Opinion

Felix Salmon

How Pimco works

Felix Salmon
Jul 31, 2012 17:08 UTC

There’s an anonymous troll on the internet who doesn’t like my latest Pimco post. And frankly it’s really hard to take any post seriously when it’s tagged “born last night, clown questions, gmafb, horseshit, STFU”. This kind of macho bullying posturing is everything I hate about Wall Street — a place which is still home to far too many overconfident frat boys with overstuffed paychecks.

So, why am I rising to the bait? Mainly because some people I respect are taking the post seriously. And also because, hidden behind the sophomoric grandstanding, there are actually a couple of substantive points being made.

To take them in order, then:

Firstly, does Bill Gross pay himself, or is he “paid by the parent company that bought his firm”? I haven’t seen a lot of reporting on this, but everything I know about Pimco says that it’s a very arm’s-length, largely independent unit of Allianz. It certainly dividends profits up to its parent, but I don’t actually believe nor have I ever seen it reported that Allianz executives make granular decisions on how much Bill Gross, or any other Pimco employee, gets paid on a year-to-year basis.

Is there a formula governing Gross’s remuneration, based on some combination of Pimco revenues, Pimco profits, and the performance of the funds he manages? I’m sure there is. And if you want to reverse-engineer a way for Gross to have been paid $200 million in 2011 despite massively underperforming that year, then that’s surely the way to get there. Pimco doesn’t want to encourage short-term gambling among its employees, and so its pay is based on long-term performance rather than year-to-year fluctuations; Gross’s long-term performance remains excellent, and he manages an astonishing amount of money. And on top of that, Pimco is attracting spectacular inflows these days.

Still, Pimco told me that the numbers in the original NYT article were “seriously inaccurate”, and I’m quite sure that Gross, given his position in the company, does have a certain amount of discretion when it comes to divvying up the remuneration pool. He might not “have to answer to congress or a goofball parade of Occupy Wall Streeters”, but he’s still a leader — and even if we don’t know for sure how much he got paid last year, a lot of big-time money managers in the company know exactly what example he is setting. If they would risk getting fired after turning in such dismal performance, then it would be downright hypocritical — and bad for the cohesion of the senior management team — were Gross to accept a $200 million paycheck in such a bad year.

And how about the people whose money Pimco is managing? Yes, it’s easy to say that they’re sophisticated investors who “pay an agreed upon and transparent management fee up front” — but that doesn’t mean they’re happy with the fees they’re paying, especially not if they start reading about $200 million paychecks. And in a world moving swiftly away from the fund model and toward the lower-fee ETF model, it behooves any long-only money manager to keep a very close eye on fees and costs. The level of money-skimming which maximizes your payday this year is not necessarily the best way to keep on building your company’s franchise over the long term, especially in a world where index investing is becoming increasingly popular.

As for the assertion that long-only “buy siders that actually run portfolios north of 200 billion are paid at this level” — well, name some names. It’s a very short list, of course. But if you can find one or two other people who were paid $200 million a year for managing funds, and who weren’t hedge-fund managers collecting 2-and-20, then I’d be much more likely to believe that Gross is paid that much, too.

Next up comes a question about Mohamed El-Erian’s tenure at Harvard Management Company. I quoted an article about how “Mohamed was having a heart attack” while he was there, because Larry Summers insisted on taking Harvard’s spare cash and investing it in an endowment which was designed to have a virtually infinite time horizon. As a result, El-Erian’s job when it came to liquidity management was made extremely difficult. But now I’m told “this isn’t true”, on the grounds that all El-Erian needed to do was “explain” to Summers and others “that their allocation was inappropriate”, and then sleep well at night since the “allocation was made by Harvard officials not by Harvard Management.”

Maybe anonymous Wall Street trolls think that way, and wouldn’t worry about Harvard’s liquidity needs even if Harvard was effectively using them as a checking account. But a responsible money manager worries about liquidity every day, especially in a situation where Harvard can and will ask for large sums of cash on a regular basis. In any case, my larger point was that El-Erian can’t be blamed for liquidity problems after he left HMC, and there doesn’t seem to be any disagreement on that front.

