There are known knowns; there are things we know that we know.
There are known unknowns; that is to say, there are things that we now know we don't know.
But there are also unknown unknowns – there are things we do not know we don't know.
— United States Secretary of Defense, Donald Rumsfeld
The famous observation above misses an important category. There are also unknown knowns; there are things you don't know you know. Until recently, I had a pocket of knowledge in the last category: what leads an activist to a particular company.
Two Minute Version:
| || |
There are three pieces to a successful activist campaign. First is some perceived hidden value. Second, the ability to buy in and own it. And finally there is the activist's ability to uncover that value, either by convincing management to take some action or forcing them to through the shareholder voting process. A company must have all three to be interesting to a (sane) activist.
Below I've unpacked it and given some common factors that lead to activism, but the beauty is that once you've grasped it, you can apply it to your own situation and find the relevant factors yourself. Its a framework that will allow you to access a company's chances of being attacked. It is also quite useful for a company, which wishes to reduce their chances of such an encounter, as well as, for an activist seeking to maximize returns.
WSJ's recent in depth write up of the opposition to DELL's going private transaction is unfortunately marred by some shoddy analysis. It is important to highlight because if these kinds of mistakes are made by DELL's advisers it could sink the whole ship. For the sake of people basing decisions based on the type of information in the WSJ's article, I must take issue with the quoted analysis of Toni Sacconaghi of Sanford Bernstein. Here is the offending paragraph:
Some analysts said a majority of Dell's shareholder base are still likely to support the deal. Sanford C. Bernstein analyst Toni Sacconaghi estimated that roughly 20% of Dell's shares are owned by merger-arbitrage traders "who will vote for the deal," he said, adding that "nearly 50% of Dell shares outstanding have turned over since the deal first leaked."
Mr. Sacconaghi added that 8% of Dell's owners are passive index funds that "will very likely vote for the deal," and that about 60% of Dell's 25 largest shareholders after Mr. Dell purchased their stock on average at less than $15 a share, "suggesting some among them may support the deal."
The three points attributed to him are all, at best, misleading. I’ll take them one at a time:
- Sanford C. Bernstein analyst Toni Sacconaghi estimated that roughly 20% of Dell's shares are owned by merger-arbitrage traders "who will vote for the deal," he said, adding that "nearly 50% of Dell shares outstanding have turned over since the deal first leaked."
I have no estimate of arbs voting for Dell’s deal and so don’t take issue with the 20% quoted, but left unsaid is Mr Sacconaghi’s guess as to what part of the 30% remaining will likely vote against the deal. Is it 5%(Good for Dell), 20%(Good for Southeastern and T. Rowe), 30%(Very Good for Southeastern and T. Rowe)? These are all very different situations, so without this number, the information contained in this statement is ZERO.
Mr. Sacconaghi added that 8% of Dell's owners are passive index funds that "will very likely vote for the deal,"
The latter half of this statement is simply not true. Passive index funds tend to vote with management, except when they don’t. The more useful statement is that they tend to vote together and they tend to vote with the crowd. The fact that the two largest shareholders, themselves respected institutional investors, have come out against the deal dramatically raises the chances that the Vanguards of the world with vote against the deal as well. (Get in touch for links to a great academic study on this point)
- and that about 60% of Dell's 25 largest shareholders after Mr. Dell purchased their stock on average at less than $15 a share, "suggesting some among them may support the deal."
This statement also seems completely empty of content. I could simply point to the fuzziness of the quantitative estimate of “some shareholders”, but let’s charitably assume that some shareholders is a lot, say even more than half. We still can’t take anything from this statement because if about half of shareholders (60%) bought their stock at less than $15 (and so some among them might support the deal), then it follows that about half of shareholders (40%) bought their stock at more than $15 (and so some among them might be against the deal). To put things more clearly the whole thing is a wash.
Investment fund, T Rowe Price, has opposed Michael Dell's plans to take Dell private. For details on the opposition see the WSJ write up here. The gist is that T Rowe Price and Southeastern Asset Management have a blocking position on the deal, but they are crucial players to get the deal done. Here is what Rotary Gallop’s Control Analysis Reveals:
This analysis assumes that Southeastern Asset and T Rowe Price vote against the deal and all other shareholders are as likely to vote for it as against it.
