, we find that all dummy variables have negative and significant coefficients only in noncompetitive industries. This result further supports the earlier evidence obtained when these indices are treated as continuous variables.

Second, we check the robustness of our results to the concern that a few very large abnormal returns are driving the results. To examine this possibility, we estimate a logit regression in which the binary dependent variable indicates positive announcement abnormal returns. Again, all governance indices continue to have negative and significant effects on R&D announcement returns only in noncompetitive industries. Therefore, our earlier results do not appear to be driven by a few very large abnormal returns.

Third, as the HI computed from Compustat includes only publicly traded companies, we also use competition measures provided by the Census Bureau, which include all public and private companies in the United States and hence are likely to be more accurate. (15) Although these measures are more comprehensive, they are available only every five years and only for manufacturing industries. Because our sample period is from 1988 to 2014, we use data from the 1992, 1997, 2002, 2007, and 2012 Censuses. For intermediate years, we use data from the latest available Census. Because of these drawbacks, we lose some observations, leaving us with 516 announcements. Untabulated results show that the negative effects of the governance indices on R&D announcement returns are pronounced only in noncompetitive industries. Similarly, the negative relations between governance indices and R&D announcement volume turnover are relatively stronger in noncompetitive industries. These findings again indicate that good-governance firms facing lower product market competition pressures make better R&D investments for their shareholders.

Finally, we measure R&D announcement abnormal returns over event days (-5, +5), (-2, +2), and (-1,0) or use the market model parameters over the 200-day period from event day -230 to event day -31 with the CRSP value-weighted return as the market return. The results on the governance indices and product market competition remain qualitatively the same. Untabulated results again confirm the robustness of our findings. (16)

III. Summary and Conclusions

We examine whether governance mechanisms interact with market competition to affect shareholder wealth on announcements of corporate R&D spending changes. Our results show that investors appear to respond rationally to announcements of corporate R&D programs. The stock price response is positive, significant, and larger for good-governance firms. This evidence suggests that governance structure (i.e., the protection of shareholder rights) has a strong and material impact on managers' efforts to make value-creating R&D investments. Furthermore, good-governance firms in noncompetitive industries experience more favorable R&D announcement returns than do good-governance firms in competitive industries, indicating that investors appear to incorporate the interaction of corporate governance and product market competition in their pricing decisions. The substantial increases in turnover around the R&D announcement days for good-governance firms in noncompetitive industries also provide further evidence regarding investors' valuation of R&D spending changes.

Our article documents how corporate governance mechanisms, the degree of product market competition, and a firm's innovative resources jointly affect its market valuation. In this regard, our study makes an important contribution by shedding light on the impact of interacting governance with competition on a firm's investment efficiency and, in particular, the shareholder wealth effects of its R&D investments.

References

Aboody, D. and B. Lev, 2000, "Information Asymmetry, R&D, and Insider Gains," Journal of Finance 55, 2747-2766.

Aggarwal, R. and A. Samwick, 1999, "Executive Compensation, Strategic Competition, and Relative Performance Evaluation: Theory and Evidence," Journal of Finance 54, 1999-2043.

Barber, B.M. and J.D. Lyon, 1997, "Detecting Long-Run Abnormal Stock Returns: The Empirical Power and Specification of Test Statistics," Journal of Financial Economics 43, 341-372.

Bebchuk, L.A. and A. Cohen, 2005, "The Costs of Entrenched Boards," Journal of Financial Economics 78, 409-433.

Bebchuk, L.A., A. Cohen, and A. Ferrell, 2009, "What Matters in Corporate Governance?" Review of Financial Studies 22, 783-827.

Brickley, J.A., J.L. Coles, and R.L. Terry, 1994, "Outside Directors and the Adoption of Poison Pills," Journal of Financial Economics 35, 371-390.

Cannolly, R.A. and M. Hirschey, 2005, "Firm Size and the Effect of R&D on Tobin's Q," R&D Management 35, 217-223.

Chan, K., Y.-H. Lin, and Y. Wang, 2015, "The Information Content of R&D Reductions," Journal of Empirical Finance 34, 131-155.

Chan, L.K.C., J. Lakonishok, and T. Sougiannis, 2001, "The Stock Market Valuation of Research and Development Expenditures," Journal of Finance 56, 2431-2456.

Chan, S.H., J. Martin, and J. Kensinger, 1990, "Corporate Research and Development Expenditures and Share Value," Journal of Financial Economics 26, 255-276.

Chauvin, K. and M. Hirschey, 1993, "Advertising, R&D Expenditures, and the Market Value of the Firm," Financial Management 22, 128-140.

Chung, K.H., P. Wright, and B. Kedia, 2003, "Corporate Governance and Market Valuation of Capital and R&D Investments," Review of Financial Economics 12, 161-172.

Core, J.E., W.R. Guay, and TO. Rusticus, 2006, "Does Weak Governance Cause Weak Stock Returns? An Examination of Firm Operating Performance and Investors' Expectations," Journal of Finance 61, 655-687.

Cremers, M.K.J. and VB. Nair, 2005, "Governance Mechanisms and Equity Prices," Journal of Finance 60, 2859-2894.

Cremers, M.K.J., VB. Nair, and U. Peyer, 2008, "Takeover Defenses and Competition: The Role of Stakeholders," Journal of Empirical Legal Studies 5, 791-818.

