Business Courts and Firm Performance Ch 2

.2 −.1 −.2 −.1 0 0 .1 .1 .2 VIII. Conclusion Overall, the results presented in this paper are broadly consistent with the assumption that access to high-quality courts benefits firm performance. The importance of these findings is substantial. Establishing good courts is a relatively cheap undertaking. For example, according to Delaware’s state budget for the fiscal year 2016, the annual cost of running Delaware’s famous Chancery Court is about $4.9 million. Hence, the costs of expert courts seem a small price to pay for a five to seven percent increase in Tobin’s q. −3 −2 −1 0 1 2 3 −3 −2 −1 0 1 2 3 Figure 2. Dependent variable: Log(ROE) Note: Regressions as described in the main text (1990-2015). Year 0 is the year in which the business court is created. Dependent variable is log (ROE). Regressions as described as in text. Both Regression controls for firm fixed effects, headquarters state year fixed effects, and complex litigation programs. The regression on the right additionally controls for takeover statutes and case law. To be included in an event sample, firms must have had at least one observation before 1992. Firms that reincorporated between 1990 and 2015 are dropped. Both regressions cluster at the level of the headquarters state. .02 .01 .02 0 .01 −.01 0 −.01 −3 −2 −1 0 1 2 3 −3 −2 −1 0 1 2 3 Figure 3. Dependent variable: Completed merger with positive CAR for target shareholders Note: Regressions as described in the main text (1994-2015). Year 0 is the year in which the business court is created. Dependent variable is a binary variable that captures whether, in a given year, a firm became the target in a completed takeover with positive CAR for the target shareholders. Regressions as described as in text. Both Regression controls for firm fixed effects, headquarters state year fixed effects, and complex litigation programs. The regression on the right additionally controls for takeover statutes and case law. To be included in an event sample, firms must have had at least one observation before 1992. Firms that reincorporated between 1990 and 2015 are dropped. Both regressions cluster at the level of the headquarters state. Table 4—Linear Probability: Determinants of the Decision Where to (Re)incorporate Dependent variable firm incorporated locally Business Court Complex Litig. Progr. Mand. Stagg. Board Constit. Stat. Pro Pois. Pill Stat. Strong Pro Pois. Pill Stat. Pro Poison Pill Case Strong Pro Pois. Pill Case Unocal Unocal Rejected Revlon Revlon Rejected Blasius Blasius Rejected Dependent variable: firm incorporated in Delaware (1) b/se (2) b/se (3) b/se (4) b/se (5) b/se (6) b/se 0.002 (0.005) -0.046∗∗∗ (0.015) 0.003 (0.004) -0.044∗∗∗ (0.010) 0.037∗∗∗ (0.009) -0.030∗∗∗ (0.006) 0.026∗∗∗ (0.007) 0.019∗∗∗ (0.005) 0.001 (0.004) -0.039∗∗∗ (0.008) 0.033∗∗∗ (0.006) -0.027∗∗∗ (0.004) 0.026∗∗∗ (0.007) 0.009 (0.006) 0.021∗ (0.011) -0.005 (0.012) -0.022 (0.019) 0.004 (0.007) -0.012 (0.010) -0.010 (0.010) 0.011 (0.007) 0.001 (0.012) 0.001 (0.004) 0.047∗∗∗ (0.013) -0.000 (0.004) 0.045∗∗∗ (0.009) -0.040∗ (0.022) 0.021∗∗∗ (0.007) -0.023∗∗ (0.009) -0.010 (0.008) 0.003 (0.004) 0.038∗∗∗ (0.007) -0.034∗ (0.017) 0.018∗∗∗ (0.007) -0.023∗∗∗ (0.008) 0.006 (0.012) -0.037∗∗∗ (0.013) 0.021 (0.013) 0.025∗ (0.014) -0.006 (0.006) 0.015∗∗ (0.006) 0.008 (0.008) -0.019 (0.012) -0.008 (0.012) N 147351 147351 147351 147351 147351 147351 R2 0.0076 0.0112 0.0123 0.0190 0.0202 0.0211 adj. R2 0.0074 0.0110 0.0120 0.0188 0.0200 0.0208 Note: Linear Probability Model. For columns 1 to 3, the dependent variable takes on the value 1 if the firm is incorporated locally in a given year, and 0 otherwise. For columns 4 to 6, the dependent variable takes on the value 1 if the firm is incorporated in Delaware in a given year, and 0 otherwise. Unlike in the other regressions in this paper, the question of whether there exists a business court or a complex litigation program and the takeover law variables refer to the headquarters state rather than to the law of the state of incorporation. All regressions include firm fixed effects and year fixed effects. Headquarters state fixed effects were omitted since they would be absorbed by firm fixed effects. All models cluster at the level of the headquarters state. * ** *** denote significance at the 10%, 5%, and 1% levels respectively. Table 5—Business Courts and Firm Performance: Regular Cohort Approach (1990-2015) Dependent Variable: Log(Tobin’s q) Business Court Complex Litigation Program Business Combination Statute Mandatory Staggered Board Constituency Statute Pro Poison Pill Statute Strong Pro Poison Pill Statute Pro Poison Pill Case Strong Pro Poison Pill Case Unocal Unocal Rejected Revlon Revlon Rejected Blasius Blasius Rejected (1) (2) (3) (4) (5) (6) 0.063∗∗∗ (0.018) 0.049 (0.036) 0.057∗∗∗ (0.018) 0.045 (0.034) 0.000 (0.015) -0.049∗∗ (0.020) -0.016 (0.024) -0.036 (0.040) 0.075 (0.057) 0.046∗∗ (0.017) 0.050 (0.043) -0.004 (0.012) -0.050∗ (0.029) -0.017 (0.025) -0.037 (0.042) 0.043 (0.054) 0.106∗∗∗ (0.037) -0.089∗∗∗ (0.029) -0.000 (0.044) -0.040 (0.046) -0.009 (0.025) 0.021 (0.021) 0.034 (0.026) 0.078∗∗ (0.030) 58166 0.034 0.0334 0.068∗∗∗ (0.014) -0.003 (0.018) 0.065∗∗∗ (0.016) -0.006 (0.018) -0.019 (0.033) -0.060∗∗∗ (0.021) -0.018 (0.018) 0.014 (0.041) -0.029 (0.062) 0.051∗∗∗ (0.017) 0.003 (0.025) -0.021 (0.028) -0.045 (0.034) -0.016 (0.019) 0.001 (0.037) -0.088 (0.067) 0.054 (0.043) -0.023 (0.038) 0.039 (0.048) -0.011 (0.037) -0.024 (0.024) 0.001 (0.019) 0.115∗∗ (0.047) 0.085∗∗∗ (0.021) 58166 0.062 0.0422 N 58166 58166 58166 58166 R2 0.033 0.034 0.062 0.062 adj. R2 0.0330 0.0331 0.0417 0.0417 Fixed Effects: Firm FE Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes No No No HQ state-year FE No No No Yes Yes Yes Note: Regular cohort approach. Dependent variable is log(Tobin’s q). Variables/controls are defined in tables B.1 and B.2. All regressions cluster at the level of the state of incorporation. To be included, firms must have at least one firm-year observation before 1992. Firms that reincorporated between 1990 and 2015 were dropped. * ** *** denote significance at the 10%, 5%, and 1% levels respectively. Table 6—Business Courts and Firm Performance: Regular Cohort Approach (1990-2015) Dependent Variable: Log(ROE) Business Court Complex Litigation Program Business Combination Statute Mandatory Staggered Board Constituency Statute Pro Poison Pill Statute Strong Pro Poison Pill Statute Pro Poison Pill Case Strong Pro Poison Pill Case Unocal Unocal Rejected Revlon Revlon Rejected Blasius Blasius Rejected (1) (2) (3) (4) (5) (6) 0.055∗∗ (0.024) -0.064∗∗∗ (0.018) 0.061∗∗ (0.024) -0.057∗∗∗ (0.020) 0.141∗∗∗ (0.035) -0.184∗∗∗ (0.050) 0.023 (0.022) -0.015 (0.033) -0.015 (0.037) 0.055∗∗ (0.021) -0.068∗∗∗ (0.022) 0.132∗∗∗ (0.044) -0.173∗∗∗ (0.060) 0.018 (0.024) -0.021 (0.031) -0.021 (0.038) 0.023 (0.032) -0.063∗∗ (0.024) 0.060 (0.048) -0.057∗∗∗ (0.021) 0.036∗∗ (0.018) 0.070∗∗∗ (0.019) 0.011 (0.035) 0.077∗∗∗ (0.023) 50675 0.012 0.0110 0.045∗ (0.027) -0.033 (0.023) 0.057∗∗ (0.026) -0.027 (0.024) 0.061∗∗∗ (0.019) -0.119 (0.076) 0.050∗∗ (0.021) 0.049 (0.039) -0.073 (0.081) 0.033 (0.024) -0.037∗ (0.020) 0.057∗∗ (0.026) -0.101 (0.087) 0.056∗∗ (0.023) 0.032 (0.035) -0.080 (0.077) 0.066 (0.077) -0.039 (0.064) 0.114∗∗ (0.053) -0.023 (0.028) 0.029 (0.025) 0.107∗∗∗ (0.014) 0.060 (0.061) 0.057 (0.038) 50675 0.041 0.0172 N 50675 50675 50675 50675 R2 0.011 0.011 0.040 0.040 adj. R2 0.0103 0.0107 0.0166 0.0168 Fixed Effects: Firm FE Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes No No No HQ state-year FE No No No Yes Yes Yes Note: Regular cohort approach. Dependent variable is log(ROE). Variables/controls are defined in tables B.1 and B.2. All regressions cluster at the level of the state of incorporation. To be included, firms must have at least one firm-year observation before 1992. Firms that reincorporated between 1990 and 2015 were dropped. * ** *** denote significance at the 10%, 5%, and 1% levels respectively. Table 7—Business Courts and Firm Performance: Regular Cohort Approach (1990-2015) Dependent Variable: Log(ROA) Business Court Complex Litigation Program Business Combination Statute Mandatory Staggered Board Constituency Statute Pro Poison Pill Statute Strong Pro Poison Pill Statute Pro Poison Pill Case Strong Pro Poison Pill Case Unocal Unocal Rejected Revlon Revlon Rejected Blasius Blasius Rejected (1) (2) (3) (4) (5) (6) 0.034 (0.029) 0.040 (0.050) 0.022 (0.027) 0.028 (0.052) 0.008 (0.023) -0.226∗∗∗ (0.030) -0.026 (0.028) -0.097∗ (0.051) 0.199∗∗∗ (0.071) 0.015 (0.026) 0.047 (0.063) 0.002 (0.027) -0.222∗∗∗ (0.037) -0.026 (0.030) -0.095∗ (0.053) 0.183∗∗ (0.069) 0.037 (0.055) -0.108∗∗ (0.045) 0.048 (0.076) -0.006 (0.041) -0.049 (0.032) 0.026 (0.035) -0.002 (0.064) 0.094∗∗ (0.046) 44559 0.029 0.0278 0.039 (0.026) 0.026 (0.038) 0.031 (0.027) 0.017 (0.043) -0.104∗∗ (0.040) -0.204∗∗∗ (0.027) -0.027 (0.027) -0.040 (0.039) 0.100 (0.077) 0.014 (0.027) 0.041 (0.053) -0.105∗∗∗ (0.036) -0.194∗∗∗ (0.039) -0.022 (0.032) -0.049 (0.039) 0.033 (0.087) -0.021 (0.085) -0.010 (0.074) 0.058 (0.078) 0.030 (0.041) -0.061∗ (0.033) -0.010 (0.036) 0.092 (0.079) 0.121∗∗∗ (0.037) 44559 0.068 0.0427 N 44559 44559 44559 44559 R2 0.027 0.028 0.068 0.068 adj. R2 0.0268 0.0273 0.0421 0.0423 Fixed Effects: Firm FE Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes No No No HQ state-year FE No No No Yes Yes Yes Note: Regular cohort approach. Dependent variable is log(ROA). Variables/controls are defined in tables B.1 and B.2. All regressions cluster at the level of the state of incorporation. To be included, firms must have at least one firm-year observation before 1992. Firms that reincorporated between 1990 and 2015 were dropped. * ** *** denote significance at the 10%, 5%, and 1% levels respectively. Table 8—Business Courts and Takeovers (Regular Cohort Approach, 1994-2015) Binary dependent variable: did corporation become a merger target in any merger in a merger with positive CAR for target shareholders negative CAR for target shareholders (1) b/se (2) b/se (3) b/se (4) b/se (5) b/se (6) b/se 0.005∗ (0.003) -0.008∗∗∗ (0.002) -0.004 (0.003) -0.008 (0.007) -0.001 (0.002) -0.003 (0.003) 0.015∗∗∗ (0.003) 0.006∗∗ (0.002) -0.010∗∗∗ (0.002) -0.004∗ (0.002) -0.013 (0.007) -0.001 (0.002) -0.003 (0.003) 0.013∗∗∗ (0.003) -0.015∗ (0.008) 0.019∗ (0.010) -0.009 (0.007) 0.005∗∗ (0.003) 0.004 (0.003) -0.003 (0.002) 0.002 (0.003) -0.009∗∗∗ (0.003) 0.006∗∗ (0.003) -0.001 (0.001) 0.003∗ (0.002) -0.009 (0.007) -0.002 (0.001) -0.001 (0.002) 0.009∗∗∗ (0.002) 0.007∗∗∗ (0.002) -0.004∗∗ (0.002) 0.003∗ (0.001) -0.012 (0.007) -0.002 (0.001) -0.002 (0.002) 0.009∗∗∗ (0.002) -0.012∗∗ (0.005) 0.016∗∗∗ (0.005) -0.002 (0.003) 0.007∗∗∗ (0.003) 0.005 (0.003) -0.002 (0.002) 0.002 (0.003) -0.009∗∗∗ (0.002) -0.000 (0.000) -0.005∗∗∗ (0.001) -0.003∗ (0.002) 0.001∗∗ (0.000) 0.001∗ (0.001) -0.002 (0.002) 0.005∗∗∗ (0.002) -0.000 (0.000) -0.005∗∗∗ (0.001) -0.004∗∗ (0.001) -0.001 (0.001) 0.001 (0.001) -0.001 (0.002) 0.004 (0.003) -0.002 (0.006) 0.003 (0.007) -0.007 (0.005) -0.002 (0.001) -0.001 (0.002) -0.002 (0.001) -0.000 (0.001) -0.000 (0.001) N R2 adj. R2 47719 0.024 0.0030 47719 0.024 0.0031 47719 0.025 0.0037 47719 0.025 0.0039 47719 0.017 -0.0039 47719 0.018 -0.0039 Baseline mean: 0.010 0.010 0.009 0.009 0.001 0.001 Business Court Complex Litigation Bus. Comb. Statute Mand. Staggered Board Constituency Statute Pro Pois. Pill Stat. Str. Pro Pois. Pill Stat. Pro Poison Pill Case Str. Pro Pois. Pill Case Unocal Unocal Rejected Revlon Revlon Rejected Blasius Blasius Rejected Controls: Takeover Statutes Yes Yes Yes Yes Yes Yes Takeover Cases No Yes No Yes No Yes Firm FE Yes Yes Yes Yes Yes Yes HQ-year FE Yes Yes Yes Yes Yes Yes Note: Regular cohort approach. Linear probability model. Dependent variable is binary and captures if corporation was, in a given year, a target in at least one completed merger (columns 1-2), or a target in at least one completed merger with positive CAR for the target shareholders (columns 3-4), or a target in at least one completed merger with negative CAR for the buyer’s shareholders (columns 5-6). All regressions control for firm fixed effects, headquarters-state year fixed effects, as well as the creation of complex litigation programs. All regressions cluster at the level of the state of incorporation. Firms that reincorporated between 1994 and 2015 were dropped. Firms must have at least one firm-year observation before 1992 to be included. B.M denotes the mean for firm-year observations in treatment states before the treatment. ”Hq-yr” means headquarters state year fixed effects. CAR and takeovers as defined in text. * ** *** denote significance at the 10%, 5%, and 1% levels respectively. Table 9—Regular Cohort Approach (1990-2015): Log(Tobin’s q) and Triple Differences Dependent Variable: Log(Tobin’s q) Business Court Business Court * High Tobin’s q Business Court * Low Tobin’s q Complex Litigation Program Business Combination Statute Mandatory Staggered Board Constituency Statute Pro Poison Pill Statute Strong Pro Poison Pill Statute Pro Poison Pill Case Strong Pro Poison Pill Case Unocal Unocal Rejected Revlon Revlon Rejected Blasius Blasius Rejected (1) (2) (3) (4) (5) (6) 0.026∗ (0.014) 0.030 (0.041) 0.051∗ (0.027) 0.046 (0.034) 0.023 (0.017) 0.030 (0.041) 0.050∗ (0.027) 0.044 (0.033) 0.019 (0.016) -0.055∗∗∗ (0.019) -0.012 (0.023) -0.026 (0.035) 0.048 (0.056) 0.015 (0.019) 0.020 (0.039) 0.052∗ (0.027) 0.045 (0.041) 0.014 (0.011) -0.053∗ (0.029) -0.016 (0.023) -0.029 (0.036) 0.015 (0.050) 0.101∗∗∗ (0.037) -0.105∗∗∗ (0.029) 0.003 (0.045) -0.052 (0.047) 0.002 (0.026) 0.019 (0.021) 0.030 (0.026) 0.103∗∗∗ (0.032) 58166 0.057 0.0550 0.021 (0.024) 0.037 (0.044) 0.074∗∗ (0.034) -0.012 (0.017) 0.020 (0.026) 0.038 (0.045) 0.075∗∗ (0.035) -0.013 (0.016) -0.004 (0.035) -0.052∗∗ (0.023) -0.011 (0.018) 0.026 (0.034) -0.060 (0.054) 0.005 (0.028) 0.032 (0.044) 0.077∗∗ (0.035) -0.008 (0.023) -0.006 (0.029) -0.040 (0.034) -0.012 (0.019) 0.014 (0.031) -0.128∗∗ (0.054) 0.058 (0.043) -0.042 (0.035) 0.026 (0.050) -0.029 (0.040) -0.011 (0.024) 0.002 (0.017) 0.109∗∗ (0.043) 0.115∗∗∗ (0.024) 58166 0.085 0.0643 N 58166 58166 58166 58166 R2 0.056 0.056 0.084 0.084 adj. R2 0.0544 0.0544 0.0636 0.0636 Fixed Effects: Firm FE Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes No No No HQ state-year FE No No No Yes Yes Yes Note: Regular cohort approach. Dependent variable is log(Tobin’s q). Variables/controls are defined in tables B.1 and B.2. All regressions cluster at the level of the state of incorporation. To be included, firms must have at least one firm-year observation before 1992. Firms that reincorporated between 1990 and 2015 were dropped. * ** *** denote significance at the 10%, 5%, and 1% levels respectively. Appendix A: Jurisdictional Framework For jurisdictional purposes, incorporating a public corporation in Delaware has two main consequences. First, Delaware courts have jurisdiction over the internal (corporate) affairs of Delaware corporations. I have described this aspect at length in part I. Second, a corporation’s decision to incorporate in a certain state has jurisdictional consequences pertaining to the corporation’s ”external affairs,” such as commercial contracts, torts, patents, etc. Corporations have a so-called place of general jurisdiction in their headquarters (”home”) state, meaning that they can generally be sued there regardless of the lawsuit’s connection to the state. By incorporating in a state, the corporation creates a second general place of general jurisdiction, meaning it can also be sued in its state of incorporation. Therefore, one might speculate that any differential impact that the creation of business courts has on locally incorporated corporations stems from more than just corporate law litigation. More specifically, one might be concerned that locally incorporated firms are sued more frequently in their headquarters state than Delaware-incorporated firms. Needless to say, this would not put the relevance of business courts to firm performance into question, but it would change the interpretation of results somewhat in that the beneficial impact on court performance could be due to to both improved litigation in corporate matters and improved litigation in external matters such as commercial litigation and contracts. However, there are strong reasons to think that third party litigation plays a very limited role, if any, in explaining the relative improvement in performance seen for locally incorporated firms. While incorporating in Delaware means creating a place of general jurisdiction there, the jurisdictional rules ensure that the effects of that change are generally slim. To begin, most non-corporate thirdparty litigation arises in the context of contractual disputes, and there the parties can freely choose the applicable forum without regard to where they are incorporated. Hence, the place of incorporation ought to have little impact on the bulk of third party litigation. Furthermore, until at least 2014, state jurisdictional rules were so generous to plaintiffs that third party plaintiffs could generally sue public corporations in any state even if they were not incorporated there: States have traditionally been very aggressive in asserting jurisdiction over plaintiffs via so-called long-arm statutes that only require minimum contacts with the state (Dammann, 2008). For example, if our hypothetical Texas-based firm produced goods, knowing that these would enter the stream of commerce and eventually be bought by consumers in Delaware, the firm could be sued in Delaware.19 In its 2014 Daimler decision,20 the U.S. Supreme Court effectively limited that practice in extreme cases, but my results do not change substantially if I focus on the time 19 See 20 134 ASUS Computer Int’l, 70 F.Supp.3d 654, 659 (D. Del. 2014)) S.Ct. 746 (2014) frame before that decision. Moreover, corporations who wish to do business in a state other than their home state, have to register to do business there. Accordingly, public corporations are routinely registered to do business in the various states including Delaware. Until a 2016 decision by the Delaware Supreme Court, Genuine Parts Company v. Cepec21 , such registration was widely assumed to be a sufficient basis for third party suits against corporations in that state.22 21 2016 WL 1569077 (Del. 2016) the previously leading case, Sternberg v. O’Neil, (550 A. 2d 1105 (Del. 1988), the Delaware Supreme Court had explicitly held that ”[i]f a foreign corporation has expressly consented to the jurisdiction of a state by registration, due process is satisfied and an examination of ’minimum contacts’ to find implied consent is unnecessary. (id. at 1113) 22 In