A DUAL DISCUSSION
Hong Kong is re-joining a global debate about dual-class shareholding structures. But the pressure to allow these structures at the behest of tech companies where owners wish to retain control of their companies masks structural problems, particularly as
The pressure for Hong Kong to allow for dual structure at the behest of tech company owners who wish to retain control of their companies is setting off debate
In the modern era of finance, the notion of “one share, one vote” lies at the heart of corporate governance. The idea that shareholders should have voting rights of the companies they are invested in, at a level proportional to their ownership is, in many ways, a basic tenet of capitalism.
Shareholder meetings, annual events where shareholders come together (online or in person) to vote on corporate strategy or boards of directors, have become capitalist carnivals in some cases. The most legendary are the annual shareholder meetings of Walmart and Berkshire Hathaway (Class A shareholders only). These events can feature day-long parties, music and dance performances, Q&A periods, speeches, and of course, voting.
In 2015, Hewlett-packard (HP) began holding online-only shareholder meetings, in a bid to encourage participation and cut costs. The previous year, the HP shareholders meeting saw Reverend Jesse Jackson push HP Chief Executive Meg Whitman to encourage more diversity in Silicon Valley at the highest levels. More recently, shareholder activists in Switzerland’s private banking and asset management industry have voiced concern over high pay packages for underperforming top executives. Such are the benefits of one share, one vote systems of shareholding and corporate governance. They allow shareholders to weigh in on the fate of their investments.
But in recent years, investors have come under increasing pressure, by technology companies in particular, to accept dual-class shares; that is, shares with reduced voting rights, or even no votes at all. Such share structures give company owners or founders access to capital market financing, but without ceding much, or any, control of their company.
Despite their recent prominence, these forms of dualclass shareholding structures, specifically where one class of share holds more voting rights than others, have actually been around for nearly a century. As far back as 1925, the Michiganbased car marker Dodge Brothers, which ultimately sold to Chrysler, was listed on the New York Stock Exchange (NYSE) with a dualclass shareholding structure that allowed for the owners of just 1.7 per cent of the equity to control 100 per cent of the voting power.
It wasn’t until 1940 that the NYSE prohibited listed companies from issuing any non-voting equity, and furthermore limited the aggregate voting power of any superior-voting stock to exceed 18.5 per cent of all outstanding votes. While this wasn’t an outright ban on dual-class (or multi-class) shares, it limited multi-class listings to the extent that in 1985 there were just ten such listings on the NYSE.
For decades, the “one share, one vote” rule guided US securities markets.
But in the late 80s and early 90s, increasing competition among US exchanges for listings business saw a change in policy. The major US exchanges began to allow multi-class structures with no restrictions on voting rights whatsoever for initial public offerings (IPO). The U.S. Securities and Exchange Commission (SEC) attempted to ban the practice in 1988 in an apparent effort to level the playing field between exchanges, but a court subsequently ruled against the equity market regulator.
But it was Google’s (now Alphabet Inc.) IPO in 2004 that triggered what was to become a wave of technology IPO’S with multi-class shares. In Google’s IPO, new investors would receive Class A common stock carrying one vote per share, while the founders would retain Class B common stock with ten times the number of votes per share. In 2012, Facebook similarly issued Class A and B stock with one and ten votes per share. Between 2012 and 2016, some 15 per cent of technology companies that went public in the US used dual or multi-class shares; up from eight per cent between 2007 and 2011, according to University of Florida Finance Professor Jay Ritter.
While the likes of Facebook and Alphabet offered shares at IPO with at least some voting power, Snap Inc., the company behind social media platform Snapchat, took things a lot further when they went public earlier this year, offering shares that carried no voting rights whatsoever. It was the first incidence of an IPO of non-voting stock on a US exchange. The Class C stock, held exclusively by co-founders Evan Spiegel and Bobby Murphy, ensure that the 27 and 28-year old CEO and CTO, controls nearly 90 per cent of the company’s voting power.
