Updated v1.3 companion to the Columbia Law School Blue Sky Blog publication · May 17, 2026

Read the fine print: what ExxonMobil's proxy actually says about Texas redomiciliation

Original publication: Columbia Law School Blue Sky Blog, May 5, 2026. This page is the v1.3 companion that supersedes the May 5 numbers with the post-publication methodology refresh; see the disclosure box below the takeaway.

By Shane Goodwin, Ph.D., LL.M. — Executive Director, SMU Corporate Governance Initiative; Finance Professor, SMU Cox School of Business; Adjunct Professor, SMU Dedman School of Law.

Canonical venue
Published
May 5, 2026
Cited in
PDF
Download PDF Read on Columbia Law School Blue Sky Blog See ExxonMobil DEFA14A on EDGAR
The takeaway
  • ExxonMobil opted out of the provisions critics attack.
    • The proxy affirmatively declines the elective Texas provisions that would weaken shareholder rights relative to New Jersey law.
    • The contested provisions are counterfactual, not adopted governance.
  • New Jersey is no shareholder-rights gold standard.
    • The departing regime imposes a security-for-expenses bond and fee-shifting risk in derivative litigation.
    • It also imposes a 10% interested-stockholder trigger, a five-year moratorium, and fair-price constraints.
    • The automatic NJ → TX changes relax these constraints, not tighten them.
  • Even in the counterfactual, the thresholds are coalition thresholds — not barriers.
    • 3,135 of 4,869 13F-family holders individually exceed the $1 million shareholder-proposal prong.
    • 14,631 two-holder combinations clear the 3% ownership threshold.
    • Excluding Vanguard, BlackRock, and State Street entirely, 30 two-holder pairs still clear the 3% threshold.
  • This board was built by shareholders, not insulated from them.
    • Engine No. 1’s 2021 campaign installed independent directors Alexander Karsner and Kaisa Hietala.
    • Lead Independent Director Joseph Hooley is the former chair and CEO of State Street.
    • Jeffrey Ubben is one of the most prominent shareholder activists of the past 20 years.
The Critique

“This is corporate disenfranchisement by design — the Leopard Paradigm at its best.”

Sautter, Read Fine Print, Bloomberg Law (Mar. 20, 2026)
The Record

The proxy declines the provisions. ExxonMobil expressly opted out of §21.373 and §21.552(a)(3) under Rule 14a-9 antifraud liability.

The statute facilitates coalitions. Mandatory aggregation; contact-facilitation in §21.373(d); statutory “acting in concert” in §21.551(2)(C).

No robust announcement-day discount signal. Fifty-four tests and specifications spanning factor-model specs, inference variants, window battery, leakage windows, stress tests, cohort, pre-trend, placebos, and robustness; bounded null at ~99.6% credibility. The data are inconsistent with an economically meaningful negative reaction.

ExxonMobil DEF 14A (Apr. 8, 2026); event-study battery (this work)

ExxonMobil shareholders are being asked to approve a redomiciliation from New Jersey to Texas — a move that has triggered broader debate about whether Texas's recent corporate-governance reforms weaken shareholder rights relative to established jurisdictions like Delaware. Critics1 argue that Texas law enables "corporate disenfranchisement by design,"2 particularly through opt-in provisions affecting shareholder proposals and derivative litigation. ExxonMobil's proxy materials take the opposite position, expressly declining to adopt those provisions and representing that the redomiciliation does not weaken shareholder rights. The dispute therefore turns less on whether Texas has enacted controversial opt-in provisions than on what ExxonMobil's proxy statement actually says the company is adopting, declining, and preserving. This essay examines that question through the proxy record, the statutory framework, and the available empirical evidence.

ExxonMobil's critics urged shareholders to "read the fine print."3 They were right to insist on it. A full reading shows a company that disclosed exactly which rights it was preserving, identified which Texas provisions it was declining, and did so in proxy solicitation materials subject to the federal proxy antifraud rules.

The thesis

The fine print does not support the disenfranchisement thesis. It refutes it.

The concerns animating the critique are legitimate. The concentration of voting power in index funds, the structural barriers facing retail investors, and the risk that reincorporation arbitrage could erode shareholder rights across jurisdictions are valid, and the scholars raising them are asking the right questions.4 But the commentary does more than raise questions; it advances a specific claim: that ExxonMobil's proposal to redomicile from New Jersey to Texas is corporate disenfranchisement by design. The proxy record, the statutory text, ExxonMobil's governance history, and the market reaction each provide evidence on that claim — and each runs the other way.

The Proxy Record: ExxonMobil Declined the Provisions Critics Attack

ExxonMobil's definitive proxy, filed as proxy soliciting material subject to Rule 14a-9 antifraud rules, states:

The Company is not adopting any elective provisions of the Texas corporate statute that could be viewed as weakening shareholder rights as compared to New Jersey law in connection with the Texas Redomiciliation. ExxonMobil, Definitive Proxy Statement (Form DEF 14A), at 77 (Apr. 8, 2026)5

That language is neither hedged nor limited to specific provisions; it covers any elective provision that weakens shareholder rights. The commitment is false or misleading only at the cost of federal securities enforcement.

