Dual class stock is back in the news. Institutional investors are seeking ways to delegitimize this governance structure, with some success as reported by the Council of Institutional Investors:
Following the egregious no-vote IPO of Snap Inc. and requests by CII and other concerned investor groups, three major index providers opened public consultations on their treatment of no-vote and multi-class structures. The FTSE Russell consultation resulted in a decision to exclude past and future developed market constituents whose free float constitutes less than 5 percent of total voting power. S&P Dow Jones' consultation resulted in a broader, but only forward-looking exclusion, which bars the addition of multi-class constituents to the S&P Composite 1500 index and its components, covering the S&P 500, MidCap 400 and SmallCap 600 indexes. MSCI's consultation concluded on August 31; a decision remains pending.
I find this hysteria vastly overblown, as I shall explain in a series of posts. Today, I begin by setting the background through a review of the relevant history. As we shall see, dual class capital structures are not an historical anomaly. Rather, it is prohibitions on them that are the oddities.
Famed oilman and takeover raider T. Boone Pickens once asserted that "equal voting rights and common stock ownership are inextricably linked. Over 200 years of experience have established equal voting rights as a fundamental tenet of American democracy."[1] Pickens' rhetorical powers are considerable, but in this case his grasp of economic history was not. Worse yet, his claim is a common misconception.
In fact, however, one share-one vote is not the historical norm. To the contrary, limitations on shareholder voting rights in fact are as old as the corporate form itself.
Prior to the adoption of general incorporation statutes in the mid-1800s, the best evidence as to corporate voting rights is found in individual corporate charters granted by legislatures. Three distinct systems were used. A few charters adopted a one share-one vote rule.[2] Many charters went to the opposite extreme, providing for one vote per shareholder without regard to the number of shares owned.[3] Most followed a middle path, limiting the voting rights of large shareholders. Some charters in the latter category simply imposed a maximum number of votes to which any individual shareholder was entitled. Others specified a complicated formula decreasing per share voting rights as the size of the investor's holdings increased. These charters also often imposed a cap on the number of votes any one shareholder could cast.[4]
Gradually, however, a trend towards a one share-one vote standard emerged.[5] Maryland's experience was typical of the pattern followed in most states, although the precise dates varied widely.[6] Virtually all charters granted by the Maryland legislature between 1784 and 1818 used a weighted voting system. After 1819, however, most charters provided for one vote per share, although approximately 40 percent of the charters granted between 1819 and 1852 retained a maximum number of votes per shareholder. Finally, in 1852, Maryland's first general incorporation statute adopted the modern one vote per share standard.
Legislative suspicion of the corporate form and fear of the concentrated economic power it represented probably motivated the early efforts to limit shareholder voting rights.[7] A variety of factors, however, combined to drive the legal system towards the one share-one vote standard. Because reform efforts were almost invariably led by corporations, apparently under pressure from large shareholders,[8] it may be assumed that one factor was a desire to encourage large scale capital investment. The ease with which restrictive voting rules could be evaded also undermined the more restrictive rules. Large shareholders simply transferred shares to strawmen, for example, who thereupon voted the shares as the true owner directed.[9] Finally, while other factors also contributed, the most important factor probably was the fading of public prejudice towards corporations.[10]
By 1900, a majority of U.S. corporations had moved to one vote per share.[11] Indeed, contrary to present practice, most preferred shares had voting rights equal to those of the common shares.[12] State corporation statutes of the period, however, merely established the one share-one vote principle as a default rule.[13] Corporations were free to deviate from the statutory standard,[14] and the trend towards one vote per share reversed in the first two decades of the 20th Century as a growing number of issuers adopted dual class governance structures.