Then there’s the question of the degree to which El-Erian’s ubiquity in the media is a Pimco marketing strategy, responsible for the large increase in assets that Pimco is seeing these days. I’m informed that the answer is a simple yes — but if that’s the case, that has interesting implications. A large chunk of Fabrikant’s article was based on the premise that Pimco’s investors wanted Gross’s bond-trading expertise, rather than El-Erian’s technocratic global-macro insights. But if indeed El-Erian’s regular TV appearances and various op-eds are responsible for the hundreds of billions of dollars which continue to flow into Pimco, then it seems that there’s a lot of appetite out there for a macro-led, rather than a trading-led, strategy.

On top of that, it’s notable that Gross, the great bond trader, has started to underperform Pimco as a whole, where investments are based very much on the global macroeconomic outlook. Pimco’s more than big enough for both Gross and El-Erian, of course. But the idea, in Fabrikant’s piece, that Pimco is effectively still Gross’s shop, and risks withering away were he ever to leave — that idea is pretty effectively demolished if in fact El-Erian’s media strategy is responsible for bringing in enormous amounts of new money. Certainly El-Erian never talks about trading strategies in such appearances.

Finally, there’s the question of Blackrock, a much bigger fund manager than Pimco, where, incidentally, the CEO, Larry Fink, was paid $21 million in 2011. How did Blackrock grow so big? In large part by buying a lot of index funds, thereby diversifying into one of the fastest-growing investment strategies in the world. And also, in part, by being a public company. And so I asked a question, and received an answer:

In order for Pimco to effectively compete with Blackrock, will it too have to go public?

No. How is that even a question? They are a wholly owned subsidiary of a firm that is significantly larger than Blackrock which allows them tremendously cheap financing if they need it. Allianz’s insurance assets also provides them with 23% of their AUM. Does JP Morgan Asset Management, SSgA, or Deutsche Bank Asset Management (all well over a trillion in AUM) need to spin off and IPO to compete with Blackrock?

I wasn’t suggesting that Pimco spin off from Allianz. But Pimco already has “shadow equity” which is traded among Pimco employees; there’s no reason that it couldn’t get listed as some kind of tracking stock. And that tracking stock could be a very valuable acquisition currency as Pimco seeks to diversify away from its historical core competence of actively-managed bond funds. There are many reasons why Pimco might well prefer to do things that way, rather than asking Allianz for “tremendously cheap financing” for an acquisition.

I’m sure that Pimco gets lots of value from having Allianz assets at its core. But Pimco is also reported to be “seeking more independence from its parent”, and in any case I don’t think it’s true that Pimco is wholly owned by Allianz, which bought only 70% of the company back in 1999.

My point about Blackrock is that by having its own stock and being master of its own strategy, it has managed to diversify, and grow, more quickly and effectively than Pimco has. Here’s a germane quote, from last year:

“The history of the asset-management business demonstrates time and time again that the most successful asset-management firms are those who are dedicated to investing rather than subsidiaries of banks and insurance companies where there can be lots of tension,” Burton Greenwald, a fund-consultant based in Philadelphia, said in an interview. “Fund companies tend to be entrepreneurial, while banks and insurance companies tend to be bureaucratic.”

There’s a case to be made that Pimco has in fact thrived under Allianz’s ownership — but it’s unclear whether that’s a function of Allianz being a great owner, or whether it’s a function of the fact that those years saw the greatest fixed-income bull market of all time. That bull market is going to come to an end at some point. And when it does, Pimco wants to be positioned much more evenly across various different asset classes and strategies than it is now. In order to do that, it’s not a completely horseshit clown question to ask whether it might want to take a leaf or two out of Blackrock’s book.

Update: David Merkel adds some very useful facts to the debate.