Proponents of such dual or multi-class structures that offer limited voting rights to shareholders argue that they give the founders, typically those with the original vision behind the company itself, the ability to focus on the long-term growth of the company rather than short-term decisions that might help bump quarterly reporting metrics at the cost of longer-term value creation. In the so-called “fast growth period” of a firm’s life cycle, company leaders may also need to focus their full attention on advancing the firm’s growth, and therefore providing some protection from outside takeover threats would be a good idea.
In the Google initial offering prospectus, Co-founder Larry Page summed up the rationale: “The main effect of this structure is likely to leave our team, especially Sergey [Brin] and me, with significant
control over the company’s decisions and fate, as Google shares change hands. New investors will fully share in Google’s long-term growth but will have less influence over its strategic decisions than they would at most public companies. Media observers have pointed out that dualclass ownership has allowed these companies to concentrate on their core, long term interest in serious news coverage, despite fluctuations in quarterly results.”
SO WHAT IS THE EFFECT ON PERFORMANCE?
Those in favour of dual-class listings consistently argue that concentrated voting power allows management to govern with little to no outside interference and eventually deliver higher returns to shareholders in return for their limited control powers. Joseph Tsai, vice chairman of Alibaba (parent company of South China Morning Post Group, which publishes The Peak), prior to the company’s IPO argued that insider control would allow management to “set the company’s strategic course without being influenced by the fluctuating attitudes of the capital markets.”
Professor Zhang Tianyu, a corporate governance specialist at the CUHK Business School states: “The benefit of allowing founders to have absolute control of their company is that they can focus on the long-term strategic development of their company without worrying about the pressure on short-term earnings from investors, especially as investors have become ever more focused on short-term gains.”
Detractors point out that dualclass structures deprive investors of their proportional voting rights, and they incentivise managers to act in their own personal interests rather than those of the company’s larger shareholders (this is known as the “agency problem”). Those most opposed to dual-class structures include corporate governance advocacy groups such as the Council of Institutional Investors (CII), the Investor Stewardship Group, and the Asian Corporate Governance Association, among others.
Past scandals offer some insight as to why such investors are leery of losing their say in corporate governance. Perhaps the most famous and egregious example of abuse of enhanced management voting rights comes from Lord Conrad Black and his publishing company Hollinger International. Lord Black owned less than 20 per cent of the company yet he controlled 68 per cent of the vote. Between 1997 and 2003, a special committee report investigation into the company found that Black and his cronies took over US$400 million (HK$3.2 billion) from Hollinger for themselves.
The report noted that Lord Black appointed all of his company’s directors, and that “the board Black selected functioned more like a social club or public policy association than as the board of a major corporation, enjoying extremely short meetings following by a good lunch and discussion of world affairs… Actual operating results or corporate performance were rarely discussed.”
Apart from possible scandals, the other key question is the extent to which dualclass shareholding structures affect long-term share price performance. Academic studies have provided mixed results, although the conclusions tend towards underperformance by dual-class companies over time.
A study in October 2012 by the Investor Responsibility Research Centre Institute investigated the relative performance of “controlled companies” listed on US exchanges. These are companies with concentrated voting control by one party, either through a substantial ownership stake or through a multi-class capital structure created specifically to facilitate disproportionate voting power in relation to equity ownership. The study found that the average tenyear total shareholder return for controlled companies with multiclass structures was 7.52 per cent, compared to 9.76 per cent for noncontrolled companies. Controlled companies with a single-share class however returned 14.26 per cent by comparison.
A more recent examination, “Does Multi-class Stock Enhance Firm Performance?” by the CII from May of this year investigated the potential impacts of different voting structures on return on invested capital (ROIC) using data on 1762 Us-listed companies on the Russell 3000 index over nine years. Overall, their models found that multi-class structures did not result in a meaningful statistical increase in long-term value creation as measured by ROIC.
MEANWHILE, IN HONG KONG
A fact often overlooked by individual investors is that stock exchanges are businesses themselves, operating in their own best interests. They compete with one another for company listings. More listings mean more trading volume and hence exchange revenue.
It was competition among the major US exchanges NASDAQ, AMEX and the NYSE, back in the 1980s that ultimately drove what many have labelled a “race to the bottom” in terms of the easing of corporate governance restrictions (i.e. allowing dual-class shareholdings) to win listings. It was this competition that the SEC sought to eliminate in 1988.
