Dual-Class Shares: From Google to Alibaba, Is It a Troubling Trend for Investors?
In March, mainland Chinese company Alibaba finally announced that it would be listing its much-anticipated IPO in New York rather than in Hong Kong. This has left regulators in Hong Kong revisiting their stance on a one-share, one-vote standard and many corporate governance practitioners around the world concerned that this may hasten a growing trend — for companies to go public with a dual-class share structure that can disenfranchise shareowners. (See our views on the topic from October 2013).
The reasoning behind dual-class share structures is that company founders want to have their cake and to eat it, too. Hong Kong wouldn’t allow that, but the New York Stock Exchange (NYSE) does. If you are wondering, CFA Institute endorses a one-share, one-vote standard.
Dual-class shares often give some shareowners more voting rights than others. In this case, Alibaba’s founders likely would be able to control the vote at the company (with more than 50% of the voting power) without having to risk their own equal capital.
A quick example demonstrates how this separation of economic ownership from control can work. If a company has a dual-class share structure in which the founder gets four votes for every one of other shareowners, and there are one million shares with four votes controlled by the founder and four million shares with one vote controlled by other shareowners, the founder can still control all decisions at the company (he has 50% of the vote) while only taking 20% of the economic risk. If that founder wishes to control 50% of the shares, why doesn’t he simply take 50% of the economic risk?
Recently, dual-class shares have been making a comeback, with companies such as Google, Groupon, Facebook, and other tech companies preferring such a structure. Before dual-class shares became all the rage with technology companies, they were quite prominent at media companies such as the New York Times, Washington Post, and News Corp. The argument was that the dual-class structure helped these media companies keep their journalistic integrity.
What Kind of Message Are These Companies Sending?
Defenders of dual-class shares claim that the structure allows them to focus more on long-term performance than on short-term returns. This is a specious argument, as there is an easy way to retain control and show shareowners that you have as much at risk as they do — just own 50% of the shares. By adopting a dual-class structure, however, companies are sending the message that they want to control a majority of the votes but not take a majority of the risk. Another way to say it is that they want the public’s capital, just not their opinions. If they don’t want to give up control of the company they can finance through debt, raise venture or equity capital, sell fewer shares so they retain control, or they can simply choose to remain a private company.
Research has shown that an ownership situation in which a founder, family, or other entity controls a company’s voting power, but does so under a one-share, one-vote standard, performs better for minority shareowners and controlling shareowners, alike. In the United States, at least, a study by the Investor Responsibility Research Center (IRRC) has shown that on average, and over time, companies with dual-class shares underperform those with a one-share, one-vote standard in which the owner’s economic risk is commensurate with his voting power. This IRRC study also found that over the long term, controlled companies with a one-share, one-vote structure tend to outperform all others. In essence, the nature of control matters. A structure in which controlling shareowners share the same link between economic risk and control as other shareowners seems to work best for all parties.
Detractors of dual-class structures state that such structures insulate bad management from the discipline of the markets. They also cite the “next generation” problem with a dual-class share structure: inevitably, the management team or individual who founded a company — no matter how brilliant — will retire, cash out, or move on to other endeavors. This often leaves the dual-class voting power in the hands of the next generation who inherits the shares, a next generation who is rarely as interested in the business or as talented as the passionate creators who came before them. For this reason, some in the governance community have called for a sunset provision (perhaps five or 10 years) in a company’s dual-class share structure.
There is also a perception problem for stock exchanges. Perhaps the New York Stock Exchange is so big that by allowing a proportionately few companies to have different listing rules does not compromise the perceived integrity of the NYSE. However, if the trickle of dual-class issuers grows to a sizable proportion of listed companies, these exchanges will have to defend against investor concerns that they are allowing lax corporate governance standards.
The answer many companies give to those who voice concerns about the corporate governance shortcomings of dual-class shares is “buyer beware.” After all, no one is forcing investors to buy these companies. Issuers often argue that as long as the dual-class structures are clearly disclosed to shareowners, they can choose to invest or look elsewhere. There are plenty of choices to invest in companies with or without dual-class shares. However, index investors and many institutions that own the market do not have such a choice, and are locked into owning companies in which their rights — and their returns — are compromised by dual-class structures.