Davidson III, W.N., I. Chhachhi, and J.L. Glascock, 1996, "A Test for Price Pressure Effects in Tender Offer Stock Repurchases," Financial Review 31, 25-49.

Doukas, J. and I. Switzer, 1992, "The Stock Market's Valuation of R&D Spending and Market Concentration," Journal of Economics and Business 44, 95-144.

Eberhart, A.C., W.F. Maxwell, and A.R. Siddique, 2004, "An Examination of Long-Term Abnormal Stock Returns and Operating Performance Following R&D Increases," Journal of Finance 59, 623-650.

Fama, E.F., 1998, "Market Efficiency, Long-Term Returns, and Behavioral Finance," Journal of Financial Economics 49, 283-306.

Fama, E.F. and K. French, 1997, "Industry Costs of Equity," Journal of Financial Economics 43, 153-193.

Gillan, S.L., J.C. Hartzell, and L.T. Starks, 2003, "Explaining Corporate Governance: Boards, Bylaws, and Charter Provisions," University of Delaware and University of Texas Working paper.

Giroud, X. and H. Mueller, 2010, "Does Corporate Governance Matter in Competitive Industries?" Journal of Financial Economics 95, 312-331.

Giroud, X. and H. Mueller, 2011, "Corporate Governance, Product Market Competition, and Equity Prices," Journal of Finance 66, 563-600.

Gompers, P., J. Ishii, and A. Metrick, 2003, "Corporate Governance and Equity Prices," Quarterly Journal of Economics 118, 107-155.

Griliches, Z., 1981, "Market Value, R&D and Patents," Economics Letters 7, 183-187.

Grinstein, Y. and R. Michaely, 2005, "Institutional Holdings and Payout Policy," Journal of Finance 60, 1389-1426.

Hadlock, C.J. and J.R. Pierce, 2010, "New Evidence on Measuring Financial Constraints: Moving Beyond the KZ Index," Review of Financial Studies 23, 1909-1940.

Hall, B.H., 1993, "The Stock Market's Valuation of R&D Investment during the 1980's," American Economic Review 83, 259-264.

Hall, B.H. and R. Oriani, 2006, "Does the Market Value R&D Investment by European Firms? Evidence from a Panel of Manufacturing Firms in France, Germany, and Italy," International Journal of Industrial Organization 24, 971-993.

He, J. and H.C. Wang, 2009, "Innovative Knowledge Assets and Economic Performance: The Asymmetric Roles of Incentives and Monitoring," Academy of Management Journal 52, 919-938.

Ho, Y.K., M. Tjahijapranata, and C.M. Yap, 2006, "Size, Leverage, Concentration, and R&D Investment in Generating Growth Opportunities," Journal of Business 79, 851-876.

Jaffe, A.B., 1986, "Technological Opportunity and Spillovers of R&D: Evidence from Firms' Patents, Profits, and Market Value," American Economic Review 76, 984-1001.

Jensen, M.C., 1993, "The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems," Journal of Finance 48, 831-880.

Karpoff, J.M., 1987, "The Relation between Price Changes and Trading Volume Turnover: A Survey," Journal of Financial Quantitative Analysis 22, 109-126.

Lin, C., Y. Ma, and Y. Xuan, 2011, "Ownership Structure and Financial Constraints: Evidence from a Structural Estimation," Journal of Financial Economics 102, 416-431.

Lyon, J.D., B.M. Barber, and C.-L. Tsai, 1999, "Improved Methods for Tests of Long-Run Abnormal Stock Returns," Journal of Finance 54, 165-201.

Masulis, R., C. Wang, and F. Xie, 2007, "Corporate Governance and Acquirer Returns," Journal of Finance 62, 1851-1889.

Oriani, R. and M. Sobrero, 2008, "Uncertainty and the Market Valuation of R&D within a Real Options Logic," Strategic Management Journal 29, 343-361.

Ovtchinnikov, A.V., 2010, "Capital Structure Decisions: Evidence from Deregulated Industries," Journal of Financial Economics 95, 249-274.

Petersen, M.A., 2009, "Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches," Review of Financial Studies 22, 435-480.

Pindado, J., V.D. Queiroz, and C.D.L. Torre, 2010, "How Do Firm Characteristics Influence the Relationship between R&D and Firm Value?" Financial Management 39, 757-782.

Schellenger, M.H., D. Wood, and A. Tashkori, 1989, "Board of Director Composition, Shareholder Wealth and Dividend Policy," Journal of Management 15, 457-467.

Shleifer, A. and R.W. Vishny, 1997, "A Survey of Corporate Governance," Journal of Finance 52, 737-783.

Sundaram, A.K., T.A. John, and K. John, 1996, "An Empirical Analysis of Strategic Competition and Firm Values: The Case of R&D Competition," Journal of Financial Economics 40, 459-486.

Szewczyk, S.H., G.P. Tsetsekos, and Z.Z. Zantout, 1996, "The Valuation of Corporate R&D Expenditures: Evidence from Investment Opportunities and Free Cash Flow," Financial Management 25, 105-110.

Tirole, J., 1988, The Theory of Industrial Organization, Cambridge, MA, MIT Press.

Viscusi, W.K., J.E. Harrington, Jr., and J.H. Vernon, 2005, Economics of Regulation and Antitrust, Cambridge, MA, MIT Press.