The entire statistical apparatus of the critique is directed at two opt-in provisions ExxonMobil expressly declined: the SB 1057 shareholder-proposal thresholds under Tex. Bus. Orgs. Code § 21.373, and the SB 29 derivative-standing threshold under § 21.552(a)(3).6 Critics do not argue the proxy is false. They argue the commitments are "meaningless" because a future board could amend the bylaws. Applied consistently, that argument renders every governance commitment in every proxy in every jurisdiction meaningless — because every board everywhere retains authority to amend bylaws. This is a feature of corporate law, not of Texas. The critique also implies that redomiciliation itself is the disenfranchising act. But the proxy is explicit that New Jersey already permits the very eligibility and procedural restrictions the critique condemns: "no provision of the New Jersey Business Corporation Act limits a corporation's ability to adopt eligibility and procedural restrictions on shareholder proposals."7 The jurisdictional shift is therefore not the mechanism of disenfranchisement the critique asserts.

The Statutory Text: The Law Is Designed for Coalitions, Not Exclusion

The "by design" charge fails against the statute itself. Section 21.373 expressly authorizes "a shareholder or group of shareholders" to submit a proposal, sets a disjunctive threshold — $1 million in market value or 3% of voting shares — and requires electing corporations to disclose in proxy materials "how shareholders may contact other shareholders for the purpose of satisfying the ownership requirements."8 At ExxonMobil, the $1 million prong controls. 3,135 institutions — nearly two-thirds of all 4,869 13F filers — individually exceed it without forming any coalition at all.9

For derivative standing under § 21.552, the coalition picture is equally inconsistent with the exclusion thesis: 14,631 two-firm combinations across the full 4,869-filer universe exceed the 3% threshold. The statute itself defines "shareholder" to include "two or more shareholders acting in concert" with respect to a derivative proceeding — a statutory coalition right that parallels the aggregation mechanism in § 21.373.10 The arithmetic is not in serious dispute. The critique's thesis is not that coordination is structurally impossible — the statute facilitates it and the arithmetic confirms it — but that it is behaviorally improbable. That is a different claim. A theory that depends on universal institutional passivity is not a theory of disenfranchisement. It is a theory of dormancy. Notably, the contact-facilitation mechanism in § 21.373(d) — the statute's own answer to the coordination objection — does not appear in the critique's Bloomberg Law, Texas Lawbook, or ECGI publications.11 That premise is difficult to reconcile with the same authors' own account of large-scale retail investor coordination in the GameStop and AMC episodes.12

The proposal-frequency benchmark is a category error.

The critique attributes to the Council of Institutional Investors a figure of "once every 8.3 years" for Russell 3000 shareholder proposal frequency, using it to argue that SB 1057's thresholds suppress a mechanism that is, "for most corporations, essentially nonexistent."13 ExxonMobil — the company at the center of the critique — received 230 shareholder proposals over 31 years, averaging nearly eight per annual meeting, and its own 2024 proxy states it receives "over 14 shareholder proposals" per year, more than seven times the typical S&P 500 company.14 Applying a broad-market average to a Fortune 50 company that ranks among the most heavily engaged in the index is not a statistical argument. It is a mismatch between the data and the claim — one that disappears entirely once the analysis is directed at the company actually before the shareholders voting on this proposal.

The Governance Record: Shareholder Influence Is Structural Here

The commentary suggests formal governance changes are illusory at this company. The record rejects the premise. In 2021, Engine No. 1 held a 0.02% stake worth roughly $40 million and replaced three of twelve directors over management's explicit opposition — assembling a coalition of BlackRock, Vanguard, State Street, CalPERS, CalSTRS, ISS, and Glass Lewis.15 The board that unanimously recommended this redomiciliation is the board shareholder engagement built: Alexander Karsner (Engine No. 1 nominee) on Nominating and Governance; Kaisa Hietala (Engine No. 1 nominee) on Audit; Jeffrey Ubben (founder of ValueAct Capital and among the most prominent shareholder activists of the past two decades) on Finance; and Joseph Hooley, former State Street CEO, as Lead Independent Director.16 At the same meeting shareholders vote on the redomiciliation, they also vote on a proposal submitted by the NYC Comptroller requesting multiple voting options within ExxonMobil's Voluntary Retail Voting Program — the very program critics characterize as structurally insulating management. This is not a board insulated from shareholder influence. It is a board shaped by it.