It will be interesting to see whether this trend continues, whether the Hong Kong exchange sticks to its principles, whether dual-class structures will continue to proliferate, and whether investor concerns are met. Stay tuned.
If you liked this post, consider subscribing to Market Integrity Insights.
Photo credit: ©iStockphoto.com/little_honey
27 thoughts on “Dual-Class Shares: From Google to Alibaba, Is It a Troubling Trend for Investors?”
Excellent post on an issue that is, unfortunately, growing in importance. These companies aren’t generally going public to raise money for capital expenditures or to hire employees. They are doing so in order to be able to extract cash from the company while maintaining control as if the company remains private.
Facebook clearly said as much in their perspective. I’ve met several Silicon Valley directors who told me to my face that they would never serve on a board without a dual-class structure. They don’t like dealing with questioning shareowners.
You have it exactly right. The trouble is, those directors generally serve at the whim of the founder. Sure, emperors can often work fine but sometimes people get off balance. Think of the Roman emperor Gaius (31 August 12 AD – 22 January 41 AD) popularly known as “Caligula.”
Index funds should consider modification to exclude dual-class companies. It isn’t investing; its usually gambling on one individual.
Thanks for your comment James. We are interested to see how this issue develops in coming years. Will we see a higher proportion of companies go public with dual class shares? Will exchanges that have forbidden such a practice change their minds? Will investors make more noise about their displeasure around dual-class shares? We will have to see. We are looking through the research on these structures globally, but much of what we have found recently is about the US market. We are looking into the wider separation of ownership (economic risk) from control. Stay tuned.
Great post Matt.
Dual class shares have few, if any, attractions for investors and the “long term” argument is specious and, frankly, disingenuous. If a board is not capable of looking after the long term interests of all shareholders, then perhaps there should be a different board. If the the “controlling” interest is matched by economic risk, I have less of a problem with it, although these companies hold special risks for long term investors. However, in the case of dual class companies, it is ONLY done to “have your cake and eat it, too”. It is certainly not done for the long term benefit of shareholders with lower voting shares. Anyone who suggests otherwise is just not serious or believable.
James, thanks for your post. I am not surprised by your anecdote about Silicon Valley. It is just frightening to think of a director not wanting to be answerable to shareholders. If such directors don’t like dealing with shareholders, they should join privately held boards only and leave public companies to those who act in the interest of all shareholders. Your suggestion of excluding them from index funds has merit. I would certainly recommend long term investors not include dual class companies in their portfolio.
I hope Hong Kong can resist pressure to compromise the one share, one vote principle. Indeed, in my view, the US exchanges should not allow them for any new issues. I am not sure that will happen any time soon, though.
Thanks for your comment John. Has there been any movement by investor groups to not invest in dual-class shares? I haven’t seen it. Let me know if I missed something. I have heard investor groups lobbying exchanges in the US to not allow dual-class shares – among other things – but have not seen any large investors coming out and saying that they will not invest in them. It would make a strong statement, but I haven’t seen it. Canada has many such structures as well. I know your native Australia doesn’t allow it – which is part of the reason Mr. Murdoch changed News Corp incorporation to the US – if our readers are looking for an interesting governance story on dual-class shares from yesteryear.
I am sad to announce that on the 1st of April, a law was promulgated in France to generalize the double voting right provisions attached to registered shares which have been hold for more than two years.
As shares are only recorded in the registry of shareholders by the major shareholders and by few individual shareholders, the “one share-one vote principle” is also violently attacked in France.
There is one solution remaining : that key investors engage and request a change in by-laws!
Thank you for your comment Loïc. And thank you for the update on the French market. I know there has been much debate recently about awarding such “loyalty shares” in other markets as well, but after much discussion, most parties came to the conclusion that such shares may solve one problem but create another – essentially giving longer-term shareholders more power keeps them from being disenfranchised, but it disenfranchises other shareowners.
Thank you Matt.
I am not so skeptical about loyalty shares. Proxinvest hosted this morning in its Paris office a workshop about loyalty warrants with an in-depth prestation by F. Samama from Amundi and support from P. Desfossé, CEO of the French biggest pension fund (ERAFP) (same presentation than this one : http://www.unpri.org/viewer/?file=wp-content/uploads/F.Samama15NovSolferino.pdf ).