Wang, C. and F. Xie, 2009, "Corporate Governance Transfer and Synergistic Gains from Mergers and Acquisitions," Review of Financial Studies 22, 829-858.

Weisbach, M.S., 1988, "Outside Directors and CEO Turnover," Journal of Financial Economics 20, 431-460.

Winston, C., 1998, "US Industry Adjustment to Economic Deregulation," Journal of Economic Perspectives 12, 89-110.

Winston, C., 2012, "Government Policy for a Partially Deregulated Industry: Deregulate It Fully," American Economic Review: Papers & Proceedings 102, 391-395.

Woolridge, J.R. and C.C. Snow, 1990, "Stock Market Reaction to Strategic Investment Decisions," Strategic Management Journal 11, 353-363.

Xu, B., M.L. Magnan, and P.E. Andre, 2007, "The Stock Market Valuation of R&D Information in Biotech Firms," Contemporary Accounting Research 24, 1291-1318.

Yermack, D., 1996, "Higher Market Valuation of Companies with a Small Board of Directors," Journal of Financial Economics 40, 185-211.

Zantout, Z.Z., 1997, "A Test of the Debt-Monitoring Hypothesis: The Case of Corporate R&D Expenditures," Financial Review 32, 21-48.

Tsai-Ling Liao and Wen-Chun Lin (*)

We thank Raghavendra Rau (Editor), an anonymous referee, and seminar participants at the Fourth Asian Conference on the Social Sciences for their valuable comments and suggestions on this paper. Financial support from the National Science Council of Taiwan (NSC 100-2410-H-126-010) is gratefully acknowledged. An earlier version of this paper was titled "Corporate Governance, Product Market Competition, and the Wealth Effect of R&D Investments. "

(*) Tsai-Ling Liao is an Associate Professor in the Department of Business Administration at the College of Management at National Formosa University, Taiwan. Wen-Chun Lin is an Associate Professor in the Department of Finance at National Taipei University of Business, Taiwan.

(1) A related paper by Chung, Wright, and Kedia (2003) indicates that the market valuation (Tobin's q) of the firm's capital and R&D investments depends critically on analyst following and board composition, but not on institutional holdings. However, Chung et al. (2003) do not consider the impact of shareholder rights and the interaction between corporate governance and product market competition on the shareholder wealth effects of R&D investments, as we do here.

(2) Supporting this view, Jensen (1993) indicates that many firms' R&D investments are not profitable and detrimental.

(3) We thank the editor for pointing out this thesis in relation to R&D-decreasing announcements.

(4) Several studies show that firms with stronger shareholder rights or fewer antitakeover provisions are associated with higher market value (Gompers et al., 2003; Bebchuk and Cohen, 2005; Bebchuk et al., 2009) and better operating performance (Core et al., 2006).

(5) Examples include R&D, R&D spending, R&D expenditures, R&D innovation, R&D outlays, R&D investment, R&D to grow, R&D change, R&D increase, R&D reduction, R&D cut, R&D reduce, and R&D decrease.

(6) The size of R&D-decreasing subsample is small compared to the R&D-increasing subsample over the same period. Given the proprietary cost associated with the disclosure of corporate R&D expenditure reductions, the scarcity of such decrease announcements is to be expected.

(7) As a robustness test, we also measure R&D announcement abnormal returns over event days (-5, +5), (-2, +2), and (-1, 0) or use the market model parameters over the 200-day period from event day -230 to event day -31. See Section II.G for further exposition.

(8) We also conduct robustness tests by replacing the HI with the competition measures provided by the US Census Bureau in the regression analyses. See Section II.G for further exposition.

(9) We do not control for the industry fixed effect because we already consider the influence of competitive industries (i.e., product market competition) in the following tests.

(10)As Gompers et al. (2003) point out, there are few changes over time in the G index, and the inclusion of firm fixed effects would force identification of the G index from only these changes. Following Petersen (2009), we hence use panel regressions with year fixed effects and obtain standard errors by clustering at the Fama-French (1997) 48-industry level.

(11) Note that this is consistent with Giroud and Mueller's (2011) findings that weak-governance firms have lower equity returns, worse operating performance, and lower firm value, but only in noncompetitive industries.

(12) We thank the editor for bringing this issue to our attention.

(13) For a summary of major deregulatory initiatives affecting entertainment, petroleum and natural gas, utilities, telecommunications, and transportation industries, see Viscusi, Harrington, and Vernon (2005) and Ovtchinnikov (2010).

(14) Because of the reduction in observations, we replace the year fixed effect with a Post-SOX indicator to capture the effect of regulatory intervention in corporate governance after 2003. With the enactment of the Sarbanes-Oxley Act of 2002, both the membership and certain functions of corporate boards are explicitly regulated.

(15) See: https://www.census.gov/econ/concentration.html.

(16) The results described but not reported in this section are available from the authors upon request.
Table I. Sample Selection Process
This table presents how the final sample is constructed. Announcements
of corporate R&D spending changes over January 1988 to December 2014
are obtained from the Dow Jones Factiva database by searching for the
words and phrases and their synonyms commonly used to describe R&D
spending changes as keys for a database search routine. The final
sample consists of 723 R&D spending change announcements.