Note 1: There are two factors that, given reliable information, would improve the calculation:
- It is not a surprise that Dell is a prime target for going private. High insider ownership is big plus for these transactions. According to our soon-to-be-released Control and Vulnerability in the S&P500 Study, Dell ranks in the top 10th percentile for insider’s control. They are number 4 in the Information Technology sector and number 1 in their sub industry. (You’ll be hearing a lot more about this study in early March)
- Because insiders will not be voting, S&T (Southwestern Asset Management + T Rowe Price) has 14% of the effective vote. That is more than 3 times the next largest shareholder.
- This give S&T Control (read the deciding vote) in 86% of all possible outcomes to this shareholder vote, and a 93% chance of winning.
- No one else has control over 12%. Only 4 other shareholders have more than 5% control.
- In order for S&T to lose, there would have to be near-uniform support for the deal from all other shareholders.
- When you add the fact that abstention counts as a no-vote it becomes effectively impossible for Dell to win without S&T's support.
- Therefore if the deal is going to move forward Dell will have to find a way to please S&T.
Note 2: The above analysis uses the current known list of shareholders, and does not account for likely large number of arbs who are now in the stock. You'll find reference in the linked WSJ story to an estimated 20% of voting stock held by arbs, and some analysts hoping that those shares will make it an easy win for DELL. No such luck! We'll ride right past the fact that arbs can take both sides of a deal, especially one with hints of trouble, and simply note that 20% doesn't come close to matching the ~15% of shareholders who are publicly against the deal when you add in any reasonable estimate of abstention, which can ranges from 10-20%. )
- An estimate of the percent of outstanding stock that will abstain from voting - This will likely fall somewhere between 5 and 12%. This factor will decrease Dells chances of winning.
- An estimate of the Arb position in the stock and what fraction of it is positioned FOR the deal - This factor may improve Dell's chances of winning, but it depends drastically on the estimate.
As we wrote last month, Starboard Value's chances of winning a proxy battle at Office Depot are less than half of what they were at AOL, which didn’t turn out well even with support from ISS. The common problem for Jeff Smith and the Starboard Value team is that in both AOL and now Office Depot, they are at the mercy of a larger shareholder with MUCH more control over the outcome of a proxy battle.
Then: Starboard Value's Proxy Battle Fate at AOL was Decided by One Large Shareholder. First lets look at what really happened at AOL. As the graph below shows, Starboard had much less control (in blue) in AOL, than top shareholders (see caption for explanation).
The Distribution of Control and Ownership Among Shareholders at AOL: This graph shows the percentage Ownership (in grey) and the %Control (in Blue) of the largest shareholder's at AOL prior to the recent Proxy Battle. Notice that in both AOL and ODP (see figure below) Starboard’s Control (circled in red) is very small compared to the larger shareholders. This means that Starboard's proxy future comes down to their ability to convince those few large holders to side with them. In the case of AOL that did not happen.
Starboard Value's control, circled above in red, was only about 17%. That means there was less than a one in five chance that starboard's vote would matter in the proxy battle. Now compare this to the control of the larger shareholders, especially Dodge and Cox who had control in nearly 60% of possible outcomes to the shareholder vote!
Now, Starboard’s chances of winning, on the face of things, didn't look terrible at AOL. After all, There was a 58% chance that they would come out victorious. However, that number changed drastically depending on the vote of a single shareholder, Dodge and Cox. In fact, to a large extent the outcome was determined by which side got Dodge and Cox's vote. When Dodge and Cox eventually decided to vote with management, Starboards chances of winning plummeted to just 29%.
Aside for proxy battle experts and aspirants: People are generally friendly and will often tell you what you want to hear. It is common for both sides to be sure they have a proxy battle in the bag, because of large shareholders being nice to both sides. It would be unsurprising to learn that both AOL's Management and Starboard Value went into the vote thinking Dodge and Cox was on their side. That is why it is so crucial to use methods of analysis that are able to handle reality: massive uncertainty in how people will vote, even if they have told you they are with you.
