Dual class shares: safeguards or minefields?

First published in Business Times on September 1, 2016

By Mak Yuen Teen

THE Singapore Exchange (SGX) has now moved one step closer to allowing companies to list with dual class shares (DCS), with the Listings Advisory Committee (LAC) endorsing it with certain “safeguards”. The SGX has given assurance that there will be a public consultation before it makes its decision. I would urge all investors and other stakeholders who feel strongly against DCS to make their voices heard in this consultation – even if they feel that this may be to no avail.

It is good that the LAC understands that there are entrenchment and expropriation risks that come with DCS. However, even though the LAC is stacked with “practitioner experience”, I wonder whether they fully understand how these risks may actually play out in a DCS company and the usefulness and practicality of the safeguards they are proposing.

According to the LAC, the SGX has proposed certain measures to mitigate against the risks of poor quality companies with a DCS structure. These measures are admission of companies based on a holistic assessment and the SGX referring potential listings with a DCS structure to the LAC for advice for an initial period, until the SGX becomes more familiar with such listings (which may be about the time that we have the sequel to the “S-chip” listing frenzy). The “holistic assessment” would include consideration of factors such as industry, size, operating track record and raising of funds from sophisticated investors.

Those who have been burnt by the slew of S-chip scandals would not have much confidence in the ability or incentive of the SGX to differentiate poor-quality companies from good ones. And what does “sophisticated investors” mean? Investors such as Aberdeen, BlackRock and Fidelity, who would presumably be considered sophisticated investors, have already publicly voiced opposition to DCS. Wouldn’t such sophisticated investors avoiding DCS companies be considered a negative in a “holistic assessment”?

In my commentary “Misadventures of Alibaba, JD.com” (BT, June 4, 2014), I had pointed out that Softbank and Yahoo, two major investors in Alibaba, had conflicts of interest because they stood to benefit from commercial arrangements, and had a number of related-party transactions, with Alibaba. Therefore, it is important to assess whether such “sophisticated investors” stand to gain private benefits from supporting a company with a DCS structure.

We seem to be reverting to a “merit-based” approach to listing and herein lies the first danger – investors believing that DCS companies that list have already passed the scrutiny of all those involved in the IPO process, the SGX and the LAC.

Entrenchment risks

Let’s now consider the proposed “safeguards” to minimise entrenchment and expropriation risks. For entrenchment risks, the first safeguard is a maximum voting differential of 10:1, which is the commonly adopted voting differential in other jurisdictions. A 10:1 ratio is the problem, not a safeguard. Consider a founder who holds only Class B shares with 10 votes each and public shareholders hold Class A shares with one vote each, and there are one million total issued shares. If the founder owns just 10 per cent of the total issued shares, he will have one million votes – or 52.6 per cent of the voting rights – while the public shareholders will have 900,000 votes. This will allow him to pass all ordinary resolutions. If he wants to be able to pass all special resolutions requiring 75 per cent support, he only needs to own about 23.5 per cent of the shares. And this is assuming all shares are voted at general meetings.

While the LAC considers this as a “safeguard” against entrenchment risks, such a voting differential creates expropriation risks too. Assume that the founder who owns 10 per cent of the total shares controls the company with DCS. If he “expropriates” $100 million from the company, for example through excessive remuneration or related party transactions, the company value should fall by $100 million but his share of the loss in company value is just $10 million – and meanwhile, his private benefit is $100 million. This is why expropriation risks are higher when there is a larger “wedge” between beneficial ownership of shares (“cash flow rights”) and voting power in the company (“control rights”).

If the LAC is really interested in providing a safeguard through the voting differential, it should have proposed limiting the voting differential to something much lower than 10:1 – which, although still not ideal, would be better from the anti-entrenchment and anti-expropriation standpoint. Of course, this will make DCS less attractive, as any real safeguard would.

The second proposed safeguard against entrenchment risks is that existing companies with a one-share-one-vote structure would not be permitted to convert to a DCS structure post-listing. The LAC explains that this is because shareholders did not invest with knowledge of the risks of a DCS structure. This is the rule in the US and it is easy to copy on paper, but has the LAC considered how it will be implemented in practice? How about an existing company with a one-share-one-vote structure delisting and then later relisting with a DCS structure? Wouldn’t investors now be able to invest with knowledge of the risks of a DCS structure?

Perhaps the SGX and LAC will scrutinise such opportunistic behaviour and reject blatant abuse. But where do they draw the line? If the company delists and relists with a new name and a slight change in business, with the same controlling shareholder, would that be permitted? If the answer is yes, it is easy to beat the safeguard. If the answer is no, does it mean that a controlling shareholder who has delisted a company with a one-share-one-vote structure can never list another company with a DCS structure? The point I am making is that there are practical difficulties in such a safeguard.

The third “safeguard” against expropriation risks is the auto-conversion of multi-vote (MV) shares into one-vote shares when MV shares are sold or transferred to parties other than “permitted holders”, or when the owner-manager relinquishes his executive chairman or chief executive officer role unless there is a “compelling reason”. Auto-conversion of MV shares is practised in companies such as Alphabet (Google) and Facebook. In its public consultation, the SGX should explain what “permitted holders” and “compelling reasons” would be considered acceptable for MV shares to be transferred without auto-conversion. However, I suspect there will be a bit of a “making it up as we go along” approach in operationalising this in practice. It will be down to the wise men and women on the LAC again.

Expropriation risks

I will now turn to the proposed “safeguards” against expropriation risks. Essentially, the LAC suggests making recommendations in the Code of Corporate Governance on independence of boards and the different committees mandatory, and having the election of independent directors voted on a one-share-one-vote basis. As many of those who follow closely the corporate governance of listed companies here would know, independent directors here can be classified into “good”, “bad” or “ugly”. Just as the “good” often avoid the S-chips, they may also avoid the DCS companies that seek to list here.

Further, we very rarely hold our independent directors accountable for failing to discharge their duties. Are we going to hold them accountable for failing to discharge their duties when the odds are stacked against them? If not, how then can independent directors really be a safeguard? This is why specialists such as Charles Elson have said that when you have DCS, you are essentially outsourcing the monitoring function to third parties – to the government, the courts and the regulators. Under a DCS structure, investors need to look to these third parties to protect their interests.

Finally, the LAC has proposed safeguards to increase investor awareness. These involve clear disclosure of shareholder rights, distinctive identification of DCS companies and investor education initiatives. Here we are back to a disclosure-based approach and “caveat emptor” again.

If the SGX does proceed with DCS, perhaps it should only make such shares available to institutional investors and certain prequalified retail investors. If DCS shares can be bought by ordinary retail investors through ATMs and Internet banking, I would suggest that a danger sign flashes on the screen, accompanied by a skull, before the investor can proceed. If, as a recent study suggests, less educated retail investors are less likely to consider a modified audit opinion as important when making investment decisions, would such investors be able to tell the difference between DCS and one-share-one-vote companies?

Finally, I would like to remind the SGX board about its role under the Singapore Code of Corporate Governance. In particular, they are expected to “identify key stakeholder groups and recognise that their perceptions affect the company’s reputation”, “ensure that their obligations to shareholders and other stakeholders are understood and met”, and “consider sustainability issues as part of its strategic formulation”.

All companies must consider wider stakeholders’ interest in enhancing long-term value, but this is even more important for a company such as the SGX, for which investors in other companies listed on the exchange are a key stakeholder group whose views must be seriously considered. The board should also consider if DCS are the way to build a sustainable exchange.