Ronald Masulis and Shawn Mobbs University of New South Wales & The University of Alabama

Size: px

Start display at page:

Download "Ronald Masulis and Shawn Mobbs University of New South Wales & The University of Alabama"

Marianna Howard
6 years ago
Views:

1 Ronald Masulis and Shawn Mobbs University of New South Wales & The University of Alabama Modified from the original presentation at The American Finance Association Annual Meeting January 4, 2009 San Francisco, CA to reflect the final version of the paper in The Journal of Finance, Vol. 66, No. 3, pp (2011)

2 Governance reforms emphasize outside directors Sarbanes-Oxley, Exchange Listings, Institutional Investors Pressured firms to decrease insider representation Little research on the role of inside directors Extensive board structure research- focuses on outside directors Two opposing theories of the role of inside directors Evidence of greater CEO influence Valuable contributors of firm-specific information Fama and Jensen (1983), Harris and Raviv (2008), Raheja (2005) Coles et al. (2008), Klein (1998) Can we distinguish between the two?

Outside directors differ in degrees of independence Mace (1971), Hallock (1997), Core et al (1999), Shivdasani and Yermack (1999), Hermalin and Weisbach (2003) External labor market for directorships

3 Outside directors differ in degrees of independence Mace (1971), Hallock (1997), Core et al (1999), Shivdasani and Yermack (1999), Hermalin and Weisbach (2003) External labor market for directorships Valuable decision management and control skills: Fama and Jensen (1983), Brickley, Linck and Coles (1999), Kaplan and Reishsus (1990) Inside directors differ in degrees of independence Independent Inside Directors hold outside directorships Indicates valuable decision management and control skills Greater career opportunities apart from their current CEO Are certified inside directors (CIDs) valuable board members?

4 Agency Perspective: Inside directors aid CEO entrenchment and extraction of private benefits of control H1 Certified Inside Directors (CIDs) are more common in firms with less powerful or entrenched CEOs. H2 Boards with CIDs are more effective, resulting in better firm operating performance and stock valuation.

5 Optimal Board Perspective: Advisory roles of inside directors enhance board decision making and monitoring, which is especially important when major decisions must be made. H3 In high R&D intensity firms, CIDs are more frequent and have a stronger positive association with firm performance and value relative to non-cids. H4 CIDs enhance board effectiveness, which leads to (i) more profitable acquisitions, (ii) larger average cash holdings, and (iii) smaller and less frequent earnings overstatements.

IRRC Data ~ S&P 1500 firms; 1997-2003 (Panel Data) Exclude

Hermalin and Weisbach (1988) Final Sample: Director

6 IRRC Data ~ S&P 1500 firms; (Panel Data) Exclude Finance and Utility Firms Firms where CEO 64 years old; Hermalin and Weisbach (1988) Final Sample: Director observations 8,742 non-ceo inside directors 10% are IIDs Firms: 2,137

7 Firm Characteristics Boone et al. (2007), Linck et al. (2008), Coles et al. (2008), Denis and Sarin (1999) Size, R&D, Capital Expenditures, Leverage, Business & Geographic Segments Past Performance, M&A CEO Influence Tenure, Age, Ownership, Founder or Relative present SOX Influence Methodology: Multivariate OLS and Probit regressions Year and Industry fixed effects Robust standard errors clustered by firm

8 Information Importance OLS Probit % Non-Certified % Certified Presence of Presence of Inside Directors Inside Directors Inside Directors Certified Inside Directors Model 1 Model 2 Model 3 Model 4 R&D/Assets ** 1.58* (0.018) (0.056) (0.708) (0.239) Capital Expenditure/Sales *** 0.01*** *** (0) (0.005) (0.284) (0.001) Ln(Sales) *** 0.37*** 0.07*** 0.28*** (0) (0) (0.001) (0) Leverage ** ** (0.014) (0.101) (0.012) (0.528) Ln(# Business Segments) * 0.01 (0.282) (0.899) (0.073) (0.792) Ln(# Geographic Segments) * -0.07* 0.12** (0.126) (0.069) (0.084) (0.036) Industry Competition * ** * (0.079) (0.042) (0.614) (0.058) CEO/Board Characteristics Ln(CEO Tenure) *** 0.25*** 0.15*** 0.17*** (0) (0) (0) (0) CEO Percent Ownership *** *** (0) (0.498) (0.008) (0.55) Board Ownership% *** *** (0) (0.601) (0) (0.183) Founder Director Present *** *** 0.06 (0) (0.508) (0) (0.529) Founder Family Director Present ** 0.03 (0.449) (0.826) (0.033) (0.797) Firm Performance & Activity Stock Volatility (0.415) (0.318) (0.795) (0.154) Operating CF (t-1) 0.02*** ** (0) (0.66) (0.015) (0.551) Recent M&A ** * 0.07 (0.048) (0.355) (0.088) (0.188) Post-SOX *** -0.75*** -0.39*** -0.56*** (0) (0) (0) (0) Number of Observations Adjusted R 2 / Psuedo-R % 5.94% 11.08% 14.18%

9 There are differences among inside directors CIDs are more likely where theory predicts insiders bring the most value to the board

Decision to have non-ceo inside directors is not random o Heckman (1979) Self-Selection Model 1 st stage Probit model Compute Inverse Mills Ratio Identification and IV - SOX indicator 2 nd stage

10 Decision to have non-ceo inside directors is not random o Heckman (1979) Self-Selection Model 1 st stage Probit model Compute Inverse Mills Ratio Identification and IV - SOX indicator 2 nd stage Regression of performance measure Firms with inside directors Inverse Mills Ratio control for private information Industry and year fixed effects, robust standard errors Firm Performance Measures o Industry Adjusted Operating Performance (CF) o Industry Adjusted Ln(M/B) Controls follow Coles et al. (2006), Anderson and Reeb (2003), Fich and Shivdasani (2006) o Firm Size, Business Segments, Firm Age o CEO & Board Ownership, Presence of Founder or Family member o Growth Opportunities, Return Volatility

11 Table 3. Firm Performance Regressions CF ln(m/b) Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Certified Inside Directors (%) ** * *** *** *** (0.017) (0.06) (0.581) (0.001) (0) (0) Non-Certified Inside Directors (%) *** (0.394) (0.783) (0.283) (0) CEO Percent Ownership * * *** *** (0.056) (0.086) (0.385) (0.007) (0.004) (0.905) CEO Percent Ownership *** *** (0.312) (0.379) (0.305) (0.002) (0.001) (0.451) Board Ownership *** *** *** *** (0) (0) (0) (0.666) (0.62) (0.007) Founder Director * * *** (0.081) (0.091) (0.47) (0.411) (0.443) (0) Founding Family Director ** ** ** ** 0.036* (0.027) (0.025) (0.282) (0.01) (0.012) (0.088) CF 1.741*** 1.744*** 0.946*** (0) (0) (0) CF (t-1) 0.553*** 0.552*** 0.228*** (0) (0) (0) CF (t-2) *** *** (0.002) (0.002) (0.316) Ln(Assets) *** *** *** *** (0.001) (0.001) (0.002) (0.379) (0.302) (0) Number of Business Segments * * * * (0.06) (0.061) (0.686) (0.09) (0.085) (0.763) R&D/Assets -0.32*** -0.32*** -0.32*** 2.09*** 2.10*** (0) (0) (0) (0) (0) (0.889) Depreciation Expense/Sales *** *** 0.006*** (0) (0) (0.003) Capital Expense/Sales 0.035*** 0.036*** (0) (0) (0.213) Ln(Firm Age) *** *** ** ** ** *** (0.004) (0.003) (0.037) (0.017) (0.023) (0) Stock Volitility *** *** *** (0) (0) (0) Lambda -.130*** -.132*** (0) (0) (0.142) (0.197) Number of Observations Censored Firms with Inside Directors Prob > χ2 / Adjusted R % %

12 Table 4. Effects of Inside Directors Acquiring Outside Directorships: A D-n-D Analysis Performance it = β 0 + β 1 Treatment i + β 2 Post-treatment t + β 3 Treatment i *Post-treatment t + Controls + ε it. CF ln(m/b) Model 1 Model 2 Model 3 Model 4 Treatment Firm (0.789) (0.654) (0.378) (0.374) Post-treatment Indicator ** ** (0.039) (0.31) (0.041) (0.194) Post-treatment x Treatment Firm 0.024* 0.022* 0.155** 0.116* (Difference-in-Difference) (0.055) (0.096) (0.013) (0.07) Number of Observations Controls No Yes No Yes Intercept Yes Yes Yes Yes R % 7.0% 4.53% 24.60%

13 Growth Opportunities R&D, Capital Expenditures, High Tech Industry Complex firms Size, Business Segments, Geographic Reach, Firm Age Competition Board composition Greater outside representation Separate CEO and Chair

14 Table 5. Inside Directors, Firm Performance, and High R&D Activity CF ln(m/b) Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 CID (%) x High R&D 0.003*** 0.004*** ** 0.015** 0.010*** (0.004) (0.003) (0.78) (0.03) (0.011) (0.005) CID (%) x Low R&D (0.296) (0.407) (0.294) (0.866) (0.854) (0.53) Non-CID (%) x High R&D ** (0.453) (0.04) Non-CID (%) x Low R&D (0.306) (0.193) F-test : CID x High R&D = CID x Low R&D 0.003* 0.003** * 0.015** 0.010** (0.067) (0.047) (0.82) (0.056) (0.017) (0.015) Number of Observations Censored Firms with Inside Directors Prob > χ2 / Adjusted R % % Strongest effect is in high R&D intensity firms

15 Table 6. CARs: Announcements of Outside Directorship Appointments Event Study Direct test of shareholder wealth impact Directorship appointments must be to unaffiliated firms 3-day CAR based on a one factor market model Panel A: Certified Inside Director Board Appointments Own Board Appointment CAR Independent Outside Board Appointment CAR N Mean Median %>0 N Mean Median %>0 Total % ** * 59%** (0.324) (0.477) (0.415) (0.049) (0.095) (0.026) Pre-SOX % % (0.896) (0.984) (0.333) (0.386) (0.599) (0.198) Post-SOX % ** ** 64%** (0.117) (0.230) (0.625) (0.013) (0.041) (0.032)

16 Panel B: Dependent Variable: Announcement CARs for Inside Director Appointments to Unaffiliated Directorships Model 1 Model 2 Model 3 High R&D 0.022** (0.037) (0.66) (0.441) 60% Independent Outsiders (0.185) (0.475) High R&D x 60% Independent Outsiders ** (0.285) (0.03) Relative Capital Expenditure/Sales 0.006*** (0.007) Relative Firm Size (0.541) Director Age < * 0.029* 0.031** (0.05) (0.069) (0.026) 2 nd Directorship (0.337) (0.334) (0.183) Busy Director *** ** (0.008) (0.023) (0.561) Director Ownership *** (0.138) (0.198) (0.006) Board Tenure (0.375) (0.55) (0.278) Firm Size 0.007* 0.007* 0.009** (0.067) (0.073) (0.042) Leverage (0.164) (0.18) (0.551) Capital Expense/Sales 0.057** 0.052** 0.076*** (0.023) (0.04) (0.001) CF (t-1) *** *** ** (0) (0) (0.011) SOX (0.322) (0.271) (0.332) Constant *** ** *** (0.004) (0.01) (0.006) Number of Observations R % 15.2% 26.8% F-test: 60% Independent Outsiders + High R&D x 60% Independent Outsiders= * (0.67) (0.058) High R&D + High R&D x 60% Independent Outsiders=0 0.03** 0.03** (0.020) (0.016)

17 Panel C: Inside Director Departures Certified Inside Director Non-Certified Inside Director CAR CAR N Mean Median %<0 N Mean Median %<0 All Departure Announcements *** ** 58%* % (0.01) (.05) (0.1) (0.89) (.99) (0.71) No Succession Announcements * % % (0.07) (.25) (0.16) (0.97) (.82) (0.68) Retirement Announcements * %* % (0.10) (.11) (0.09) (0.59) (.62) (0.87) Outside Firm Promotions % % (0.11) (.45) (0.74) (0.19) (.22) (0.87) Pre-SOX % % (0.12) (.26) (0.21) (0.63) (.58) (0.46) Post-SOX ** * 62%* % (0.04) (.08) (0.07) (0.16) (.14) (0.96)

18 M&A Decisions Cash Holdings Earnings Restatements

19 Panel B: Multivariate analysis 1 if 5-day CAR >0, 1 if withdrawn, Total 5-day CAR 0 otherwise 0 otherwise Wealth Gain Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 CID (%) 0.137*** 0.148*** 0.171*** 0.010*** *** (0.01) (0) (0) (0) (0.37) (0.01) CID (%) x High R&D 0.233** (0.02) CID (%) x Low R&D (0.51) 5-day CAR * (0.08) 5-day CAR x CID (%) * (0.07) Non-CID Inside Directors (%) ** (0.03) Deal Characteristics % Cash Financed 0.022*** 0.017*** * 0.022*** (0.001) (0.01) (0.11) (0.12) (0.32) (0.07) (0) Relative Deal Size -2.88*** -1.54** -3.14*** -3.12*** -0.1* 0.94*** (0.01) (0.04) (0) (0) (0.06) (0) (0.82) Unsolicited Bid * 2.832*** (0.35) (0.97) (0.83) (0.96) (0.1) (0) (0.32) Diversifying Bid (0.44) (0.48) (0.66) (0.8) (0.65) (0.65) (0.66) Public Target x Stock Deal -3.90*** -4.00*** -2.81*** -2.84*** -0.08* 1.00** (0) (0) (0) (0) (0.08) (0.01) Private Target x Stock Deal ** ** (0.04) (0.02) (0.15) (0.16) (0.22) (0.22) Public Target x All-cash Deal -2.86*** -2.41*** -2.39** -2.45** *** (0) (0) (0.02) (0.02) (0.14) (0.01) Private Target x All-cash Deal -2.42*** -2.26*** (0) (0) (0.29) (0.31) (0.55) (0.37) Subsidiary Target x All-cash Deal *** ** (0.01) (0.02) (0.33) (0.36) (0.95) (0.3) Cross-border Deal (0.84) (0.99) (0.45) (0.47) (0.14) (0.53) Firm Characteristics Ln(Assets) ** *** ** ** *** (0.01) (0.01) (0.01) (0.02) (0) (0.15) (0.24) Leverage * (0.25) (0.07) (0.47) (0.41) (0.21) (0.95) (0.62) ln(tobin's Q) (0.17) (0.22) (0.37) (0.47) (0.48) (0.34) (0.2) Stock Runup -0.94** -1.02** -1.22* -1.26* -0.08*** -0.91*** -1.07* (0.03) (0.02) (0.1) (0.08) (0.01) (0) (0.06) G-Index (0.66) (0.41) (0.98) (0.986) (0.89) (0.584) (0.67) RD / Assets (0.59) Inverse Mills Ratio (0.68) (0.88) (0.77) Number of Observations Year/Industry fixed effects yes/yes yes/yes yes/yes yes/yes yes/yes yes/yes yes/yes Adjusted R 2 / Prob > χ 2 / Pseudo-R % 8.14% 0.00% 0.00% 0.00% 26.25% 16.63%

20 Table 7. M&A decisions M&A Announcements Non-CID: negative 5-day CAR Multivariate analysis CIDs have a significantly greater 5-day CAR For all acquisition bids & completed acquisitions Greater effect in High R&D intensity firms Withdrawal decisions More sensitive to announcement effect in firms with CIDs Total wealth created CID firms: significantly greater than zero Non-CID firm: significantly less than zero

21 Cash Holdings Model 1 Model 2 Model 3 Model 4 Model 5 Certified Inside Directors (%) 0.014** 0.016*** 0.015** 0.016** (0.02) (0.01) (0.02) (0.03) Certified Inside Directors (%) x High R&D 0.031*** (0.01) Certified Inside Directors (%) x Low R&D (0.18) G-Index *** *** * (0) (0) (0.1) (0.11) Ln(Board Size) *** *** *** *** (0) (0) (0) (0) Insider Ownership * * *** *** (0.05) (0.05) (0) (0) CEO Duality (0.61) (0.51) (0.94) (0.9) Ln(Assets) ** ** *** 0.13*** (0.925) (0.04) (0.05) (0) (0) Leverage -1.75*** -1.72*** -1.63*** -2.60*** -2.58*** (0) (0) (0) (0) (0) R&D/Assets 4.06*** 4.06*** 4.10*** 7.04*** 6.92*** (0) (0) (0) (0) (0) Capital Expense / Sales 0.03*** 0.03*** 0.03*** 0.42*** 0.42*** (0) (0) (0) (0) (0) Stock Volatility 5.84*** 5.48*** 4.88*** 7.28*** 7.27*** (0) (0) (0) (0) (0) Inverse Mills Ratio 0.99*** 0.97*** (0) (0) Number of Observations Adjusted R 2 / Prob > χ % 48.08% 15.68% 0.00% 0.00%