                                     Observations

Total number of R&D spending change
announcements made by firms listed   1,799
on the NYSE, AMEX, and NASDAQ
Less firms whose industry has no
firm with the same four-digit
Standard industrial                    (66)
classification (SIC) code
Sample R&D spending change
announcements available              1,733
Less external governance data of
the announcing firms unavailable
from the                              (941)
RiskMetrics database
Less stock return data of the
announcing firms unavailable from
the annual CRSP                        (45)
database
Less financial data of the
announcing firms unavailable from
the Compustat                          (24)
database
Final sample                           723

Table II. Sample Distribution of R&D Spending Change Announcements
This table summarizes the sample distribution of R&D spending change
announcements from 1988 to 2014. The sample distribution is reported in
Panel A by Fama-French (1997) 48-industry classification, in Panel B by
announcing year, and in Panel C by R&D spending change type.

                        Panel A. Industry Distribution
Industry                Observations  %      Industry

Agriculture               5            0.69  Aircraft
Food products             2            0.28  Shipbuilding, railroad
                                             Equipment

Candy & soda              2            0.28  Defense
Beer & liquor             6            0.83  Precious metals
Tobacco products          1            0.14  Petroleum and natural
                                             gas

Entertainment             1            0.14  Utilities
Consumer goods            6            0.83  Communication
Healthcare                2            0.28  Business services
Medical equipment        14            1.94  Computers
Pharmaceutical          272           37.62  Computer software
products
Chemicals                41            5.67  Electronic equipment
Rubber and plastic        2            0.28  Measuring and control
products                                     equipment
Construction materials    3            0.41  Business supplies
Construction              1            0.14  Shipping containers
Steel works, etc.         2            0.28  Retail
Machinery                 7            0.97  Restaurants, hotels,
                                             motels

Automobiles and          17            2.35  Trading
trucks
                        Panel B. Year Distribution
Year                    Observations  %      Year

1988                      2            0.28  2002
1989                      3            0.41  2003
1990                      8            1.11  2004
1991                     12            1.66  2005
1992                     13            1.80  2006
1993                     15            2.07  2007
1994                     15            2.07  2008
1995                     14            1.94  2009
1996                      9            1.24  2010
1997                     13            1.80  2011
1998                     43            5.95  2012
1999                     54            7.47  2013
2000                     65            8.99  2014
2001                     25            3.46  Total
                        Panel C. R&D Spending Change Type
                        Observations  %
R&D-increasing          570            78.84
R&D-decreasing          153            21.16


Industry                Observations  %
Agriculture              35             4.84
Food products             1             0.14


Candy & soda              9             1.24
Beer & liquor             1             0.14
Tobacco products          9             1.24


Entertainment             1             0.14
Consumer goods            6             0.83
Healthcare               13             1.80
Medical equipment        40             5.53
Pharmaceutical           83            11.48
products
Chemicals               115            15.91
Rubber and plastic       15             2.07
products
Construction materials    6             0.83
Construction              1             0.14
Steel works, etc.         2             0.28
Machinery                 1             0.14

Automobiles and           1             0.14
trucks

Year                    Observations  %

1988                     49             6.78
1989                     22             3.04
1990                     68             9.41
1991                     39             5.39
1992                     58             8.02
1993                     23             3.18
1994                     39             5.39
1995                     12             1.66
1996                     32             4.43
1997                     17             2.35
1998                     21             2.90
1999                     14             1.94
2000                     38             5.26
2001                    723           100.00

R&D-increasing
R&D-decreasing

Table III. Firm Characteristic Statistics and Pearson's Correlations
Coefficients
This table shows the sample statistics and Pearson's correlation
coefficients. Herfindahl index is calculated as the sum of squares of
the financial market shares of all Compustat firms in the Fama-French
(1997) 48-industry classification. G index covers 24 unique
antitakeover provisions followed by RiskMetrics, from which Gompers,
Ishii, and Metrick (2003) construct their governance index. BCF index
covers six unique antitakeover provisions, from which Bebchuk, Cohen,
and Ferrell (2009) construct their governance index. Staggered board is
a dummy equal to one for firms with a staggered board. ATI index covers
three unique antitakeover provisions, from which Cremers and Nair
(2005) construct their governance index. Market capitalization
(Smillion) is measured as the capitalization of the announcing firm in
fiscal year -1. R&D intensity is the ratio of firm R&D expenditures to
sales in fiscal year -1. Debt ratio is the ratio of total debts to
total assets in fiscal year -1. Free cash flow is defined as operating
income before depreciation minus interest expense, taxes, preferred
dividends, and common dividends, divided by book value of total assets,
in fiscal year -1. Tobin's q is estimated as the ratio of the market
value of the firm's assets to the book value of the firm's assets in
fiscal year -1, where the market value of assets is estimated as the
book value of assets minus the book value of common equity plus the
market value of common equity.