Now: Starboard Value's Fate at Office Depot is Also Likely to be Decided by One Large Shareholder The graph below is the same as above, only now showing the current shareholders in Office Depot. Notice how eerily similar Starboard's strategic position at Office Depot is to AOL:
BC Partners with just over 20% of the vote has a whopping 86% control! (BC partners is a preferred shareholder who gets to vote their shares on an as-converted basis. We'll cover this more in the next post.) Since BC Partners is currently contractually required to vote with the recommendation of Office Depot's board(a condition of their preferred investment), we know they will vote against Starboard in any potential proxy battle. That only leaves Starboard Value a 14% chance of winning a proxy battle at Office Depot. That is a rough position for someone who is by far the largest common stock holder!
- Again Starboard Value is far from the largest shareholder, though much closer here
- Again we have a large shareholder - BC Partners -who basically controls the outcome of a potential proxy battle.
The Distribution of Ownership and Control Among Shareholders at Office Depot: This graph shows the percent ownership (in grey) and the %Control (in Blue) of the largest shareholders at Office Depot today. Notice that in both AOL (above) and ODP, Starboard’s Control (circled in red) is very small compared to the larger shareholders, especially in the case of BC Partners at Office Depot. Note: Here %Ownership is actually percent of the vote, because BC Partners has a large chunk of the vote through preferred shares that do not show up as common-stock ownership.
What does this mean for Starboard Value (and coattailers)?It's not all bad news for Starboard: Office Depot is deja vu in more ways than one. Though they lost the proxy battle at AOL, some of their proposals were instituted and AOL's stock appreciated by ~70% while they held it. In that sense too, Office Depot looks very similar to AOL. Namely the share price is up 100% since Starboard started buying in. Starboard Value may be losing battles, but so far they are winning the war! And despite the long odds, Starboard appears to be gearing up for a fight, with their recent retainer of Bob Nardelli, the former chief of companies such as Home Depot and Chrysler, and Joseph Vassalluzzo, a former vice chairman of Staples Inc. Both are powerful choices for an opposition slate of directors!
Next time we'll cover the Office Depot innovation that Starboard was attempting to counter with their recent letter to independent directors. Depending on your view, it is either the greatest or the worst thing to hit the field of corporate governance in years. I call it the Permanent White-Knight Defense. Starboard might call it the Chained White Knight Defense. Whatever your take, it makes the poison pill, the subject of much gnashing of teeth, look like a kitten in a pretty pink bow, so tune in next week!
We had a great talk with Paula Schaap at The Deal recently about the Starboard Value/Office Depot Activist Situation which we've been following closely It is a fascinating case study in corporate defense. Starboard's chances of winning a proxy battle at Office Depot are about half as much as they were at AOL, where they recently lost. We'll be covering some of our in-depth analysis here over the next couple of weeks so stay tuned. Until then Paula covers our thoughts on Starboard best chance of turning things around, so check out the article
Owen Thomas over at Business Insider had a great article on the ownership breakdown of one of our favorite, power to the people companies, Square: REVEALED! Here's Who Owns How Much Of Square.
So we thought we'd take the time to see how much control each of the shareholders at Square has.
As readers of Rotary Gallop know, Control is almost always very different from what percentage ownership implies. I calculated control among the shareholders at Square based on the ownership numbers in Thomas's article and thought you might be interested.
Two interesting points:
- Jack Dorsey retains much more significant control than one might have guessed. With 28% ownership he retains ~75%control and an ~87% chance of winning a shareholder vote.
- Also very interesting is that while Khosla owns nearly twice as many shares as McKelvey they have nearly identical control!
Caveats and Addendums: I don't have a copy of Squares articles of incorporation. There are two things to check to verify the accuracy of these calculations if available. One is that the line of passage is 50%. And the other is that each share gets only one vote. However my understanding is that in the vast majority of VC deals with major shops these things are true. VCs don't like to have different rules in every company they are in.
One thing that probably does apply to square is that VCs have preferred shares that vote without common for veto rights on certain issues. This however doesn't change the calculations, it just mean they apply only to the majority of things that do not require a special preferred vote.
At Rotary Gallop, we believe that measuring the right thing, with the right tool is REALLY important!
One of the biggest tragedies in the field of Investor Relations and Defense/Contested-Situation Advisory is one brought on by a terrible tool: Projections. Proxy solicitors popularized the name projections, but really all the experts we’ve talked to in the area of winning shareholder votes, (from successful activists, to Corporate Secretaries, heads of IR, IR consultants) are doing essentially the same thing. In fact, the problem described below is endemic to all analysis that involves starting at the top and working your way down a shareholder list (for convenience we’ll refer to all of this broader class as “projections”). Projections not only destroy useful information, but force people into measuring the wrong thing. These methods are wrong and destructive. So why is this terrible tool so wide spread? Because projections are intuitive and, in the absence of a better technology, were the only available choice. What is wrong with Projections?