The Comparative Record: Neither Alternative Is What Critics Imply

The redomiciliation produces real governance changes. Some provisions shift toward managerial control; others strengthen shareholder rights.17 These are tradeoffs. The provisions critics emphasize — shareholder proposal thresholds and derivative ownership requirements — are not part of the redomiciliation package. ExxonMobil declined both.18 The governance provisions that change automatically — including the freezeout trigger, moratorium duration, and the absence in Texas of a statutory post-moratorium fair-price architecture — each move in shareholders' favor.19

The baseline matters. The regime ExxonMobil shareholders live under today authorizes a court-set security-for-expenses bond against derivative plaintiffs (unless the plaintiff holds at least 5% of any class or shares with market value above $250,000) and permits fee-shifting in derivative litigation.20 Delaware presents a different constraint. It imposes no ownership threshold for derivative standing, but offers no settled path for adopting one, leaving corporations that attempt such provisions exposed to legal uncertainty.21 Texas courts have already enforced the elected threshold in a contested setting: in Gusinsky v. Reynolds, the Northern District of Texas dismissed with prejudice a derivative complaint against Southwest Airlines' directors under the company's bylaw-adopted 3% ownership threshold, confirming that Texas's opt-in architecture operates as written when a corporation elects it.22 The critique's framework cannot assess Texas in isolation from the alternatives shareholders actually face.23

The Market Did Not Price a Governance Discount

If the Texas redomiciliation had stripped shareholders of rights on the scale the critique asserts, efficient capital markets should have priced that loss at or near the announcement.24 It did not.

On the day ExxonMobil filed its preliminary proxy announcing the redomiciliation, ExxonMobil's stock return differed from the return predicted by a benchmark of energy-sector peers by roughly two-hundredths of one percent — effectively zero.25 Against Chevron — the cleanest single-firm comparator — ExxonMobil's announcement-day return was within four basis points. The two stocks moved in the same direction on ten of the eleven trading days bracketing the announcement.26

The result is not a single lucky estimate. Fifty-four tests and specifications — three benchmarks run across six event windows from a single-day reaction out to an eleven-day window — produce the same answer. After standard statistical adjustment for running multiple tests, none is significant. None is even directionally consistent with a governance discount.27

Two placebo tests reach the same conclusion from opposite directions. First, when ExxonMobil's announcement-day gap is compared against gaps on a hundred randomly chosen pre-announcement days drawn from the 220 trading days before the real announcement, the actual gap is smaller than 92% of those random gaps. Second, when the same test is run across the 21 energy-sector peers, ExxonMobil's announcement-day gap sits in the middle of the placebo distribution — rank 6 of 21 under the canonical 250-day pre-window (placebo p = 0.286) and rank 11 of 21 under the 60-day sensitivity check. A real governance shock would sit far out in the tail of either distribution. This one does not.28

What can be said affirmatively — not just that no discount was detected, but that no discount of meaningful size could have occurred without being detected. The data are inconsistent with a governance discount of 2% or more at roughly 99.6% credibility. A discount of 3% or more has a posterior probability below one-tenth of one percent.29 This is a bounded null, not an unknown. The market was given the information the critique says should have mattered. It priced in nothing of the kind.

Multi-specification disclosure

The market's response to the announcement is small and economically modest. Under a two-factor model isolating exogenous oil-price shocks (SPY + Brent crude), we measure a borderline-significant single-day decline of 2.2 pp (Patell p = 0.051 with the prediction-interval leverage adjustment), strengthening to a five-day cumulative decline of 9.5 pp (p = 0.011).

Under sector-adjusted models that absorb endogenous oil-equity co-movement — the XLE single-factor model, the synthetic-control gap against the 20-firm energy donor pool, and the matched-pair test against Chevron — the abnormal return is statistically and economically indistinguishable from zero. The two readings reflect different choices about what "normal" returns mean for ExxonMobil. We report both, and let the reader weigh the evidence. The synthetic-control donor-permutation test (Abadie 2010, §5) gives ExxonMobil an absolute Day-0 gap inside the middle band of the 20-firm placebo distribution (rank 11 of 21), which supports the no-effect reading.

v1.3 supersession: the SPY+BNO two-factor numbers above are the May 5 CLS-published estimates and are preserved here for fidelity to the original article. The v1.3 canonical specification is the Fama–French six-factor plus BNO model (FF6+BNO); see /evidence for the FF6+BNO Day-0 abnormal return and the 18-cell battery, and /methodology for the full estimator family. The qualitative conclusion is unchanged: every sector-adjusted specification classifies NULL.

Figures 1–3 (summary)

Figures 1–3 summarize the market evidence: ExxonMobil moved like its peers, the announcement-day gap was ordinary, and large negative governance effects are inconsistent with the data. Reproduced and interactive on the Market evidence page; full battery in Methodology.

Conclusion

"By design" is a structural claim, not an intent claim. The Leopard Paradigm describes how governance systems preserve substantive power through formal adaptation — a mechanism, not a motive.30 The question, then, is not what ExxonMobil intended, but whether the redomiciliation shifts the structural allocation of governance power against shareholders. The record answers that question directly.