Variable                  Observations  Mean       SD

 1. Herfindahl index       723                0.08       0.08
 2. G index                723                9.18       2.47
 3. BCF index              723                1.63       1.26
 4. Staggered board        723                0.41       0.49
 5. ATI index              723                1.58       0.96
 6. Market capitalization  723           74,170     94,965
 7. R&D intensity          723                0.22       0.81
 8. Debt ratio             723                0.51       0.20
 9. Free cash flow         723                0.10       0.09
10. Tobin's q              723                3.45       2.96

Variable                  Median      1            2

 1. Herfindahl index            0.05   1.00
 2. G index                     9.00  -0.10 (***)   1.00
 3. BCF index                   2.00  -0.11 (***)   0.39 (***)
 4. Staggered board             0.00  -0.13 (***)   0.56 (***)
 5. ATI index                   1.00  -0.12 (***)   0.65 (***)
 6. Market capitalization  38,050     -0.06         0.03
 7. R&D intensity               0.12  -0.06        -0.01
 8. Debt ratio                  0.51   0.21 (***)  -0.13 (***)
 9. Free cash flow              0.10  -0.07 (*)     0.05
10. Tobin's q                   2.49  -0.12 (***)   0.07 (*)

Variable                  3            4            5

 1. Herfindahl index
 2. G index
 3. BCF index               1.00
 4. Staggered board         0.55 (***)   1.00
 5. ATI index               0.61 (***)   0.81 (***)   1.00
 6. Market capitalization  -0.25 (***)  -0.26 (***)  -0.21 (***)
 7. R&D intensity           0.06         0.13 (***)   0.11 (***)
 8. Debt ratio              0.08 (**)    0.09 (**)   -0.01
 9. Free cash flow         -0.17 (***)  -0.24 (***)  -0.12 (***)
10. Tobin's q              -0.10 (***)   0.06         0.08 (**)

Variable                  6            7            8

 1. Herfindahl index
 2. G index
 3. BCF index
 4. Staggered board
 5. ATI index
 6. Market capitalization   1.00
 7. R&D intensity          -0.08 (**)    1.00
 8. Debt ratio             -0.23 (***)   0.08 (**)    1.00
 9. Free cash flow          0.30 (***)  -0.46 (***)  -0.38 (***)
10. Tobin's q               0.46 (***)  -0.01        -0.35 (***)

Variable                  9           10

 1. Herfindahl index
 2. G index
 3. BCF index
 4. Staggered board
 5. ATI index
 6. Market capitalization
 7. R&D intensity
 8. Debt ratio
 9. Free cash flow         1.00
10. Tobin's q             0.34 (***)  1.00

(***) Significant at the 0.01 level.
(*) Significant at the 0.05 level.
(*) Significant at the 0.10 level.

Table IV. Cumulative Abnormal Returns (CARs) around R&D Spending Change
Announcements
This table presents CARs surrounding the announcements of 723 R&D
spending changes from 1988 to 2014, including 570 R&D-increasing and
153 R&D-decreasing announcements. Abnormal returns are estimated using
the market-adjusted returns model with the CRSP value-weighted return
as the market index. Day 0 is defined as the initial R&D announcement
date.

Period Relative                                       Mean
to Announcement  Observations                         CARs (%)

                 Panel A. R&D Spending Change Sample
(-30, -2)        723                                   0.30
(-1, +1)         723                                   0.36
(+2, +30)        723                                   0.36
                 Panel B. R&D-Increasing Subsample
(-30, -2)        570                                   0.39
(-1, +1)         570                                   0.67
(+2, +30)        570                                   0.52
                 Panel C. R&D-Decreasing Subsample
(-30, -2)        153                                  -0.06
(-1, +1)         153                                  -0.77
(+2, +30)        153                                  -0.25

                                        p-Value for
Period Relative               Median    Wilcoxon
to Announcement  (-Statistic  CARs (%)  z-Statistic

(-30, -2)         0.63         0.47     0.18
(-1, +1)          3.95 (***)   0.10     0.00 (***)
(+2, +30)         1.16        -0.07     0.62

(-30, -2)         0.97         0.56     0.51
(-1, +1)          6.37 (***)   0.39     0.00 (***)
(+2, +30)         1.18         0.10     0.54

(-30, -2)        -0.03        -0.03     0.46
(-1, +1)         -4.63 (***)  -1.00     0.00 (***)
(+2, +30)        -0.23        -0.14     0.62

(***) Significant at the 0.01 level.

Table V. Cross-Sectional Regression Results of Corporate Governance on
R&D Announcement Abnormal Returns (-1, +1)
The dependent variable is three-day (-1, +1) cumulative abnormal
returns (%). Four governance indices are used: G index, BCF index,
staggered board, and ATI index. G index covers 24 unique antitakeover
provisions followed by RiskMetrics, from which Gompers, Ishii, and
Metrick (2003) construct their governance index. BCF index covers six
unique antitakeover provisions, from which Bebchuk, Cohen, and Ferrell
(2009) construct their governance index. Staggered board is a dummy
equal to one for firms with a staggered board. ATI index covers three
unique antitakeover provisions, from which Cremers and Nair (2005)
construct their governance index. Firm size is the nature logarithm of
market capitalization of the announcing firm in fiscal year -1. R&D
intensity is the ratio of firm R&D expenditures to sales in fiscal year
-1. Debt ratio is the ratio of total debts to total assets in fiscal
year -1. Free cash flow is defined as operating income before
depreciation minus interest expense, taxes, preferred dividends, and
common dividends, divided by book value of total assets, in fiscal year
-1. Tobin's q is estimated as the ratio of the market value of the
firm's assets to the book value of the firm's assets in fiscal year -1,
where the market value of assets is estimated as the book value of
assets minus the book value of common equity plus the market value of
common equity. The t-values in parentheses are computed with clustered
standard errors at the Fama-French (1997) 48-industry level.