Your expert has LOTS of valuable information, which they generally use to predict a result (e.g. Projection - 63% of the vote For Managment). Unfortunately, the expert feeds this valuable information into their traditional method of analysis - projections. The problem is projections are inherently incapable of handling the vast number of outcomes (defined as combinations in which your shareholders might vote) possible in a shareholder vote! Projections are essentially a pencil and paper method that, with the help of an excel sheet and a really ambitions practitioner, might be able to consider, say, 100 different outcomes.
To drive home how inadequate this really is, below on the left is the number of atoms in the known universe. On the right are the number of unique outcomes possible in an election of just 1000 shareholders! A mind-boggling number!
Atoms in the Known Universe: 100,000,000,000,000,000,000,000,000,000,
Possible Outcomes in a 1000 Shareholder Election: 1,000,000,000,000,000,000,000,000,000,000,
So, in order to reduce the number of outcomes to something manageable your expert is forced to start adding in shaky information, and eventually even making outright assumptions. If you mix good information with even a tiny amount of bad information, especially in a flawed methodology, the bad information wins every time. Thus the output of projections is garbage. “Garbage in – Garbage out” isn’t just a saying. It is an inarguable fact about our world, like the area of a triangle, resulting from the way errors propagate.
Garbage in - Garbage out: Projections force the use of shaky information, poisoning what solid intelligence is available and inevitably producing inaccurate results.
Your expert has spent a career building the expertise and connections that make up their incredibly valuable knowledge. It is precious. We should be protecting it from harm. Don’t let Projections ruin your experts most valuable assets!
At Rotary Gallop, because we can handle the vast number of outcomes possible in your next shareholder vote, we can pool only the most certain information from all your experts and calculate useful, relevant, actionable intelligence - such as your control, your chances of victory, the control of each shareholder, and how much a given shareholder could change your chances of victory by siding for you or against you.
Next time I’ll explain how in addition to destroying your experts precious information, projections measure the wrong thing! It will be obvious once you see our latest innovation: The Map! If you have any questions email me at T@RotaryGallop.com
I have a confession to make: unlike both of my business partners, I am not a natural-born negotiator. Like a lot of Americans, I have been conditioned by shopping in stores where the price displayed is the price paid. We agree at the outset or we don’t do business. Anything else is a rude and uncouth argument. I’ve witnessed my partners in action enough to know that that’s not the way the real world works. (On the lighter side I once witnessed, open-mouthed, one of my partners negotiate an $8 piece of cake for free in return for assurances that if they liked it they might buy it next time and even tell their friends about it!)
I knew this was an area I could really improve on, so I started read a lot of books about how to negotiate. For me, Roger Dawson’s “Secrets of Power Negotiating” stands head and shoulders above the rest. That book taught me to stop viewing negotiation as an argument, and to start seeing it as a fun and fascinating game of strategy.
So when the e-mails between John Cleghorn of Canadian Pacific Railway and Bill Ackman of Pershing Square Capital were released, I was extremely excited to see such high-stakes negotiation play out right before our eyes. I turned to Roger Dawson to help analyze the negotiating tactics between Cleghorn and Ackman.
I suggested to Roger, “between your negotiating expertise and my ability to measure the control of each side under each proposed scenario, we could really dissect and analyze this thing play-by-play and tell the whole story!”
He agreed but added, “the only thing that really stands out as interesting is that Bill Ackman seems extremely aggressive and confrontational. Do you know if this style of negotiation generally works for him, because it’s never worked for me, and I’ve never really seen it work well? Generally, this sort of tone shuts negotiations down.”
That got me thinking … I have never met Bill Ackman, but given his line of business and his success, I imagine he’s not a bad negotiator. Why would he make such a rookie mistake? What if he didn’t make a mistake?
This is the difference between an enthusiastic practitioner and a real master of his craft. I was lost in the trees, simply spotting glorious negotiating tactics flying this way and that, until Roger gently pulled me back – “Did you notice this forest?” The forest was quite fascinating! What I saw was that Ackman was walking a tight wire out of a subtle trap – the negotiations themselves.