I concluded that those loyalty warrants were quite acceptable as they respect the “one share-one vote” principle. They must be surely used carefully to create a small incentive for longer-term investors which would not dilute too much the other shareowners.
I also welcome the impact on the increasing cost of shares borrowing which is a major barrier to an efficient voting process and monitoring of the listed firms.
Great article. This is widespread in Brazil, where many companies implement the separation of ownership and control by issuing 2 or even 3 different classes of shares. This is the case of most state controlled companies, such as Petrobras and Eletrobras, and many corporate governance conflicts often arise from that. Usually, the shares with less political power or even no voting rights, have preference over dividends and minimum dividends guaranteed. There has been a trend favoring new issues to come on a one share/one vote basis, but these examples from US will likely provide arguments to the supporters of the abusive structures that have prevailed in the country in the past.
Thank you for your comments Bruno.
And thank you for informing us about the dual-class share structures in Brazil. We are familiar with the one-share, one-vote rules for Novo Mercado-listed companies, and will have to stay informed about the prevalence of dual-class share structures in Brazil and around the world. Take care, Matt
The other problem is that the the Nasdaq 100 index decided to include only tge the non Google shares in its membership. Hence all equity option contract have been now readjusted with the non voting GOOG share as underlying. This will cause the voting share to lose liquidity very quickly. Pushing investors again towards the non voting share.
Quite an awkward proposition forthose shareholders who will still want to hang on to their voting shares…the exchange is basicly telling you. Well little guy, i am sorry but we decide to side with the big guys. If you still want..some marketability…switch…or else !
Great article Matt. The comments that have been posted have also been very insightful.
I think it is great that you clearly stated what the CFA Insitute position is. That is a very helpful.
Is there anything more we can do at the CFA Institute to push the NYSE in this case to instill best in class governance for listings? Have we thought about developing a listing “code” perhaps to guide these transactions?
Just one additional comment. One reason that HK is so insistent against dual class shares is that local companies tend to be closely held by a family, whereas mainland companies that list in HK are often PRC-state owned enterprises. The justified fear is that dual class shares will allow the controlling shareholder to take all of the profits and leave the ordinary shareholder with nothing. The rule is relatively new (adopted in 1987) and I think it was specifically set up with Mainland China in mind. The Mainland has adopted similar rules.
In addition, the HK markets are more “reputation based” than “disclosure based.” The rules for disclosing financials are weak, and people will invest in companies based on the reputation of the shareholders. The one class rule makes sure that the controlling shareholder (who in the case of Mainland China is often nominally an agency of the Chinese government, and in practice is the Communist Party) won’t take all of the money, but will share any profits with the shareholder. Most of the profits *will* go to the controlling shareholder, but the rule is in place to make sure that they don’t get greedy and try to take everything.
This has been a particular problem with state-owned enterprises, and much of the problem with Russia has been the lack of a legal structure to prevent controlling shareholders from taking everything.
One other reason the one class rule is important has to do with the unique nature of state owned enterprises in China. Rather than being controlled by one agency, it is often the case that an SOE will be owned by several different agencies, and several different local governments. Allowing different classes of shares would allow the most politically powerful actor to take all of the money, whereas having the one class rule allows a provincial government or city government to invest its money into a company with funds from the central government and be reasonably assured that they won’t be squeezed out completely. The fact that the rule is being enforced by a politically neutral arbiter (i.e. the Hong Kong Stock Exchange) is an added bonus.
As a Canadian it always infuriated me that there were quite a few firms with dual share structures with different voting rights. I came across this article http://www.benefitscanada.com/investments/asset-classes/one-share-one-vote-9502 which while not recent comes to the same conclusion as IRRC using Canadian companies as its basis.
I should note that in the Canadian study on some metrics the dual share class companies outperform the single share class companies in some metrics but that their share price performance been inferior to their peers and indexes. Sorry for the multiple post. I shouldn’t post before I get my morning coffee.
Thank you for your comments Joseph.
Dual-class shares are indeed a hot issue globally with the imminent Alibaba IPO. Many Investors in the US are concerned about the issue, and the recent regulatory filing by Alibaba which was short on crucial details — such as where the company will be listed — has not allayed any of those concerns.