Governance Measure  G Index        BCF Index     Staggered Board
                    Model 1        Model 2       Model 3

Intercept             1.42           0.66          0.85
                     (1.92) (*)     (1.09)        (1.35)
Governance           -0.12          -0.13         -0.63
                    (-2.89) (***)  (-1.90) (*)   (-3.02) (***)
Firm size            -0.05          -0.07         -0.09
                    (-0.73)        (-1.08)       (-1.40)
R&D intensity         0.07           0.08          0.08
                     (2.43) (**)    (2.55) (**)   (2.59) (***)
Debt ratio            0.60           0.81          0.90
                     (1.35)         (1.85) (*)    (2.02) (**)
Free cash flow        2.01           1.86          1.37
                     (1.80) (*)     (1.66) (*)    (1.19)
Tobin's q             0.01           0.01          0.03
                     (0.14)         (0.08)        (0.67)
Year fixed effects  Yes            Yes           Yes
Adjusted [R.sup.2]    0.02           0.01          0.02
Observations        723            723           723

Governance Measure  ATI Index
                    Model 4

Intercept             1.05
                     (1.56)
Governance           -0.30
                    (-2.63) (***)
Firm size            -0.08
                    (-1.34)
R&D intensity         0.07
                     (2.28) (**)
Debt ratio            0.80
                     (1.83) (*)
Free cash flow        1.80
                     (1.56)
Tobin's q             0.02
                     (0.55)
Year fixed effects  Yes
Adjusted [R.sup.2]    0.02
Observations        723

(***) Significant at the 0.01 level.
(**) Significant at the 0.05 level.
(*) Significant at the 0.10 level.

Table VI. Controlling for Product Market Competition: Herfindahl Index
The dependent variable is three-day (-1, +1) cumulative abnormal
returns (%). Four governance indices are used: G index, BCF index,
staggered board, and ATI index. Noncompetitive is a dummy equal to one
for firms ranking in the highest Herfindahl index tercile and zero
otherwise. Firm size is the nature logarithm of market capitalization
of the announcing firm in fiscal year -1. R&D intensity is the ratio of
firm R&D expenditures to sales in fiscal year -1. Debt ratio is the
ratio of total debts to total assets in fiscal year -1. Free cash flow
is defined as operating income before depreciation minus interest
expense, taxes, preferred dividends, and common dividends, divided by
book value of total assets, in fiscal year -1. Tobin's q is estimated
as the ratio of the market value of the firm's assets to the book value
of the firm's assets in fiscal year -1, where the market value of
assets is estimated as the book value of assets minus the book value of
common equity plus the market value of common equity. The t-values in
parentheses are computed with clustered standard errors at the
Fama-French (1997) 48-industry level.

Governance Measure             G Index        BCF Index
                               Model 1        Model 2

Intercept                        0.85           0.37
                                (1.10)         (0.61)
Governance                      -0.03          -0.03
                               (-0.60)        (-0.34)
Noncompetitive                   2.38           0.76
                                (2.93) (***)   (2.65) (***)
Noncompetitive (*) Governance   -0.24          -0.26
                               (-2.59) (***)  (-2.14) (**)
Firm size                       -0.07          -0.06
                               (-1.20)        (-0.94)
R&D intensity                    0.05           0.06
                                (2.05) (**)    (2.11) (**)
Debt ratio                       0.39           0.66
                                (0.85)         (1.43)
Free cash flow                   2.04           1.89
                                (1.78) (*)     (1.66) (*)
Tobin's q                        0.01           0.01
                                (0.29)         (0.19)
Year fixed effects             Yes            Yes
Adjusted [R.sup.2]               0.03           0.02
Observations                   723            723

Governance Measure             Staggered Board  ATI Index
                               Model 3          Model 4

Intercept                        0.70             0.84
                                (1.12)           (1.26)
Governance                      -0.26            -0.11
                               (-0.91)          (-0.73)
Noncompetitive                   0.61             1.21
                                (2.51) (**)      (3.30) (***)
Noncompetitive (*) Governance   -1.08            -0.65
                               (-2.53) (**)     (-2.69) (***)
Firm size                       -0.09            -0.09
                               (-1.43)          (-1.41)
R&D intensity                    0.08             0.06
                                (2.49) (**)      (2.17) (**)
Debt ratio                       0.73             0.52
                                (1.55)           (1.13)
Free cash flow                   1.57             1.84
                                (1.32)           (1.56)
Tobin's q                        0.01             0.01
                                (0.27)           (0.14)
Year fixed effects             Yes              Yes
Adjusted [R.sup.2]               0.03             0.03
Observations                   723              723

(***) Significant at the 0.01 level.
(**) Significant at the 0.05 level.
(*) Significant at the 0.10 level.