COMMENT

I think people focus on the “amount” that Bill Gross gets paid rather than the value that he brings to PIMCO / Allianz or Bill’s opportunity cost (how much he could be making elsewhere). By these measures, I calculate he significantly is underpaid at $200M or less per year. While I made a lot of big assumptions, my back of the napkin math suggests that Bill Gross should be getting paid in range of $700 – $1 billion per year. The math is described in the article:

http://www.learnbonds.com/bill-gross-com pensation/

Also, we have a poll – how much do you think Bill Gross should be paid?

http://www.learnbonds.com/bill-gross-sal ary/

Posted by LearnBonds | Report as abusive

How SecondMarket works

Felix Salmon
Apr 12, 2011 20:23 UTC

I spent most of this morning at SecondMarket, having a long conversation with Adam Oliveri, the person in charge of their private company market. That’s the part of the company which gets the most attention: it’s where stock in companies like Twitter and Facebook change hands, for instance. I learned a huge amount while I was there, and have now changed my mind on whether Facebook is going to go public: I finally understand exactly why companies need to do an IPO once they have more than 500 shareholders.

Once the 500-shareholder limit is breached, companies have to start reporting detailed financial information to the SEC. Which isn’t in and of itself a compelling reason to go public — lots of private companies with public debt file that information, after all. But there’s something else which gets triggered when you have more than 500 shareholders: you have to register your equity securities with the SEC. And at that point, your shares can be traded by anybody at all in the public over-the-counter markets, even if you haven’t had an IPO.

It’s conceivable that companies could continue to encrust their shares with various contractual restrictions which prevented shareholders from trading their shares in the OTC markets, even after those shares were formally registered with the SEC. But in practice, it’s almost impossible for companies to prevent OTC trading in their publicly-registered securities. And when a stock trades in the OTC markets, that trade is registered and printed in public. At that point, with a company’s stock being traded by anyone at all, at any time, at a public price, on the basis of public information filed with the SEC, the company is to all intents and purposes public already. So it might as well just make it official by having an IPO.

Now that Facebook has said that it passed the 500-shareholder limit this year, then, it’s pretty certain to go public in 2012. (Kara Swisher thinks it might be even earlier than that: I have a bet with her that it won’t happen before September 21 of this year. If it does, I need to go to San Francsico and buy her dinner, but if it’s still private at this point she needs to come to New York and take me out.)

SecondMarket actually has two platforms for trading private-company stock. The main one is Adam’s private company market, which is about two years old at this point, and has seen equity in 50 different growth stocks change hands. It’s pretty much restricted to growth stocks: none of those 50 companies has ever paid a dividend, and they’re overwhelmingly in the technology space.

For other companies, SecondMarket has set up a much more nascent market, which kicked off in January with those trades in Pimco stock. It’s also designed to help trade stock in partnerships (like McKinsey, say), or maybe even large, established private companies like Mars or Cargill. But mostly it seems that SecondMarket has its eyes on companies like Pimco which are subsidiaries of larger companies but which still use their own equity as a recruitment and compensation tool. Reddit is one company which might try to price stock on SecondMarket, as a way of helping it attract talent and grow while still remaining a part of Conde Nast.

Adam also helped answer my question of why SecondMarket is taking off now. Look at the three companies which really got this market started: Facebook, LinkedIn, and eHarmony. They’re all highly visible companies, with metrics that can be measured externally with quite a lot of specificity by companies like ITG Investment Research. It’s also much easier these days to find such companies’ articles of incorporation and the like online — and of course huge amounts of information about these firms is published by the fast-growing blogosphere. So while the amount of information that would-be investors have is surely lower than if there was a formal SEC-registered prospectus, the rise of the internet has made it much easier to do reasonably good diligence on how much a company might be worth. And that’s especially true when the company is young enough that its revenues don’t matter very much.

On top of that, webby companies like these are generally pretty capital-efficient: there’s very little risk that existing shareholders will be unpleasantly diluted by some big upcoming capital-raising round. It’s no coincidence that SecondMarket hasn’t seen trading in green-tech or biotech startups, which are much more capital-intensive.

And then there’s the big picture, which is simply that we’re seeing fewer IPOs of small companies, and that most companies when they do IPO are more like 8-10 years old rather than 3-4 years old. At that point, you’re likely to have had a reasonable amount of turnover in terms of employees, and early employees who have long since left the company are reasonably going to want a way to cash in their equity stakes. That demand for liquidity — along with long-term employees who have a lot of paper wealth but still live relatively frugally and who would like to monetize some of their stake — is what helped get SecondMarket’s equity business started.