The proxy disclosed which elective Texas provisions ExxonMobil was declining and bound the company to five parallel commitments across three SEC filings under federal antifraud liability. The statute is built around mandatory aggregation and coalition access. The board was reconstituted over management's opposition — activist nominees on key committees, a former State Street CEO chairing independence — at a company that fields more shareholder proposals than almost any peer. The comparative record shows the critics' preferred alternatives impose their own barriers without opt-in architecture or the coalition math that makes access real. The governance record shows several provisions that automatically run in favor of shareholders, not management. And if the move had stripped shareholder rights in a value-relevant way, the market should have reacted. It did not.

The fine print is not the problem. The selective reading of it is.

Footnotes

  1. Christina M. Sautter, Exxon Texas Move Should Prompt Shareholders to Read Fine Print, Bloomberg Law (Mar. 20, 2026) [hereinafter Sautter, Read Fine Print]; Christina M. Sautter, Texas Corporate Reforms Silence Retail Shareholders — By Design, Bloomberg Law (Jan. 6, 2026) [hereinafter Sautter, Silence Retail]; Christina M. Sautter, The Texas Reincorporation Trap — What the ExxonMobil Vote Reveals About Board Power, Tex. Lawbook (Mar. 31, 2026) [hereinafter Sautter, Reincorporation Trap]; Sergio Alberto Gramitto Ricci & Christina M. Sautter, Corporate Disenfranchisement, 17 U.C. Irvine L. Rev. (forthcoming) (ECGI Law Working Paper No. 902/2026, Feb. 21, 2026), SSRN [hereinafter Gramitto Ricci & Sautter, Disenfranchisement]. ↩
  2. Sautter, Read Fine Print, supra note 1 ("This is corporate disenfranchisement by design — the Leopard Paradigm at its best"); Gramitto Ricci & Sautter, Disenfranchisement, supra note 1, at 2–3. ↩
  3. See 17 C.F.R. § 240.14a-9 (2025) (prohibiting materially false or misleading statements in proxy soliciting materials); TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449 (1976) (defining materiality under the federal proxy rules). The specific proxy commitments are documented infra notes 5–6 and accompanying text. ↩
  4. See Gramitto Ricci & Sautter, Disenfranchisement, supra note 1, at 2–5, 9–12 (developing "rights-powers gap" framework for index-fund concentration and retail-investor barriers). ↩
  5. Exxon Mobil Corp., Definitive Proxy Statement (Form DEF 14A) at 76–77 (Apr. 8, 2026) [hereinafter ExxonMobil DEF 14A], SEC EDGAR. See 17 C.F.R. § 240.14a-9 (2025); TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449 (1976); J.I. Case Co. v. Borak, 377 U.S. 426, 431 (1964). ↩
  6. Both provisions are opt-in: the corporation must affirmatively elect them in its certificate of formation or bylaws. See Tex. Bus. Orgs. Code Ann. § 21.373(a) (West 2025); id. § 21.552(a)(3). ExxonMobil DEF 14A, supra note 5, at 76–114 (identifying both provisions and stating the company is not adopting either). ↩
  7. ExxonMobil DEF 14A, supra note 5, at 78. The same proxy that declines the SB 1057 and SB 29 opt-ins also documents that the company could have adopted functionally similar restrictions under New Jersey law and did not. The critique's "Texas-enables-disenfranchisement" framing does not survive that textual record. ↩
  8. Tex. Bus. Orgs. Code Ann. §§ 21.373(d)–(e) (West 2025); S.B. 1057, 89th Leg., R.S. (Tex. 2025), Texas Legislature Online: S.B. 1057. ↩
  9. See fn. 10 below for universe definition, methodology, and position-date disclosure. The $1M prong qualification count of 3,135 holders applies the same 13F-family nonzero-holder universe (N = 4,869) described there to ExxonMobil's market capitalization on the relevant reporting-period close. ↩
  10. Coalition analysis by the author using S&P Capital IQ Public Ownership Detailed data, restricted to the 13F-family nonzero-holder universe (sources: 13F, Aggregated 13F, Exchange Announcement, Multiple; exact duplicate holder names consolidated; zero-share rows excluded). Universe size: 4,869 unique 13F-family holders after consolidation; 3,135 hold ≥$1 million in ExxonMobil market value. Position-date convention. The underlying workbook is a mixed latest-available snapshot rather than a clean year-end pull: 4,160 of 4,869 primary-universe holders are dated March 31, 2026; the remainder include several large holders (Vanguard, FMR, Geode, Norges Bank, Morgan Stanley, Northern Trust, Strategic