Table VII. Average Daily Volume Turnover around R&D Spending Change
Announcements
In Panel A, volume turnover (trading volume divided by shares
outstanding) over the three event intervals is converted into average
daily volume turnover in each interval. The adjusted average daily
volume turnover for each interval is computed as a times of the average
daily volume turnover for the entire period (-30, +30). Analysis of
variance (ANOVA) is used to test for the null hypothesis that average
daily volume turnover is equivalent among the three intervals. The
superscript letters indicate when pairs of means are significantly
different at the 0.05 level using the Tukey test. The Scheffe test
confirms these results. When the letters are the same for a pair in the
column, it means that we cannot reject the null hypothesis of equal
means; when the letters are different, we can reject the null. In
Panels B and C, the dependent variable is the average daily volume
turnover over the R&D announcement days (-1, +1). Four governance
indices are used: G index, BCF index, staggered board, and ATI index.
Noncompetitive is a dummy equal to one for firms ranking in the highest
Herfindahl index tercile and zero otherwise. Firm size is the nature
logarithm of market capitalization of the announcing firm in fiscal
year -1. R&D intensity is the ratio of firm R&D expenditures to sales
in fiscal year -1, Debt ratio is the ratio of total debts to total
assets in fiscal year -1. Free cash flow is defined as operating income
before depreciation minus interest expense, taxes, preferred dividends,
and common dividends, divided by book value of total assets, in fiscal
year -1. Tobin's q is estimated as the ratio of the market value of the
firm's assets to the book value of the firm's assets in fiscal year -1,
where the market value of assets is estimated as the book value of
assets minus the book value of common equity plus the market value of
common equity. The t-values in parentheses are computed with clustered
standard errors at the Fama-French (1997) 48-industry level.

                         Average Daily        Adjusted Average Daily
Intervals  Observations  Volume Turnover (%)  Volume Turnover (Times)
           Panel A. Average Daily Volume Turnover
Panel A.1. R&D Spending Change Sample
(-30, -2)  723           0.77 (a)              0.99 (a)
(-1.+1)    723           0.94 (b)              1.14 (b)
(+2, +30)  723           0.78 (a)              1.00 (a)
ANOVA                    6.59 (***)           44.05 (***)
F-value
Panel A.2. R&D-Increasing Subsample
(-30, -2)  570           0.77 (a)              0.98 (a)
(-1.+1)    570           0.94 (b)              1.16 (b)
(+2, +30)  570           0.79 (a)              1.00 (a)
ANOVA                    5.86 (***)           38.33 (***)
F-value
Panel A.3. R&D-Decreasing Subsample
(-30, -2)  153           0.76 (a)              1.00 (a)
(-1.+1)    153           0.95 (b)              1.09 (b)
(+2, +30)  153           0.75 (a)              0.99 (a)
ANOVA                    6.02 (***)            5.83 (***)
F-value

Governance Measure             G Index        BCF Index
Panel B. Regression Results of Corporate Governance on Volume Turnover
                               Model 1        Model 2

Intercept                        3.93           3.19
                               (10.77) (***)   (8.95)
Governance                      -0.13          -0.16
                               (-6.15) (***)  (-3.62)
Firm size                       -0.18          -0.20
                               (-5.94) (***)  (-6.06)
R&D intensity                    0.01           0.02
                                (0.35)         (0.70)
Debt ratio                       0.10           0.13
                                (0.35)         (0.44)
Free cash flow                   0.20           0.01
                                (0.26)         (0.01)
Tobin's q                        0.01           0.01
                                (0.26)         (0.43)
Year fixed effects             Yes            Yes
Adjusted [R.sup.2]               0.14           0.10
Observations                   723            723
Panel C. Controlling for Product Market Competition: Herfindahl Index
Intercept                        3.37           2.84
                                (8.65) (***)   (7.38)
Governance                      -0.04          -0.04
                               (-1.98) (**)   (-0.78)
Noncompetitive                   2.28           0.88
                                (4.89) (***)   (4.69)
Noncompetitive (*) Governance   -0.23          -0.34
                               (-4.81) (***)  (-4.08)
Firm size                       -0.20          -0.19
                               (-6.75) (***)  (-5.65)
R&D intensity                    0.01           0.01
                                (0.21)         (0.11)
Debt ratio                       0.30           0.02
                                (1.13)         (0.08)
Free cash flow                   0.22           0.07
                                (0.29)         (0.09)
Tobin's q                        0.01           0.02
                                (0.07)         (1.39)
Year fixed effects             Yes            Yes
Adjusted [R.sup.2]               0.19           0.14
Observations                   723            723

Governance Measure               Staggered Board  ATI Index
Panel B. Regression Results of Corporate Governance on Volume Turnover
                                  Model 3          Model 4

Intercept                          3.28             3.47
                                  (9.87) (***)    (10.31) (***)
Governance                        -0.60            -0.29
                                 (-5.43) (***)    (-4.97) (***)
Firm size                         -0.22            -0.22
                                 (-7.12) (***)    (-6.95) (***)
R&D intensity                      0.02             0.01
                                  (0.68)           (0.21)
Debt ratio                         0.21             0.11
                                  (0.69)           (0.42)
Free cash flow                     0.42             0.01
                                  (0.51)           (0.01)
Tobin's q                          0.02             0.01
                                  (1.34)           (0.88)
Year fixed effects               Yes              Yes
Adjusted [R.sup.2]                 0.12             0.12
Observations                     723              723
Panel C. Controlling for Product Market Competition: Herfindahl Index
Intercept                          3.12             3.23
                                  (8.95) (***)     (9.06) (***)
Governance                        -0.23            -0.06
                                 (-2.16) (**)     (-1.40)
Noncompetitive                     0.62             1.43
                                  (4.33) (***)     (6.30) (***)
Noncompetitive (*) Governance     -1.06            -0.81
                                 (-5.19) (***)    (-6.05) (***)
Firm size                         -0.22            -0.22
                                 (-6.86) (***)    (-6.86) (***)
R&D intensity                      0.02             0.01
                                  (0.50)           (0.10)
Debt ratio                         0.03             0.21
                                  (0.11)           (0.82)
Free cash flow                     0.22             0.05
                                  (0.28)           (0.06)
Tobin's q                          0.01             0.01
                                  (0.02)           (0.60)
Year fixed effects               Yes              Yes
Adjusted [R.sup.2]                 0.17             0.20
Observations                     723              723

(***) Significant at the 0.01 level.
(*) Significant at the 0.05 level.