The IPO of mainland China internet giant Alibaba in late 2014 provided a stern test of Hong Kong’s listing rules, which explicitly prohibit dual-class share structures. Alibaba sought a partnership
structure allowing a group of 27 managers to appoint the majority of Alibaba’s directors, thereby giving them control of the board. While not an explicit A and B class share structure like other recent technology listings (although reportedly Alibaba did at first propose this), the structure falls into a general dual-class category, hence the Hong Kong exchange’s refusal to list the company.
The Alibaba IPO would go on to raise US$21.8 billion for the company on the NYSE, becoming the largest US IPO in history.
Corporate governance activists applauded the fact that the Hong Kong exchange had stuck to its guns. But nearly three years after Alibaba dropped their pursuit of a Hong Kong listing in favour of the US, the question of dual-class listings is now firmly back on the agenda in Hong Kong.
Earlier this year, the Hong Kong Securities and Futures Commission (SFC) backed plans for a public consultation proposed by the stock exchange regarding the creation of a third board (after the HKSE and GEM boards), aimed at technology companies, that would allow for dual-class listings. This follows a similar consultation announcement made towards the end of last year regarding dual-class listings by the Singapore stock exchange, a regional competitor to Hong Kong’s stock exchange.
“I am concerned about dualclass shareholdings because all the reasons given by the SFC to not proceed in 2015 still exist today,” says Robert Knight, former partner and CEO of law firm Heidrick & Struggles Financial Services Practice in Hong Kong and former Asia CEO of Standard Life. “The fact that just two years on, and only after Singapore’s SGX issued their consultative paper on dual listings in February, they are now reconsidering, allows one to assume that Hong Kong is more interested in competing with Singapore for future listings, rather than protecting the Hong Kong Stock Exchange, or the best interests of its shareholders, particularly the smaller shareholders.”
Jamie Allen, secretary general of the Asian Corporate Governance Association (ACGA) and a vocal advocate of “one share, one vote” for all listings, says, “Our view on dualclass shares has not changed. We think all companies, whether new economy or old economy, should follow the same high standards of investor protection and corporate governance. This is what is best for the long-term development of the Hong Kong market.”
On whether this makes Hong Kong less competitive, Allen goes on to say that, “there are many reasons why Hong Kong does not attract high-tech firms to list here. To reduce the debate to the dual-class share issue is overly simplistic.”
Perhaps the largest and most powerful opponent of such dualclass listings comes from the Council of Institutional Investors (CII), a Us-based nonpartisan, non-profit association of employee benefit plans, foundations and endowments with combined assets under management
exceeding US$3 trillion. Their member funds include major long-term shareowners with associate members being a range of asset managers with more than US$20 trillion in assets under management. The “one share, one vote” principle was among their first member-approved policies when the CII was founded in 1985.
In a letter sent earlier this year to the CEO and Chief Regulatory Officers of Singapore Exchange Limited, Kenneth Bertsch of the CII wrote that a move by the SGX to permit dual-class shares would “take a significant step in the wrong direction,” saying that such a move “conflated the interests of intermediaries with the interest of the two core market participants: investors and public companies.” He added that accepting dual-class shares in Singapore would “lead to regional competition to lower listing standards, and reflect accordingly on Singapore.”
In remarks to the US SEC Investor Advisory Committee in March, Bertsch said: “It is clear that Singapore and Hong Kong are responding to competitive pressure from low standards at the NASDAQ and the NYSE, just as NYSE was pressured to relax its rules in 1986 by the lack of restrictions on dual-class listings at NASDAQ.”
Jamie Allen from the ACGA points to Hong Kong having one of the most successful IPO markets in the world. “One of the reasons we have high liquidity and investor confidence is because we have higher standards than most other markets in this region. Dual-class shares would only erode the long-term value of our market.”