Letting employees sell some of their vested stock doesn’t disalign incentives — quite the opposite, in many cases. After all, venture-capital owners of fast-growing tech startups are looking for high-risk home-runs and have diversified portfolios. Employees, by contrast, are always going to be more risk-averse, and letting them cash out in the growth phase can give them enough money to be willing to take the kind of risks their VC paymasters want to see.

It’s also worth clearing up some of the misconceptions in Dennis Berman’s column today on SecondMarket and SharesPost. For instance:

Many in Silicon Valley and Washington regard SharesPost and rival SecondMarket as small saviors of American capitalism. These markets give young companies and their employees new ways to raise capital or sell private stock without the arduous financial and legal disclosure of fully public companies.

This is partly true, but I’m pretty sure that neither SharesPost nor SecondMarket has ever let a company raise capital using their platform. I asked SecondMarket about this today, and in principle they’re open to exploring the idea in future, but for the time being they’re concentrating on simple transfers of shares, rather than the capital-raising issuance of new equity.

Berman continues:

SEC boss Mary Schapiro seems conflicted about these new markets’ purpose. The agency is investigating potential abuses in these secondary markets, including conflicts of interest and insider trading.

There’s no hyperlink here, so I have no idea what Berman thinks he’s talking about. It’s conceivable, I suppose, that he has an SEC source feeding him secret information about an internal SEC investigation that nobody else knows about. But if he did, one imagines he’d write a news story about that, rather than mentioning it in passing in a column which leads with the death of his grandmother 20 years ago. Certainly I’ve seen nothing to indicate that the SEC is investigating SharesPost or SecondMarket for potential abuses including insider trading; this seems to me to be both inflammatory and false.

After quoting Ben Horowitz as someone who is skeptical about such markets (but not mentioning that Horowitz spent $80 million buying shares of Twitter on SecondMarket in the secondary market), Berman comes out with this:

For a market to work best, investors need to be comfortable that they can trade at will.

This manages to completely miss the point of SecondMarket and SharesPost. They’re emphatically not trading vehicles: they’re designed to facilitate one-off transactions. In the two-year history of SecondMarket’s private-companies market, the company has seen maybe half a dozen instances of what you might call tertiary trades: someone who bought at one point and then sold later, once the price had gone up. SecondMarket gives an opportunity to invest in private equity, and private equity by its nature is illiquid. In fact, that’s why many investors like it: they want to capture the illiquidity premium, happy holding on to their stake for many years and knowing that they have an asset which isn’t highly correlated with public markets.

Going forwards, of course, SecondMarket would love it if the 500-shareholder restriction was relaxed. When the rule was introduced in 1964 it was pretty arbitrary, but it was set at a level which wasn’t particularly onerous: the 500-shareholder limit was very rarely triggered before a company went public. After all, in those days you could go public when you were still small; today, that’s much harder. Today, the 500-shareholder limit is a real constraint on how companies do business, how they compensate their employees, and how they structure themselves internally. Is there any good fundamental reason to change the way you incentivize and compensate employees just because you’re hiring lots of people? Of course not — but that’s the effect the rule has.

So while I worry about the public-policy effects of having fewer public companies, I also see no reason for the SEC to keep this rule at its anachronistic 1964 level. On the other hand, I think it might make sense for the SEC to regulate SharesPost and SecondMarket more explicitly than it does at present, rather than having them operate in the shadow of exemptions which were written long before they were founded. If the SEC set clear rules for how private exchanges like this could operate, then that might open the way to bring the rules for companies listing on public exchanges into the 21st Century.

Update: SecondMarket’s Mark Murphy emails to say that Horowitz’s secondary-market acquisition of Twitter shares did not take place through SecondMarket.

COMMENT

Maybe inflammatory, but no longer false. http://www.sec.gov/news/press/2012/2012- 43.htm

Posted by Setty | Report as abusive

When insiders trade

Felix Salmon
Apr 4, 2011 03:35 UTC

On Thursday, Pimco CEO Mohamed El-Erian told an audience at Thomson Reuters how he was buying “shadow equity” in Pimco on SecondMarket. On Friday, John Carney put up an intriguing post alleging that deals done on SecondMarket are subject to insider-trading laws. If he’s right, it seems to me that almost anybody buying or selling shares of a private company on SecondMarket would be breaking the law, at least if that company puts out little or no public information about itself. And then today, on Sunday, Zero Hedge picked up on a passing comment in David Sokol’s CNBC interview, where Sokol said that Charlie Munger owned 3% of Chinese auto and battery maker BYD before recommending the stock to Berkshire Hathaway.