Table VIII. Controlling for Other Firm Traits and Governance
Mechanisms
In this table the dependent variable is three-day (-1, +1) cumulative
abnormal returns (%). Four governance indices are used: G index, BCF
index, staggered board, and ATI index. Noncompetitive is a dummy equal
to one for firms ranking in the highest Herfindahl index tercile and
zero otherwise. Firm size is the nature logarithm of market
capitalization of the announcing firm in fiscal year -1. R&D intensity
is the ratio of firm R&D expenditures to sales in fiscal year -1. Debt
ratio is the ratio of total debts to total assets in fiscal year -1.
Free cash flow is defined as operating income before depreciation minus
interest expense, taxes, preferred dividends, and common dividends,
divided by book value of total assets, in fiscal year -1. Tobin's q is
estimated as the ratio of the market value of the firm's assets to the
book value of the firm's assets in fiscal year -1, where the market
value of assets is estimated as the book value of assets minus the book
value of common equity plus the market value of common equity. Firm age
is measured by the natural logarithm of the number of years the firm is
listed with a nonmissing stock price on Compustat. Stock price run-up
is measured as the buy-and-hold abnormal return over the 200-day window
(event days -230 to -31) before the R&D announcement by using the CRSP
value-weighted market index as the benchmark. Block is the percentage
stock ownership by a firm's largest institutional blockholder. PP is
the aggregate percentage stock ownership in a firm by public pension
funds. Board size is defined as the number of directors on a board.
Board independence is measured as the proportion of independent
directors on the board. Post-SOX is a dummy equal to one if sample
firms announce their R&D expenditure changes in the post-Sarbanes-Oxley
period. The t-values in parentheses are computed with clustered
standard errors at the Fama-French (1997) 48-industry level.

Governance Measure             G Index        BCF Index
                               Model 1        Model 2

Intercept                        0.38           -0.33
                                (0.30)         (-0.40)
Governance                      -0.04           -0.03
                               (-0.59)         (-0.27)
Noncompetitive                   2.32            0.73
                                (3.00) (***)    (2.12) (**)
Noncompetitive (*) Governance   -0.24           -0.24
                               (-2.82) (***)   (-1.76) (*)
Firm size                       -0.07           -0.05
                               (-0.58)         (-0.41)
R&D intensity                    0.04            0.06
                                (1.36)          (1.60)
Debt ratio                       0.56            0.90
                                (1.37)          (2.27) (**)
Free cash flow                   2.88            2.74
                                (4.23) (***)    (3.44) (***)
Tobin's q                        0.03            0.04
                                (0.76)          (0.89)
Firm age                        -0.15           -0.17
                               (-4.11) (***)   (-4.85) (***)
Stock price run-up               0.08            0.15
                                (0.64)          (1.48)
Block                            1.07            0.53
                                (0.64)          (0.39)
PP                               0.27            0.21
                                (0.35)          (0.30)
Board size                      -0.03           -0.03
                               (-0.94)         (-0.64)
Board independence               0.92            0.59
                                (1.30)          (0.72)
Post-SOX                        -0.18           -0.04
                               (-0.93)         (-0.19)
Adjusted [R.sup.2]               0.06            0.04
Observations                   466             466

Governance Measure             Staggered Board  ATI Index
                               Model 3          Model 4

Intercept                        0.13             0.11
                                (0.14)           (0.12)
Governance                      -0.30            -0.13
                               (-1.62)          (-1.28)
Noncompetitive                   0.56             1.16
                                (2.82) (***)     (2.90) (***)
Noncompetitive (*) Governance   -0.93            -0.62
                               (-2.94) (***)    (-2.78) (***)
Firm size                       -0.01            -0.01
                               (-0.07)          (-0.10)
R&D intensity                    0.08             0.06
                                (1.70)*          (1.61)
Debt ratio                       0.86             0.72
                                (2.01) (**)      (1.55)
Free cash flow                   2.43             2.65
                                (2.50) (**)      (3.43) (***)
Tobin's q                        0.02             0.02
                                (0.41)           (0.65)
Firm age                        -0.16            -0.16
                               (-5.41) (***)    (-4.59) (***)
Stock price run-up               0.08             0.12
                                (0.67)           (1.02)
Block                            0.71             0.49
                                (0.52)           (0.33)
PP                               0.40             0.05
                                (0.53)           (0.07)
Board size                      -0.02            -0.02
                               (-0.37)          (-0.58)
Board independence               0.46             0.48
                                (0.53)           (0.57)
Post-SOX                        -0.12            -0.10
                               (-0.57)          (-0.47)
Adjusted [R.sup.2]               0.05             0.05
Observations                   466              466

(***) Significant at the 0.01 level.
(**) Significant at the 0.05 level.
(*) Significant at the 0.10 level.
COPYRIGHT 2017 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Liao, Tsai-Ling; Lin, Wen-Chun
Publication:Financial Management
Article Type:Report
Date:Sep 22, 2017
Words:12333
Previous Article:CEO age, risk incentives, and hedging strategy.
Next Article:Market illiquidity and conditional equity premium.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters