On May 6, Chinese e-commerce giant Alibaba filed a registration statement in the United States, its first official step toward becoming a public company. In addition to its size and nationality, Alibaba’s plan to list under a dual-class voting structure is noteworthy. Multi-class share voting structures are frequently criticized for exacerbating the agency problem created by the separation of corporate ownership and management. These structures arguably make management less responsive to shareholders and more inclined to act in its own narrow interests. Multi-class share structures do not equate in every case to insider control, but they certainly make it much easier to effectuate. The counterargument is that a multi-class share structure allows management to avoid outside interference and the sort of short-term focus that comes at the expense of the long-term planning that is ultimately best for the company and its shareholders. In this brief, we look at the growth of multi-class share structures and ask what it means for investors.

The Rise of Multi-Class Share Structures

The number of companies with multi-class share structures has grown dramatically, from 19 in 2001 to 44 by first quarter 2014. Consumer discretionary firms account for just over one-third of the total; firms in this sector and in the consumer staples, financial, and IT sectors make up more than 80% of all such firms. All sectors except industrials, materials, telecom, and utilities have seen significant growth in the percentage of firms with multi-class share structures over the past decade. From a market-cap perspective, the trend has been even more notable, from about 2% of the S&P 500 in 2001 to 12% today.^[1] The numbers are similar for the broader Russell 3000® Index: 9.4% of constituents—11.8% of market cap—had multi-class share ownership structures as of April 30. The average sector weight of multi-class share firms by number of constituents is 7.1%, but 8.8% on the basis of market capitalization. Firms with multiple voting shares account for 22.5% of the market cap of S&P 500 consumer discretionary firms and an even more surprising 26.0% of the market cap of S&P 500 IT firms.

Following in Google’s Footsteps

The seminal event for tech firms was arguably a 2004 letter from Larry Page (who also wrote on behalf of co-founder Sergey Brin) in connection with Google’s forthcoming initial public offering (IPO). Page wrote that the “standard structure of public ownership may jeopardize the independence and focused objectivity that have been most important in Google’s past success and that we consider most fundamental for its future. Therefore, we have implemented a corporate structure that is designed to protect Google’s ability to innovate and retain its most distinctive characteristics. We are confident that, in the long run, this will benefit Google and its shareholders, old and new.”

As precedents, Page cited the voting structures of some media firms (the New York Times Company, the Washington Post Company, and Dow Jones) as well as Berkshire Hathaway. In its subsequent IPO, Google issued class A shares with one vote per share, while the class B shares (held by existing shareholders) carried ten votes per share, a common structure. Following the IPO, which was oversubscribed, Google’s executive management team and directors collectively controlled over 61% of the company’s voting rights; Brin, Page, and CEO Eric Schmidt together controlled more than 37%.

Since Google’s IPO, a number of other firms have listed under a multi-class share (normally dual class) capital structure, signifying management’s intent to retain control indefinitely. In recent years this has included Facebook, Groupon, LinkedIn, Yelp, and Zynga; after Facebook’s IPO, founder Mark Zuckerberg owned 18% of the company but controlled 57% of the voting rights. The April 2014 listing of investment bank Moelis & Company seems almost a parody of the genre—founder Ken Moelis retained 97% voting power. The previous month Google had split its class A shares, giving shareholders one non-voting class C share for each A share they held.

Reports suggest Alibaba is listing in the United States only because Hong Kong does not permit a dual-class structure. In its F-1 Alibaba emphasizes its long-term outlook, including its hope of lasting at least 102 years so its existence spans three centuries. The company plans to manage “our various business units to a single profit and loss … rather than setting compartmentalized P&L targets for each business unit,” as this would “negatively [impact] the long-term profit potential of our business.”^[2]In an email to employees written just prior to the F-1 filing, founder Jack Ma said the company would adhere after its IPO to the “customer first, employee second, shareholder third” principle … Continue reading Alibaba downplays the dual-class share issue by pointing out that its partnership structure “rejuvenates itself through admission of new partners each year.” The F-1 filing states, “unlike dual-class ownership structures that employ a high-vote class of shares to concentrate control in a few founders, our approach is designed to embody the vision of a large group of management partners.” Partners, who presently number 28, must “maintain a meaningful level of equity interests.”