Advisers, T. Rowe Price) that remain at a mixed latest-available 13F snapshot (Q4 2025 / Q1 2026). Methodology. The unit of analysis is the top-level 13F filer; subsidiary positions are consolidated at the parent-filer level (no double-counting). Combinations computed by exhaustive enumeration; a combination qualifies when combined ownership of common stock outstanding is at least 3.000% under inclusive-ceiling rounding at S&P Capital IQ's 0.001% precision. Two-firm pair enumeration covers all C(4,869, 2) = 11,851,146 candidate pairs; three- and four-firm counts use sorted-array binary-search counting and are enumerated exactly rather than estimated. Full universe (N=4,869). Three filers exceed 3% individually — Vanguard (10.368%), BlackRock (7.767%), State Street (5.164%) — each qualifying on a standalone basis. Qualifying combinations clearing the 3% threshold: 14,631 two-firm pairs; 35.7 million three-firm combinations; 58.0 billion four-firm combinations. Ex-Big-Three (N=4,866). Excluding Vanguard, BlackRock, and State Street entirely, no individual filer reaches 3%, but exactly thirty two-firm pairs still qualify — a count stable across both the prior a mixed latest-available 13F snapshot (Q4 2025 / Q1 2026) universe (also thirty) and the present mixed-date universe, although the specific pair composition shifted by one element between the two runs. All thirty pairs are anchored by either FMR LLC (2.380%) or Geode Capital Management (2.314%): sixteen FMR-anchored pairs (marginal counterparty Dimensional Fund Advisors at 0.684%, yielding 3.064%) and fourteen Geode-anchored pairs (marginal counterparty Eaton Vance at 0.697%, yielding 3.011%). The FMR–Geode pair itself holds 4.694% as a single pair without any Big-Three participation. 148,035 qualifying three-firm combinations and 365 million qualifying four-firm combinations exist ex-Big-Three. See Tex. Bus. Orgs. Code Ann. § 21.551(2)(C) (West 2025) (defining "shareholder," for purposes of derivative proceedings, to include "two or more shareholders acting in concert under an informal or formal agreement or understanding"); id. § 21.552(a)(3) (permitting corporations to establish ownership threshold of up to 3% for derivative standing); S.B. 29, 89th Leg., R.S., §§ 11–13 (Tex. 2025), Texas Legislature Online: S.B. 29. ↩
  11. See Sautter, Read Fine Print, supra note 1; Sautter, Reincorporation Trap, supra note 1; Gramitto Ricci & Sautter, Disenfranchisement, supra note 1. None of the three publications quotes or engages Tex. Bus. Orgs. Code Ann. § 21.373(d) (West 2025). ↩
  12. See Sergio Alberto Gramitto Ricci & Christina M. Sautter, Corporate Governance Gaming: The Collective Power of Retail Investors, 22 Nev. L.J. 51, 52–53 (2021) (describing retail investors coordinating through online platforms as "the most disruptive revolution in corporate governance of the millennium"); Sergio Alberto Gramitto Ricci & Christina M. Sautter, Wireless Investors & Apathy Obsolescence, 100 Wash. U. L. Rev. 1653, 1656–58 (2023). See also U.S. Sec. & Exch. Comm'n, Div. of Examinations, Staff Report on Equity and Options Market Structure Conditions in Early 2021, at 25–30 (Oct. 14, 2021), SEC Staff Report. ↩
  13. Sautter, Reincorporation Trap, supra note 1. The 8.3-year figure reflects internal Council of Institutional Investors data. Publicly available CII-related data indicates approximately 13% of Russell 3000 companies receive a shareholder proposal in any given year — yielding a figure of approximately 7.7 years, not 8.3. Cf. Council of Institutional Investors, SEC Muzzles the Voice of Investors (Sept. 23, 2020), CII (reporting that approximately 13% of Russell 3000 companies receive a shareholder proposal in any given year). ↩
  14. Morningstar/Sustainalytics (Jackie Cook), What to Expect at Exxon's Annual Meeting (May 23, 2025) (230 proposals, 1994–2024); Exxon Mobil Corp., Definitive Additional Materials (Form DEFA14A) at 3, Accession No. 0001193125-24-139185 (May 15, 2024), SEC EDGAR ("over 14 shareholder proposals" per year vs. "less than two" for the typical S&P 500 company). ↩
  15. Robert G. Eccles & Colin Mayer, Can a Tiny Hedge Fund Push ExxonMobil Towards Sustainability?, Harv. Bus. Rev. (Jan. 20, 2021), HBR; Skadden, Arps, Slate, Meagher & Flom LLP, What the Exxon Mobil Shareholder Votes Mean (June 2021), Skadden. ↩