Measuring (and gaining) control of a board of directors is difficult. Two seats are definitely better than one, but not exactly twice as good. It is better to replace a sitting member, than to pull up a new chair at the table, and better still to replace one of the opposition. It can all get very muddled. But when you can measure control directly, you can see the forest plain as day. Let me show you:
Exhibit 1 shows Managements’ (Blue) and Pershing’s (Red) control of the board of directors under various hypothetical scenarios. The first scenario is the current 15-member board of directors. Management has 77% control, meaning they hold the deciding vote in 77% of all possible voting outcomes on the board of directors .
As we move to the right on the horizontal axis we see the power structure if Ackman is given 1, 2, 3, 4, 5, or 6 seats added to the board.
The brick wall divides the scenarios likely to result from negotiation, from those that would almost certainly require a proxy battle to achieve. Under the possible negotiated outcomes, management retains significant control of the board. If Pershing were to accept 1 or even 2 board seats, they would have very little chance of causing change at Canadian Pacific.Exhibit 1: Canadian Pacific Railway and Pershing’s control of the Board of Directors
Exhibit 1: Bill Ackman and Pershing square have very little chance of causing change on the board of directors with a negotiated settlement of two or less seats. Such a settlement would likely result in another year for current strategies to play out, while Ackman must wait for the chance to gain additional seats next year. To have an effect on Canadian Pacific now Pershing needs to engage in and win a proxy battle.
Now we see that Ackman cannot win negotiating. The best he can hope for from negotiations is to be stuck on a board he cannot influence for a year while he waits to try again next year.
It’s a tricky situation. Ackman is trying to combat the seemingly cooperative and congenial offer of a seat for himself on the board of directors that both sides know comes with essentially zero power to accomplish anything. If this wasn’t clear at the start, Canadian Pacific’s decision to add two new handpicked members to the board made it loud and clear, before they even made the offer. Now we can understand Ackman’s strategy. He has to negotiate, without appearing unreasonable, and without coming off as too much of a bad guy. To do otherwise would alienate other shareholders. At the same time, he cannot be trapped in a negotiated settlement. Hence the aggressive email followed by the befuddled apology, on the advice of his team.
The pairing of personalities at Pershing Square allows them to naturally implement one of the most powerful and well-known negotiating techniques: good-guy/bad-guy. A more useful, if less pithy, description of good-guy/bad-guy might be sympathetic-and-flexible/emotionally-charged-and-firm
(Here is Roger on good guy/bad guy
).The email exchange is a fine example of this technique’s opening rounds, with Ackman’s team nicely set up in the good-guy role. Good-guy/bad-guy is a powerful negotiating gambit, and not easily countered when genuine, as is the case at Pershing Square
. Ackman was smart enough to surround himself with cooler heads, and to know that he needs to keep them there. I suspect the cooler heads at Pershing see the value in Ackman’s passion as well. We’ve all seen enough cop-drama to know that when the team gets broken up, they’re useless without each other.
Control measurement reveals several more fascinating insights to the Canadian Pacific vs. Pershing Square case, which we doubt even the players involved fully understand. Check out this article on how significant representation for Pershing Square will double the power of existing independent board members!
: For these calculations we have assumed that management controls 4 of the 15 votes on the board of directors. We have no inside knowledge and have never spoken to anyone on the board. However, we do believe this is a conservative assumption, given that 5 board members have been on the board for 10 years, one board member is the CEO, two board members joined the board in the same month as the CEO, and two more were added in response to Pershing Square.
Though the influence of ISS varies from company to company, the proxy voting advisory firm’s impact can be measured. (Originally published on Directorship.com)There has been much ink spilled, angst and teeth gnashing about the outsized influence that Institutional Shareholder Services (ISS) holds, much of it based on broad averages—such and such a percentage of institutions subscribe to ISS, for example, or the average company in the S&P 500 has X percent of it shareholder base subscribed to ISS.
So exactly how much attention should you pay to these numbers? The truth is that the influence of ISS varies drastically from company to company. The best part—this can be measured.