Questions Shareholders Should Be Asking

The argument over what corporate structure is best for shareholders is an old one, with analysis complicated by the level of insider ownership and control rights, corporate domicile, governance provisions, and other factors. One analysis of a large sample of dual-class US firms from 1995 to 2002 found a positive relationship between firm value and greater cash flow rights and a negative relationship between firm value and greater insider voting rights, particularly where insiders “owned more than half of the voting rights … but less than half of the cash-flow rights.”^[3]Paul A. Gompers, Joy Ishii, and Andrew Metrick, Extreme Governance: An Analysis of Dual-Class Firms in the United States (2009). Another study of S&P Composite 1500® firms from 2002 to 2011 found that, contrary to the rationale advanced by their proponents, controlled companies underperformed non-controlled companies^[4]Controlled firms encompass a larger number of firms than just multi-class share firms. The Institutional Shareholder Services’ Investor Responsibility Research Center (IRRC) Institute has defined … Continue reading over a decade (while outperforming over one, three, and five years). At the same time, consistent with the theory, single-class share controlled firms materially outperformed non-controlled firms over the decade, and multi-class share controlled firms outperformed non-controlled firms over one-year periods.

So investors should be asking questions about investments in multi-share class and other controlled firms. Some to consider include:

• Are such companies overrepresented relative to their market share in investors’ long-only and hedge fund portfolios?
• Do managers apply any kind of a discount (premium) to these firms versus those with traditional voting structures?
• Do managers expect to hold positions in such firms for a longer period, consistent with company management’s stated rationale for maintaining control, or do they foresee shorter-term trading opportunities potentially arising from momentum swings driven by investors more focused on the short term?
• Do managers assess the financials and corporate actions of these companies any differently?
• Do controlled firms provide the same level of disclosure as those with a more traditional structure?
• How do dividends compare versus companies with a traditional voting structure?
• How do managers evaluate insiders’ incentives, and the conflicts they face?
• How do managers assess these firms’ creditworthiness?^[5]One just-published study found that debtholders of dual-class public companies in the United States were positively affected by excess insider control rights, manifested for example in lower cost … Continue reading

It is also worth pondering what a continuation of the trend toward controlled firms might mean for US equities’ long-term returns as well as the effect on active managers. This trend has coincided with (1) a 38.3% decline since 1997 in the number of companies listed on the NYSE and Nasdaq and (2) a large increase in US activist funds’ assets under management, now reportedly $75 billion. Might funds collectively take larger positions in the smaller number of firms with a single class of shares—driving prices up, and perhaps increasing volatility? How will activist investors be affected? Will they increasingly focus their attention on other markets, as they already appear to be doing in Europe?

The Bottom Line

The multi-class share structure employed by Alibaba is another indicator of an evolution in corporate governance. Investors should be mindful of the trend, at least in the United States, toward greater insider control of listed firms. They should be asking questions, if they have not already been doing so, about the potential impact on equity markets broadly and on their own portfolios specifically.

Contributors

Seth Hurwitz, Managing Director
Ramin Sobhany, Investment Associate

Exhibit Notes

Percentage of S&P 500 Companies with Multi-Class Share Structure
Sources: Factset Research Systems and Standard & Poor’s.

Breakdown by Sector of S&P 500 Companies with Multi-Class Share Structure
Sources: Factset Research Systems and Standard & Poor’s.
Note: Data are annual with data for 2014 as of March 31.

Percentage of Multi-Class Share Companies within Different Sectors of the S&P 500
Sources: Factset Research Systems and Standard & Poor’s.
Note: Percentage of total multi-class companies is the number of multi-class companies divided by the total number of multiclass companies.

Percentage of Multi-Class Share Companies within Different Sectors of the S&P 500 by Market Cap
Sources: Factset Research Systems and Standard & Poor’s.
Notes: Market capitalizations are in US dollars (billions). Percentage of total multi-class market cap is the sum of multi-class company market caps by sector divided by the total market cap of multi-class companies.

Ownership and Voting Structure of US Multi-Class Companies: Selected Recent Examples
Sources: Company Filings, Dealogic, and Factset Research Systems.
Notes: Voting structure of multi-class shares are generally 1 vote per class A share, and 10 votes per class B share. Exceptions for the selected sample companies are Zynga, with a common stock structure consisting of three share classes: A shares, B shares, and C shares, which are entitled to 1, 7, and 70 votes, respectively, and Groupon, with two common stock classes: A shares and B shares, entitled to 1 and 150 votes per share, respectively.