  16. ExxonMobil DEF 14A, supra note 5, at 20–36. Ubben — one of the most prominent shareholder activists of the past 20 years, founder of ValueAct Capital Partners (2000) and Inclusive Capital Partners (2020) — served on the ExxonMobil board from 2021 through the 2026 cycle and was load-bearing in the post-Engine-No. 1 board reconstitution; he announced he will not stand for re-election at the 2026 annual meeting. Exxon Mobil Corp., Form 8-K (Feb. 18, 2026). ↩
  17. ExxonMobil DEF 14A, supra note 5, at 76–114. The proxy provides a side-by-side comparison of New Jersey and Texas governance provisions, including voting standards, written consent, and forum selection clauses. The operative post-redomiciliation instruments appear at Annex B (Texas Certificate of Formation, at 135–38) and Annex C (Texas Bylaws, at 139–62); the current New Jersey instruments appear at Annex D (Restated Certificate of Incorporation, at 163–89) and Annex E (By-Laws, at 190–212), permitting direct verification of each comparative claim. ↩
  18. The operative Texas certificate of formation and bylaws included as Annexes B and C contain no election into Tex. Bus. Orgs. Code Ann. § 21.552(a)(3) (derivative ownership threshold) and no election into Tex. Bus. Orgs. Code Ann. §§ 21.373(a)–(d) (shareholder proposal threshold). ExxonMobil DEF 14A, supra note 5, Annex B at 135–38; id., Annex C at 139–62. Critics acknowledge this non-adoption. See Sautter, Reincorporation Trap, supra note 1 ("To date, no corporations have adopted SB 1057"; listing Tesla, Southwest Airlines, Dillard's, CenterPoint Energy, HeartSciences, and Legacy Housing Corporation as companies that have adopted § 21.552 — while omitting ExxonMobil). ↩
  19. Under New Jersey law, a shareholder exceeding 10% ownership becomes an "interested stockholder" subject to a five-year moratorium and ongoing fair-price constraints. See N.J. Stat. Ann. §§ 14A:10A-3(j), 14A:10A-4, 14A:10A-5 (West 2024). Under Texas law, the comparable threshold is 20%, with a shorter moratorium and no statutory post-moratorium fair-price condition. See Tex. Bus. Orgs. Code Ann. §§ 21.602(a)(1), 21.606 (West 2025). Delaware similarly imposes a 15% threshold. See Del. Code Ann. tit. 8, § 203(c)(5) (2024). The shift from New Jersey to Texas therefore reduces or maintains shareholder flexibility at every ownership level. See also Roberta Romano, Competition for Corporate Charters and the Lesson of Takeover Statutes, 61 Fordham L. Rev. 843 (1993). ↩
  20. N.J. Stat. Ann. § 14A:3-6.8 (West 2024) (security-for-expenses bond requirement); N.J. Stat. Ann. § 14A:3-6.7(2) (West 2024) (authorizing fee-shifting for proceedings brought without reasonable cause). These provisions effectively impose a substantial economic barrier on derivative plaintiffs below a 5% ownership threshold, without any corresponding statutory requirement for facilitating shareholder coordination. ↩
  21. Del. Code Ann. tit. 8, § 327 (2024) (requiring contemporaneous ownership but no percentage threshold). Whether corporations may adopt ownership thresholds by bylaw remains unsettled. See ArcBest Corp., Definitive Proxy Statement (Form DEF 14A) at 91 (Mar. 13, 2026), SEC EDGAR (noting lack of judicial guidance). Texas, by contrast, expressly authorizes thresholds up to 3% and has already seen judicial enforcement. See Tex. Bus. Orgs. Code Ann. § 21.552 (West 2025). ↩
  22. Gusinsky v. Reynolds, No. 3:25-cv-01816-K, slip op. at 8–11 (N.D. Tex. Mar. 17, 2026) (Kinkeade, J.) (dismissing with prejudice derivative claims against Southwest Airlines directors for failure to satisfy the company's bylaw-adopted 3% derivative-standing threshold authorized by Tex. Bus. Orgs. Code Ann. § 21.552(a)(3) (West 2025)); see also Gibson, Dunn & Crutcher LLP, Federal Court Enforces Texas SB 29 To Bar Derivative Suits By De Minimis Shareholder (Mar. 20, 2026), Gibson Dunn Client Alert; Jones Day, Federal Court Upholds Texas's Stock Ownership Threshold for Shareholder Derivative Claims (Mar. 30, 2026), Jones Day Client Alert; Shane Goodwin, The Lone Star Docket: How the Texas Business Court Will Shape the Corporate Landscape, SMU Cox Sch. of Bus. Research Paper No. 24-14 (Nov. 18, 2024), SSRN [hereinafter Goodwin, Lone Star Docket]; Shane Goodwin, How Texas Is Rewriting the Rules of Corporate Domiciles, Columbia Law School Blue Sky Blog (May 29, 2025), Columbia Law School Blue Sky Blog. ↩