In calculating ISS’s influence at a target company I once asked a well-known proxy solicitor working the case with me, “Roughly what percent of ISS’s subscribers end up voting along with ISS’s recommendations?” After some thought, he said “Oh, about 50 percent.” There is no need to break out the spreadsheet here. In this case, despite boasting a subscriber base holding over 20 percent of outstanding shares, ISS had exactly 0 influence. You’d expect the same level of agreement with the phases of the moon.
That said, I’ve worked on other cases where ISS was the absolute key to victory. (By the way, having a big effective-shareholder like ISS, who does what it says, isn’t all bad when compared to many large shareholders who will tell you what you want to hear and then vote the other way.)
Knowing ISS’s influence on your elections isn’t hard and isn’t abstract. You just need two things: a little information about your shareholder base and an analysis of where the voting power lies. The former tells you who subscribes to and votes with ISS. The latter tells you how much influence those shares have over the election.
Here is an example from a company I worked on recently: 12 shareholders representing 16 percent of the total outstanding shares subscribed to ISS. Of those, only seven with nine percent of outstanding shares regularly voted with ISS’s recommendations. In this company, this bloc of ISS-swayed voters, had a 53 percent chance of winning an election and a five-percent chance of swinging the election. An ISS recommendation for management would increase management’s odds of victory by 14 percent, while a recommendation against would only decrease their winning odds by 3 percent.
There you have it. ISS has been weighed and has been measured.
It won’t be the same answer for your company but the key point is that there is an answer. ISS’s influence is not abstract and cannot be calibrated by “a gut feeling.” It is measurable. Once you know how much sway ISS has on your election, you can plan accordingly, stop worrying about what the talking heads are saying and put that chunk of bandwidth back into making the company better.
(Originally a Guest Post for The Activist Investor. )There is a perception among many that granting an activist significant board representation over and above what ownership might indicate grants unfair and detrimental control to the activist. Such an action is often derided as a stealth takeover, or a takeover with no premium.
However, granting an activist significant board representation, to balance the secured votes of management, actually transfers significant power to the unaligned board members already serving. Those unaligned board members (with significant ties to neither management nor the activist) are utterly familiar with the company and its problems, but have not had sufficient voice to cause change. It may be that the best effect an activist has, even one without a detailed, 30 point, 5-year plan, is to cause this shift in power towards these already serving board members. The unfolding drama between the current management of Canadian Pacific Railway and Bill Ackman of Pershing Square Capital Management, L.P. offers a prime example. Pershing is seeking 5 board seats on a 15 member board while owning only 14% of outstanding shares. To some, this may seem like an undue amount of representation. However, we find that far from concentrating power in the hands of Pershing Square, granting Pershing significant representation would actually result in a much more even distribution of power on the board of directors.
Since things are in the late negotiating stages, with management offering Pershing a single seat, we looked at what would happen to the distribution of power on Canadian Pacific Railway’s board of directors if Pershing Square were granted 1, 2, 3, 4, 5, or 6 additional seats. As Pershing is granted more seats the unaligned board members’ power increases significantly, as seen in the graph below. In fact, when Pershing Square is granted four seats, matching management’s secured votes, the power of the unaligned board members more than doubles to 25%!
Exhibit 1: As Bill Ackman and Pershing square receive more seats the unaligned board members’ control increases, more than doubling at 4 seats for Pershing. Notice that the unaligned board members’ power grows even as their percentage of the vote shrinks steadily. As we move to the right on the horizontal axis we see the percentage of the total vote (black) and Control (Green) of each unaligned board member if Bill Ackman is given 0, 1, 2, 3, 4, 5, or 6 seats added to the board.
It can, and perhaps ought to, be argued that this spreading of power on the board is exactly what one would hope for at a time when a new direction is necessary. Such a leveling effect will allow new ideas to more effectively compete, whether they come from Pershing or the unaligned board members familiar with Canadian Pacific’s problems.
We wonder how much of activists’ proven effectiveness at delivering returns stems from this simple redistribution of power to previously disenfranchised, but established board members. Let us know what you think in the comments below!
 For these calculations we have assumed that management controls 4 of the 15 votes on the board of directors. We have no inside knowledge and have never spoken to anyone on the board. However, we do believe this is a conservative assumption, given that 5 board members have been on the board for 10 years, one board member is the CEO, two board members joined the board in the same month as the CEO, and two board members were appointed in the face of Pershing’s activism.