Assuming efficient market hypothesis, as investors start becoming aware of the concept of dual class shares, they will start pricing it accordingly. The disclosures need to be stringent in this case.
Alternatively, promoters can offer additional benefits/yields for the voting right trade off. This could benefit a class of investors who would have not voted in any case.eg technical analysts
Investors already do discount shares of companies with dual-class shares. Here is a link to just one such study: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1894910.
Yes, transparency is key. However, many shareowners who own the whole market (indexers) have no choice but to own such companies, and push for a share structure such as “one-share, one vote” that would unlock that value.
Thanks for your article. There is no question in my mind that the dual class stock structure can create material financial risk for investors due to conflicts of interest between the controlling, minority holder and the majority of shareholders . A great example of this is Swift Transportation.
Swift’s dual class stock structure provides the company’s founder and CEO Jerry Moyes majority control over a company for which he holds minority stake and he uses the company’s coffers like his personal piggy bank. Moyes has pledged 62.8% of his equity stake in the company as collateral for personal loans which has created material risk for public shareholders, and the company has engaged in hundreds of millions of dollars in related party transactions with other Moyes-controlled businesses. A Teamster shareholder proposal this year seeking an end to Swift’s dual class stock structure received 79.2% support by shareholders of the company’s Class A, publicly traded shares.
The Swift story is not that of a start-up looking for an on-ramp to the public markets. It was founded in 1966. Moyes stepped down as CEO and Chairman of Swift in 2005 after agreeing to pay $1.26 million to settle an insider trading Department of Justice investigation. In 2007 he took the company private and then public again in 2010 but, this time, with a dual class stock structure.
Thank you for your comment.
Studies tend to show that dual-class shares underperform. It is therefore incumbent upon management to prove that a dual-class structure is best for shareowners. In many cases this does not appear to be the case.
The argument that such a structure is needed to ensure that management has the room to plan for the long term simply does not ring true. An owner/founder could simply decide to keep 50% of the shares of the stock without the need to separate ownership from control.
Power corrupts. Absolute power corrupts absolutely. There is a much greater chance that management will act for their own interests instead of shareholders’ when giving the absolute power.
Thanks for your comment. That is definitely a concern among shareowners. Shareowners argue that management should simply keep more shares with equal voting rights if they wish to maintain control. The separation of ownership from control is where problems can arise.
Hechinger’s and Wang Labs were some other examples of dual share class strategies that did not end well.
If I recall Hechinger’s had A shares for everyone with extra voting rights, I believe 4 or 5 per share and then issued B shares. The B shares had only one vote per share but the B paid a $0.12 dividend while the A share paid $0.10. Large holders including funds and institutions converted to earn the extra dividend to add fractionally to their returns. This was at the expense of losing their voting rights and when the company started its long slide they had little power to affect change. At some point this should become a discussion of fiduciary responsibility as well.
Thank you for your addition to the discussion.
That is why we often see a discount for dual-class shares. Investors — some of whom are fiduciaries — understand the risks of dual-class shares and bake that into the valuation.
I think one thing is missing in this discussion. The proponents of dual class structure usually argue that If a company has room to grow but do not have enough money to invest more in the company and the owners do not want to lose control so for them dual class structure will be suitable. The other argument which the proponents of dual class structure usually gives is that mostly family owned firms choose this structure type and they hesitate to give controls to others outside their family. Family owned firm will not destroy the shareholders value.
Thank you for your comment. Our argument for a one-share-one-vote policy is that we like to see economic risk commensurate with voting power. We argue that by going to the public markets, founders are agreeing to share ownership of the company with the other shareowners in some way. If founders want to control the companies’ voting power, they should control 50% of the shares. If those founders implement a dual-class share structure, those founders are gaining more power, while taking less economic risk.
Studies have also shown that over time, most family firms underperform when control of the company is handed off to the next generation, which is likely not as motivated as their predecessors were to make the company a success.
1) Does a company need to incorporate in a certain state (e.g., Delaware, which has more flexible laws for corporations) to have dual-class shares, or is this available throughout the US? I’m specifically wondering about California.
2) You mention it’s allowed on the NYSE. What about NASDAQ?