We can certainly assume that El-Erian knows material information about Pimco: he is the CEO, after all. And that such information is non-public. Should that then bar him from buying Pimco stock through SecondMarket? And as for Munger, assuming that Sokol’s allegations are true, first read Nocera on Sokol:

How is this not, on its face, evidence of insider trading? A guy buys stock in a company and then talks his boss into buying the company. The fact that his boss is Warren Buffett makes it even more “material,” to use the word the S.E.C. favors when it investigates insider trading. If a company executive trades on material information, knowing that he is privy to stock-moving news that hasn’t yet been divulged to other shareholders, he is likely to be committing a crime. When Warren Buffett buys a company, the stock price goes up. Everybody knows that.

If you apply this logic to Munger, with his monster personal 3% stake in BYD, he looks significantly guiltier than Sokol. Maybe this is why Buffett went so easy on Sokol: he didn’t want to set a precedent which would drag down Munger.

The SEC, when it goes after insider-traders, nearly always looks for trades in which third parties — people like Raj Rajaratnam — trade on information they’ve carefully cultivated from insiders, often taking great care to ensure the utmost secrecy of what they’re doing. But there’s a case to be made that much if not most insider trading is much more overt, and simply never punished. In this sense, pretty much every stock trade by any CEO is a trade done with knowledge of material non-public information. Those trades might well end up being disclosed, but that doesn’t make them any less insidery.

I’m not for a minute saying that all these trades should be prosecuted, although a zealous prosecutor wanting to make a name for himself might well feel like giving it a try. I do think though that people like David Sokol and Charlie Munger should ask themselves why they’re trading stock for their personal accounts in the first place. Is any profit they make ever going to be worth the possible downside in terms of public embarrassment?

COMMENT

KenG: you wrote “the insider information he did have access to was that Lubrizol was trying to sell itself.”

no – that’s not true – he didn’t have that information when he bought the stock. LZ came up on a screen generated by Citi of potential companies Sokol might be interested in – not companies that were out soliciting bids.

and Najdorf: you wrote “If someone at Goldman Sachs did what Sokol did, you would all be tearing into him as an obvious criminal and very bad man.”

yes, and if the bankers at Citi had done what Sokol did, we’d also be tearing in to them. The same information can be insider trading if one party uses it (the bankers) but not if another party does (Sokol – let me explain:)

see, your next statement highlights the tricky part: “Sokol had the information because he was an insider.”

no – Sokol had the information because “the information” was the thoughts inside his own brain. He decided that LZ was an interesting company. THAT is the information. His own actions were the information – he can’t “un”-know them.

Posted by KidDynamite | Report as abusive

Why are private markets booming now?

Felix Salmon
Apr 1, 2011 00:37 UTC

I’m spending Friday at the Kauffman Foundation’s Economics Bloggers Forum in Kansas City. A dozen bloggers are giving short presentations, and I’m flattered to be one of them, along with the likes of Tyler Cowen, Bryan Caplan, Ryan Avent, Dean Baker, Steve Waldman, and Virginia Postrel. We’ve all been encouraged to write a post about our talk.

Following on from a question I asked Mohamed El-Erian at the Thomson Reuters Newsmaker event today, I’m planning to talk a bit about a favorite subject of mine, private stock markets like SecondMarket and SharesPost. They’re hot these days, and one of the questions I’m hoping to put to the bloggers at the Forum is why that might be.

Certainly it’s easy to see why these private markets serve a useful function. Take Pimco: it gives out shadow equity to a lot of employees, which vests over a period of years. But once they have that equity, what are they meant to do with it? Pimco can unilaterally announce a price for it, and say that it will buy back its own stock at that price. But that seems a bit unfair: one of the reasons that many banking partnerships went public was that they were buying back their stock at book value, even as the public markets were valuing banks at significant multiples of that. And in general, if there’s only one buyer for an asset, that buyer always has an incentive to lowball the price.