THE PASSIVE PROBLEM
There is of course a basic point that supporters of dual-class structures would make: if you are opposed to investing capital in companies that offer reduced or no voting, don’t invest. However, those who invest in passive funds may have little choice. Passive assets under management, which track an underlying benchmark equity index, have continued to grow as investors have been withdrawing from active funds that typically charge higher fees.
According to Vanguard Group, the largest provider of mutual funds and second largest ETF provider globally, and Morningstar, the Chicago-based investment research and management firm, US$264.5 billion flowed out of actively managed US equity funds in 2016, with passive index funds and ETFS taking US$236.1 billion. In the last decade, more than a trillion dollars has shifted from active to passive US equity funds, and 2016 represented the largest calendar-year change in active to passive assets.
For this growing body of passive investors, who now range from retail investors to the largest pension funds and institutions, the indices the work to may bring them into conflict with dual-class structures, depending on the
constituent stocks in the index.
For example, of the largest ten constituents of the US benchmark S&P500 index, four carry reduced voting rights. These include Facebook A shares, Berkshire Hathaway B shares, and both Alphabet’s Class A and Class C shares (which carry no voting rights whatsoever). The inclusion of dual-class shares into key benchmark equity indices will therefore drive investors into the stock, regardless of their opinions about dual-class shares and the implications for corporate governance. A current bone of contention for index providers surrounds the question of whether Snap Inc., with its non-voting shares issued earlier this year, should be included in their indices.
If the company is included, then passive investors become forced buyers of a company whose corporate governance structure they vociferously object to, and hence industry groups are heavily lobbying index providers to exclude Snap Inc.
“The growth and popularity of passive investment vehicles like index funds and exchange traded funds places a higher emphasis on how index methodologies handle shareholder rights and corporate governance issues,” says Brendan Ahern, chief investment officer for Kraneshares ETFS in China, who has been involved in the sector since 2001 when he joined Barclays Global Investors during the rollout of the ishares family of ETFS. “Institutional investors have raised their voice to protest having to buy Snap if it is included in indices despite providing shareholders absolutely no recourse on management decisions and accountability.”
The Investment Association, a trade body representing over 200 UK investment managers who collectively manage over £5.7 trillion, has lobbied major index providers FTSE Russell, S&P Dow Jones and MSCI Global to lodge strong objections to Snap Inc.’s inclusion in equity indices. Likewise, the CII has continued to lobby index providers to exclude company structures like Snap Inc. in particular.
Index providers appear to be vacillating. In March, MSCI said that Snap Inc. would qualify for its indexes, but a day later said that Snap Inc's inclusion would be re-assessed after feedback from investors. FTSE Russell announced it would defer its consideration of Snap Inc inclusion at upcoming quarterly index reviews and reconstitutions likewise pending a consultation with index users. And S&P Dow Jones Indices have stated they would require six to 12 months post- Snap Inc IPO to study the company structure accordingly. The continued decline of Snap Inc’s share price may help sharpen that discussion.
Advocates for “one share, one vote” point out that traditional equity structures have not hindered the ability of other major technology companies to deliver value to shareholders or raise capital. As Allen from the ACGA points out that successful tech firms such as Apple, Amazon and Microsoft do not have dual-class shares. Closer to home, Hong Kong-listed Chinese technology giant Tencent Holdings Ltd. has successfully built a HK$2.54 trillion-dollar company ( by market capitalisation) without the need for dual-class shares with different voting rights.
What’s clear, especially in the aftermath of Alibaba’s decision to list in the US rather than in Hong Kong, is that large technology companies will continue to shop around for exchanges that allow them to balance the need for raising capital with their preference for reduced corporate governance constraints in the form of dual-class voting structures.
As a result, both Hong Kong and Singapore are likely to find themselves under increasing pressure from their exchanges to loosen restrictions, especially if more large and potentially lucrative technology companies come out of China looking to raise capital. Of course, even if Hong Kong and Singapore were to agree on keeping a single class structure only, competition from New York is always just over the horizon.
ABOVE Google's listing in 2004 on Nasdaq started the wave of dual share listings favored by tech companies.
ABOVE Alibaba Group opted to IPO through the New York Stock Exchange because of the pre-ipo structure that allows for the board to maintain greater control and voting power.