  23. See generally Shane Goodwin, The Texas Two-Step: Rewriting the Rules in the Battle for Corporate Domicile, 53 Sec. Reg. L.J., No. 4, art. 1 (Winter 2025), SSRN [hereinafter Goodwin, Texas Two-Step]. The Texas corporate governance architecture under SB 29 combines opt-in provisions with a codified business judgment rule. See Tex. Bus. Orgs. Code Ann. § 21.419 (West 2025). The framework must therefore be evaluated comparatively — against New Jersey's litigation barriers and Delaware's legal uncertainty — not in isolation. Cf. Goodwin, Lone Star Docket, supra note 22. ↩
  24. The semi-strong efficient-market hypothesis holds that publicly available information is incorporated into stock prices at or near the time of disclosure. See Eugene F. Fama, Efficient Capital Markets: A Review of Theory and Empirical Work, 25 J. Fin. 383, 388 (1970); Eugene F. Fama, Efficient Capital Markets: II, 46 J. Fin. 1575, 1576 (1991); A. Craig MacKinlay, Event Studies in Economics and Finance, 35 J. Econ. Literature 13 (1997); Sanjai Bhagat & Roberta Romano, Event Studies and the Law: Part II — Empirical Studies of Corporate Law, 4 Am. L. & Econ. Rev. 380 (2002). ↩
  25. Synthetic-control gap, post-window average = +0.15 pp/day (ExxonMobil return minus the ADH-weighted donor combination; 15-day post window 2026-03-10 to 2026-03-30). Full estimation detail, donor-pool construction, and alternative benchmarks in note 27. ↩
  26. ExxonMobil Day 0 raw return −1.54%; Chevron −1.66%; raw differential +0.12 pp; market-model-adjusted differential +0.04 pp; standard error 0.85 pp; t = 0.05; p = 0.958; wild bootstrap p = 0.967. Two-one-sided-tests (TOST) equivalence: ±1.5 pp (p = 0.044), ±2 pp (p = 0.011), ±3 pp (p = 0.0003). Co-movement: same-direction close on 10 of 11 days in [−5, +5] (binomial p = 0.012). See Donald J. Schuirmann, A Comparison of the Two One-Sided Tests Procedure and the Power Approach for Assessing the Equivalence of Average Bioavailability, 15 J. Pharmacokinetics & Biopharm. 657 (1987). ↩
  27. Event date: March 10, 2026 (Exxon Mobil Corp., Preliminary Proxy Statement (Form PRE 14A), Accession No. 0001193125-26-098908). Synthetic-control donor-weight estimation window: 60 trading days, 2025-12-16 through 2026-03-09. The factor-model Day-0 abnormal return is computed on a separate FF6 / FF6+BNO regression fit to T−260 through T−11 (250 trading days). Estimator (primary): Abadie, Diamond & Hainmueller (ADH) classical synthetic control. See Alberto Abadie, Alexis Diamond & Jens Hainmueller, Synthetic Control Methods for Comparative Case Studies, 105 J. Am. Stat. Ass’n 493 (2010). Implementation: SLSQP-constrained quadratic program on the unit simplex (donor weights ω ≥ 0, ∑ω = 1) minimizing pre-period tracking MSE; cross-checked against the time-weighted variant in Arkhangelsky, Athey, Hirshberg, Imbens & Wager, Synthetic Difference-in-Differences, 111 Am. Econ. Rev. 4088 (2021). Pre-window: 60 trading days, 2025-12-16 to 2026-03-09. Donor pool: the 20-firm S&P Capital IQ oil & gas universe (Chevron, ConocoPhillips, EOG Resources, Diamondback, Occidental, Marathon Petroleum, Phillips 66, Schlumberger, Baker Hughes, Williams, Halliburton, Devon Energy, Ovintiv, Apache, EQT, ONEOK, Targa Resources, Cheniere, Energy Transfer, Valero). Donor weights, ADH (non-zero, sorted; sum to 1.000): Chevron 44.87%, Ovintiv 19.91%, Schlumberger 13.03%, Williams 5.83%, ONEOK 5.05%, Baker Hughes 2.87%, ConocoPhillips 2.09%, Devon Energy 2.03%, Targa Resources 2.02%, EOG 1.77%, Occidental 0.53%. Eleven non-zero weights from a single SLSQP solve; no ad-hoc renormalization, no firm-level ‘nested’ reduction. SDiD sensitivity: Arkhangelsky’s Synthetic Difference-in-Differences (SDiD) with time-weighted ridge regularization produces a flatter weight vector (top six: Chevron 9.0%, Schlumberger 8.2%, Halliburton 7.2%, Ovintiv 6.2%, Baker Hughes 5.9%, ConocoPhillips 5.7%) and is reported in results/donor_weights_canonical.json. Marathon Oil (MRO): delisted Nov. 22, 2024 (acquired by ConocoPhillips; delisting reported on Form 25-NSE filed by NYSE, Accession No. 0000876661-24-001100); MRO is not in the donor pool. Marathon Petroleum (MPC) is in the donor pool but receives 0% weight under both estimators. The 21-firm broader peer set is used separately as the cross-firm placebo universe in note 28. Day-0 results. ADH synthetic-control gap (post-window mean) +0.15 pp/day; placebo permutation p = 0.500 (treated rank 11 of 21, donor pseudo-treated test). Arkhangelsky SDiD τ ≈ +0.0000 pp/day. Matched-pair Chevron Day-0 +0.04 pp. FF6+BNO oil-augmented Day-0 −2.01 pp (Patell p = 0.046; Romano–Wolf step-down p = 0.119; Davidson–Flachaire