22 Table 8. Cash Holdings Cash Holdings CID firms: significantly greater cash holdings Consistent with better monitoring by a wellinformed board avoids poor expenditure of cash

23 M isreported Amount Restated Irregularity Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Certified Inside Directors (%) * * ** *** ** * (0.089) (0.082) (0.015) (0.007) (0.028) (0.073) Certified Inside Directors (%) x High R&D ** (0.031) Certified Inside Directors (%) x Low R&D (0.963) Ln(Sales) 0.174*** 0.135*** 0.006*** * *** 0.004*** (0) (0.001) (0.001) (0.073) (0.147) (0.002) (0) Certified Inside Directors (%) x Ln(Sales) 0.015* 0.015* 0.001*** 0.002*** ** ** (0.062) (0.053) (0.005) (0.009) (0.387) (0.034) (0.039) G-Index 0.043** 0.042** 0.002** (0.014) (0.021) (0.043) (0.889) (0.897) (0.234) (0.202) Operating CF *** *** ** *** ** (0) (0.001) (0.015) (0.55) (0.457) (0) (0.028) Market-to-Book *** 0.005*** ** (0.537) (0.902) (0.115) (0) (0) (0.019) (0.478) Ln(Board Size) ** *** ** (0.273) (0.902) (0.015) (0.373) (0.537) (0) (0.028) Outside Director Holdings ** (0.964) (0.927) (0.115) (0.779) (0.638) (0.019) (0.478) CEO Age ** ** ** (0.031) (0.043) (0.221) (0.327) (0.291) (0.017) (0.18) Founder Present 0.276** (0.03) (0.127) (0.82) (0.118) (0.165) (0.508) (0.105) Post-SOX ** *** ** (0.024) (0) (0.015) (0.813) (0.807) (0.216) (0.89) Recent M&A 0.219** 0.238** 0.024*** (0.019) (0.013) (0) (0.564) (0.703) (0.132) (0.699) RD / Assets (0.896) Inverse M ills Ratio 0.022** 0.015*** (0.02) (0.004) F-test: * * * *** * ** * CID + CID x Ln(Sales)=0 (0.09) (0.08) (0.054) (0.007) (0.096) (0.028) (0.081) Number of Observations Clustering firms firms firms firms firms firms firms Industry/Year fixed effects no/yes yes/yes yes/yes no/yes no/yes no/yes no/yes Pseudo R 2 / Prob > χ 2 / Adjusted-R % 14.25% 0.00% 26.99% 23.23% 8.08% 0.00%

24 Tables 9. Earnings Restatements Earnings restatements CID firms: Less likely to overstate earnings Restating firms: CID firms: that do restate earnings have smaller restatement levels

25 Alternative CID measure: CID indicator Ratio of CIDs to all insiders (include the CEO) Outlier Adjustments Median regressions Winsorize data at 1% and 99% levels Endogeneity Self-Selection of inside directors (undiscovered IIDs) Treatment model (non-ceo inside directors) 2SLS IV regressions (IVs: SOX Indicator and CEO tenure) Firm Fixed Effects Results are invariant to these alternative approaches

Non-CEO Inside directors are heterogeneous and independent insider can increase firm value Outside directorships is one important mechanism for distinguishing inside directors who are more valuable

26 Non-CEO Inside directors are heterogeneous and independent insider can increase firm value Outside directorships is one important mechanism for distinguishing inside directors who are more valuable to the board and shareholders Taking into account differences among inside directors and among firms can be important in: Future research on corporate boards Policy/governance reforms CIDs create significantly more shareholder wealth than non-cids CIDs are not associated with entrenched CEOs and are associated with better board decision making

Information Content Change under SFAS No. 131 s Interim Segment Reporting Requirements

Vol 2, No. 3, Fall 2010 Page 61~75 Information Content Change under SFAS No. 131 s Interim Segment Reporting Requirements Cho, Joong-Seok a a. School of Business Administration, Hanyang University, Seoul,