That’s why Pimco turned to SecondMarket, to take advantage of their web-based Dutch auction system. Pimco employees entered bids and offers into the system — the amount they were willing to sell their shares for, and the amount of money they were willing to pay for new shares. The bids and offers then cleared at a certain price (I didn’t get the opportunity to ask what the price was, and I’m sure that El-Erian wouldn’t have told me if I had), and that price by its nature is a fairer value for Pimco stock than anything determined according to a formula.

El-Erian said, and I believe him, that the auction didn’t see a lot of volume: there simply weren’t all that many people interested in selling their equity. It’s easy to see why: Pimco is doing very well these days, and it pays well too. Some employees might have difficulty paying their tax bill as their equity grants vest, but most are rich enough that they can do so quite easily. Most of them, too, probably reckon that employees-only equity in Pimco is likely to be a better investment over the medium term than anything they might be able to do in the current market with the proceeds of any sale. That’s the reason that shares of SecondMarket itself have yet to trade on SecondMarket: there aren’t any sellers, only buyers.

In the specific case of Pimco, there are always rumors that it might get sold or spun off from its parent, Allianz. A strategic buyer might well be prepared to pay a substantial control premium to get its hands on more than $1.2 trillion of assets, and Pimco’s employees would surely like to collect that premium themselves rather than sell the right to receive it to someone else.

So as is quite common in these situations, buyers — including El-Erian himself — outnumbered sellers. Which raises a common objection to private markets: that they don’t allow for short selling, which is an important contributor to the way in which public markets perform their role of price discovery.

There’s also my public-policy objection: that even if these markets make sense for the companies in question, they’re bad for the general public, which gets shut out of the opportunity to own equity in strong, attractive companies. Pimco’s not a great example on this front, since it’s possible to buy shares in Allianz. But if you look at the poster child for private markets, Facebook, the only people able to buy equity are a small group of elite and well-connected global rich people. And that doesn’t seem fair.

Historically, workers have been able to get access to the class of investments available to the rich: they have had defined-benefit pensions, run by pension plans which qualify as rich and sophisticated investors and which therefore have a large universe of asset classes to invest in. But as we move from defined-benefit to defined-contribution plans, that universe shrinks.

It’s quite easy to see why companies like Facebook might not want to go public — although the NYT is now reporting that an IPO might happen early next year. Being public is expensive and annoying: CEOs almost always prefer to be public. Public shareholders are litigious; public stocks tend to move in lockstep with each other rather than due to company-specific fundamentals; and the minute that your stock price is public, everybody pays an enormous amount of attention to it and judges you by it. Given the amount of money in private markets, what upside is there in going public? Especially when SecondMarket and its ilk allow shareholders to liquidate their holdings quietly and at a good price.

What’s less easy to see is why this trend is happening now. Is there something special about the USA circa 2011 that makes private exchanges particularly timely? Yes, there’s a lot of liquidity sloshing around — but that was true in the mid-oughts, too. Yes, Sarbox has made life for public companies that much harder — but Sarbox has been around for a while as well. Maybe it’s simply the fact that SecondMarket and SharesPost have come along and executed well in a space which no one bothered to put much effort into before. I’m reminded of the current rise in couponing: there’s absolutely no reason why Groupon, LivingSocial, and the like should all be exploding now, rather than during say the first dot-com boom of the 1990s.

Markets aren’t that efficient: sometimes it takes a while for an entrepreneur to spot a big gap in the market. But if there is a good reason why private markets are booming in the USA today, I’m sure the bloggers assembled at Kauffman will be able to think of it.

COMMENT

Here’s an additional thought…

How many times a year does a company release significant news? News that has the potential to increase the value of the company by 1% or more?

* Quarterly earnings reports.
* Mid-quarter outlook releases.
* Occasional product-specific events.

There are ~250 trading days each year, and only a dozen or so significant events. EVERYTHING ELSE is macro-driven noise, which by its nature affects the entire market (or at least entire sector) similarly.

So yes, the minute-by-minute correlations are very strong. In the absence of news they SHOULD be strong. But it is the newsworthy events that drive returns for investors (as opposed to traders).

You can obsess about the noise or you can simply look past it. Absent HFT algorithms, the latter is the better approach.

Posted by TFF | Report as abusive
  •