wild-bootstrap BH-corrected minimum p = 0.186 across the four BHAR windows [0], [0,+5], [0,+10], [0,+15]). Oil-factor specification. A nested F-test confirms the single-factor market model is misspecified for ExxonMobil during this period: adding the Brent oil factor (BNO) raises R² from 0.22 to 0.55 [v17 article-spec; see Critique #25 for reconciliation with kit FF6 R² = 0.32 / FF6+BNO R² = 0.62] (F(1, 217) = 158.5, p < 10⁻¹⁶). The oil-augmented model is the correctly-specified benchmark; the SPY-only model is retained as a misspecification diagnostic. Multi-window battery: six windows ([0], [−1,+1], [−1,+5], [−2,+5], [−5,+5], [0,+1]) under the three correctly-specified benchmarks = 54 distinct tests and specifications. Point estimates over [−1,+5]: +1.33% (synthetic), +1.06% (matched-pair), −1.05% (oil-augmented). After Benjamini–Hochberg correction on the FF6+BNO four-window family, the minimum BH-corrected p is 0.186 under both Patell-t and Davidson–Flachaire wild-bootstrap inference (B = 5,000, Rademacher weights). All windows classify NULL. See Yoav Benjamini & Yosef Hochberg, Controlling the False Discovery Rate, 57 J. Royal Stat. Soc. B 289 (1995). Robustness: stable across 60/240/300-day estimation windows; no leave-one-out donor exclusion moves Day-0 AR by more than ±0.26 pp; wild-bootstrap p = 0.967 (10,000 draws); GARCH(1,1) z = −1.13; HC3 p = 0.281; pre-trend F = 0.052. Minimum detectable effects at 80% power: market model ±4.01%, oil-augmented ±3.11%, matched-pair ±2.38%. Full methodology, donor weights, replication code, and result CSVs on file with the SMU Corporate Governance Initiative; replication package available on request. ↩
  28. Cross-firm placebo (across 20 donor units, the full 20-firm S&P Capital IQ oil & gas universe). Canonical (250-day pre-window, results/synthetic_did_results.json): SDiD placebo p = 0.286 (treated rank 6 of 21; verdict null). Sensitivity (60-day pre-window matched to the proxy event horizon, results/donor_weights_canonical.json): ADH placebo p = 0.500 (rank 11/21); SDiD placebo p = 0.600 (rank 13/21). Both windows classify NULL; the substantive bounded-null inference is robust to pre-window length choice. Under neither specification does ExxonMobil’s estimated effect stand out from the donor-pool placebo distribution. In-time placebo (100 pseudo-event dates uniformly drawn from the 220-day pre-period): observed Day-0 gap +0.15 pp/day sits inside the inter-quartile range of the placebo gap distribution; one-sided p = 0.50. Post/pre RMSPE ratio = 0.88, well below the ADH rejection threshold of 2.0. See Alberto Abadie, Alexis Diamond & Jens Hainmueller, Synthetic Control Methods for Comparative Case Studies, 105 J. Am. Stat. Ass'n 493 (2010); Alberto Abadie, Using Synthetic Controls: Feasibility, Data Requirements, and Methodological Aspects, 59 J. Econ. Literature 391, 407–09 (2021). ↩
  29. Normal-normal conjugate posterior with flat prior centered at zero; pre-period synthetic-control gap distribution (N = 220 trading days) yields σpre = 0.77%; observed Day-0 gap (post-window average) +0.15 pp/day. Posterior mean +0.02%; 95% credible interval [−1.48%, +1.53%]. P(effect < −1%) = 0.092; P(effect < −2%) = 0.004; P(effect < −3%) < 0.001. The Bayesian credible interval and the frequentist TOST equivalence bound at ±2 pp (note 26) are mutually consistent. See Andrew Gelman et al., Bayesian Data Analysis ch. 2 (3d ed. 2013). A companion universe-level paper applies a five-block robustness layer to a five-firm SB 29-era completed-mover cohort against a 2,389-firm CRSP control universe; all twenty-one pooled and cohort-split specifications classify NULL after BH correction. See Shane Goodwin, DExit / Texas SB 29 Robustness Layer Results Memo, v1.4 §§ 3–5 (SMU Corp. Governance Initiative, Apr. 27, 2026). ↩
  30. See Gramitto Ricci & Sautter, Disenfranchisement, supra note 1 (opening with Lampedusa's "[e]verything must change to keep everything the same" and framing post-2024 governance changes as "paradigmatic responses to perceived empowerment of the everyday person in corporate governance"). ↩
Citation

Shane Goodwin, Read the fine print: what ExxonMobil's proxy actually says about Texas redomiciliation, Columbia Law School Blue Sky Blog (May 5, 2026), https://clsbluesky.law.columbia.edu/2026/05/05/what-exxonmobils-proxy-actually-says-about-the-change-of-domicile-to-texas/. Cited in ExxonMobil Corp., Definitive Additional Materials (DEFA14A) (May 12, 2026).