Two distinct deviations from the one share-one vote standard emerged in the years prior to the Great Crash. One involved elimination or substantial limitation of the voting rights of preferred stock. In particular, it became increasingly common to give preferred shares voting rights only in the event of certain contingencies (such as non-payment of dividends). While controversial at the time,[15] this practice is the modern norm.[16]
The more important development for present purposes was the emergence of nonvoting common stock. One of the earliest examples was the International Silver Company, whose common stock (issued in 1898) had no voting rights until 1902 and then only received one vote for every two shares.[17] After 1918, a growing number of corporations issued two classes of common stock: one having full voting rights on a one vote per share basis, the other having no voting rights (but sometimes having greater dividend rights).[18] By issuing the former to insiders and the latter to the public, promoters could raise considerable sums without losing control of the enterprise.[19]
While disparate voting rights plans were gaining popularity with corporate managers in the 1920s, and investors showed a surprising willingness to purchase large amounts of nonvoting common stock, an increasingly vocal opposition also began emerging. William Z. Ripley, a Harvard professor of political economy, was the most prominent (or at least the most outspoken) proponent of equal voting rights. In a series of speeches and articles, eventually collected in a justly famous book, he argued that nonvoting stock was the "crowning infamy" in a series of developments designed to disenfranchise public investors.[20] In essence, this was an early version of the conflict of interest argument made below: promoters were using nonvoting common stock as a way of maintaining voting control for themselves.
The opposition to nonvoting common stock came to a head with the NYSE's 1925 decision to list Dodge Brothers, Inc. for trading. Dodge sold a total of $130 million worth of bonds, preferred stock and nonvoting common shares to the public. Dodge was controlled, however, by an investment banking firm, which had paid only $2.25 million for its voting common stock.[21] In January 1926, the NYSE responded to the resulting public outcry by announcing a new position:
Without at this time attempting to formulate a definite policy, attention should be drawn to the fact that in the future the [listing] committee, in considering applications for the listing of securities, will give careful thought to the matter of voting control.[22]
This policy gradually hardened, until the NYSE in 1940 formally announced a flat rule against listing nonvoting common stock.[23] Although there were occasional exceptions, the most prominent being the 1956 listing of Ford Motor Company despite its dual class capital structure, the basic policy remained in effect until the mid-1980s.[24]
Long before 1940, Ripley had proclaimed the demise of nonvoting common stock.[25] He was somewhat premature: in the years between 1927 and 1932, at least 288 corporations issued nonvoting or limited voting rights shares (almost half the total number of such issuances between 1919 and 1932).[26] But the Great Depression, with an assist from the opposition led by Ripley and the NYSE's growing resistance, killed off many disparate voting rights plans.[27]
Even so, however, dual class governance structures did not disappear. Far from it. Famous firms such as Ford and Hershey retained (as they still do) dual class capital structures. In the period 1988-2007, moreover, dual class firms held more or less steady at approximately seven percent of the total number of public corporations.
In sum, while their popularity has waxed and waned many times, neither dual class stock nor complaints about it are new. Granted, the one share-one vote rule fairly early became the default rule. But it remained only a default rule and departures from it remained common into the 1930s. Today's dual class capital structures thus were almost more of a revival of the historical norm than a departure from it.
[1] Testimony of T. Boone Pickens, General Partner, Mesa Limited Partnership, before the Securities and Exchange Commission 10 (July 22, 1987).
[2] See 4 Joseph S. Davis, Essays in the Earlier History of American Corporations 324 (1912); 1 William R. Scott, The Constitution and Finance of English, Scottish and Irish Joint-Stock Companies to 1720 228 (1912).
[3] Samuel Williston, History of the Law of Business Corporations Before 1800 (Part II), 2 Harv. L. Rev. 149, 156 (1888). Taylor v. Griswold, 14 N.J.L. 222 (1834), is often cited for the proposition that the common law rule, in the absence of controlling statute or charter provisions, was one vote per shareholder. E.g., Jeffrey Kerbel, An Examination of Nonvoting and Limited Voting Common Stock -- Their History, Legality and Validity, 15 Sec. Reg. L.J. 37, 47 (1987); Note, Constitutionality of Non-Voting Stock, 39 W. Va. L.Q. 345, 348 n.22 (1932). Professor Ratner, however, argues that the common law in fact had no fixed rule as to corporate voting rights. See David L. Ratner, The Government of Business Corporations: Critical Reflections on the Rule of "One Share, One Vote," 56 Cornell L. Rev. 1, 9-11 (1970).
[4] For example, a Maryland charter, typical of the period 1784 to 1818, provided: "for one share, and not more than two shares, one vote; for every two shares above two, and not exceeding ten, one vote; for every four shares above ten, and not exceeding thirty, one vote; for every six shares above thirty, and not exceeding sixty, one vote; for every eight shares above sixty, and not exceeding one hundred, one vote; but no person, co-partnership or body politic shall be entitled to a greater number than thirty votes ...." Joseph G. Blandi, Maryland Business Corporations 1783-1852 66 (1934). See generally Davis, supra note 2, at 323; E. Merrick Dodd, American Business Corporations Until 1860 215 (1954); 1 Scott, supra note 2, at 340-41; Williston, supra note 3, at 157.
[5] Davis, supra note 2, at 324.
[6] Blandi, supra note 4, at 65-66. The experience in other states is summarized in Ratner, supra note 3, at 5-8.
[7] See Davis, supra note 2, at 323-24; Armand B. DuBois, The English Business Company after the Bubble Act 1720-1800 288 (1938); Dodd, supra note 4at 255.
[8] See Davis, supra note 2, at 324; Kerbel, supra note 3, at 48-49.
[9] This practice is noted, for example, in the preamble to an English act of 1766. 7 Geo. III., ch. 48. Nonetheless, it apparently was (or became) lawful. See, e.g., Moffat v. Farquhar, 7 Ch. D. 591 (1878). The same practice arose in the United States, the Maryland legislature going so far as to require each voting shareholder to swear an oath that the shares he was voting were his property and had not been acquired with the intent of increasing the number of votes to which the shares were entitled. 1836 Md. Laws ch. 264. See also Annals of Cong. 923 (1819) (resolution proposing to prohibit transfers of stock made for the purpose of evading the limits on voting rights of shares in the first Bank of the United States).
[10] See Kerbel, supra note 3, at 49.
[11] Stevens, Stockholders' Voting Rights and the Centralization of Voting Control, 40 Q.J. Econ. 353, 354 (1926).
[13] New York's General Corporation Law of 1909, for example, entitled each shareholder to one vote per share "[u]nless otherwise provided in the certificate of incorporation." 1909 N.Y. Laws, ch. 28, 23, reprinted in Joseph A. Arnold, New York Business Corporations 39 (4th ed. 1911).
[14] E.g., St. Regis Candies, Inc. v. Hovas, 3 S.W.2d (Tex. App. 1928); General Inv. Co. v. Bethlehem Steel Corp., 100 A. 347 (N.J. Ch. 1917); People ex rel. Browne v. Keonig, 118 N.Y.S. 136 (1909); Bartlett v. Fourton, 38 So. 882 (La. 1905).
[15] E.g., Stevens, supra note 11, at 389-91.
[16] Victor Brudney & Marvin A. Chirlstein, Corporate Finance 224-26 (3d ed. 1987).
[17] Stevens, supra note 11, at 355.
[18] 1 Arthur S. Dewing, The Financial Policy of Corporations 163 (5th ed. 1953).
[19] Adolf Berle & Gardiner Means, The Modern Corporation and Private Property 75-76 (1932).
[20] William Z. Ripley, Main Street and Wall Street 77 (1926).
[21] See Joel Seligman, Equal Protection in Shareholder Voting Rights: The One Common Share, One Vote Controversy, 54 Geo. Wash. L. Rev. 687, 694-97 (1986).
[24] The evolution of the NYSE policy is traced in Kerbel, supra note 3, at 57-63; Seligman, supra note 21, at 697-700.
[25] Ripley, supra note 20, at 122 ("[n]onvoting common stock, then, bears every appearance of being dead -- dead beyond recall."). See also Berle & Means, supra note 19, at 76.
[26] 1 Dewing, supra note 18, at 161.
[27] Dewing attributes their demise solely to the Great Depression, id. at 162, but it seems likely that the other factors played a role.
