Felix Salmon

Sell-side snafu of the day, JP Morgan edition

Felix Salmon
Jul 16, 2010 06:21 UTC


This is the stock chart for ATP Oil and Gas over the past five days: it plunged at the open on Tuesday and then soared at the close today. And it’s all because of Joseph Allman, the analyst for ATP at JP Morgan

ATP stock closed at $10.43 on Monday and then dropped to $9.12 on Tuesday morning — a fall of more than 12.5% — when JP Morgan put out a very bearish research note with the title “Trading at a Premium; Financing Needs Likely Greater Than Market Thinks; Downgrading to Underweight.”

The headline was bad enough, including as it did a downgrade on the stock, but the meat of the note was worse: Allman calculated that ATPG would need $500 million of external capital — more than its entire market capitalization.


At the end of the trading day today, Allman put out a second research note, saying that the first note was wrong:

In our July 13 note, we stated that it appeared that ATPG would need $500MM of external capital. This model corrects that error and reduces that need to $50MM.

Yep, JP Morgan was out by an entire order of magnitude in the calculations which prompted the downgrade. And when it admitted its mistake today, the stock went straight back up to $10.12, a rise of 11.1% on the day and of a whopping 14.4% from the day’s low.

When JPM’s mea culpa appeared today, ATP’s stock immediately shot back up — but of course by that point it was too late for anybody who had been stopped out on the way down.

And if you were lucky enough to have any kind of early access to JP Morgan’s research product, you could have made a fortune first selling ATPG on the way down and then buying it back on the way up.

There are three lessons to this story, I think. Firstly, small stocks are volatile. Secondly, paying too much attention to sell-side analysis can be hazardous to your wealth. And thirdly, prices really do move on sell-side research. Even when it’s utterly wrong-headed.


^what TFF just said. Sell-side analysis, perhaps only slightly more useful than a 2006-vintage MBS rated AAA by Moodys / S&P.

Posted by McGriffen | Report as abusive

Congress approves Basel III before it even exists

Felix Salmon
Jul 15, 2010 23:14 UTC

Assistant Treasury secretary Mike Barr celebrated the passage of the financial reform bill today by phoning me up for a chat. So of course I asked him how much has really been achieved, and how much the really important stuff has been left to the Basel III negotiation process.

Barr said that the reforms passed today “were absolutely essential to the process,” and added something I didn’t know before — which is that they include Congressional authority for regulators to adopt all the Basel III standards. In other words, there’s no risk of Basel III getting caught up in Congressional opposition, as Basel II did. Once it’s agreed in Switzerland, US regulators are free to implement it immediately. “We got all the authority that we needed in this legislation that just passed,” Barr said. “The regulatory community will be ready to implement it in the US.”

As for timing, Barr was still hopeful that Basel III will be done this year; I’ll believe it when I see it. Then the various ratios will be phased in over many years. In terms of the bill which has now passed, different bits of it take effect at different times: resolution authority, for instance, is effective immediately, while the rules on interchange fees take effect in 9 months, and the new Volcker Rule — which has yet to even be defined, and which will surely the the subject of much financial-industry lobbying — becomes law in 18 months.

What I didn’t ask Barr about, sadly, was who he’d like to see head the Consumer Financial Protection Bureau. Shahien Nasiripour says, plausibly enough, that Tim Geithner is opposed to tapping Elizabeth Warren for the job, despite the fact that she’s the obvious choice. I hope he doesn’t get his way. The bureau would never have come into being without Warren pushing it hard; it’s only fair she gets a chance to run it at inception, and shape the way it does business. Even if she has been harsh in her public questioning of Geithner.

Update: Turns out that Geithner does support Warren for the bureau after all. Treasury just sent me this statement:

“Given her strong leadership on consumer protection, Secretary Geithner believes that Elizabeth Warren is exceptionally well qualified to lead the new bureau, and, ultimately, that’s a decision the President will have to make.” – Andrew Williams, Deputy Assistant Secretary for Public Affairs


I think there’s an implicit assumption in Barr’s comments – namely that Basel III will meet the minimum standards (i.e., the Collins amendment) and comply with the other prescriptions (i.e., countercyclical capital charges for systemically risky activities) in the bill.

Should Basel come in weaker than expected, it seems that US regulators would indeed have to reconcile with Dodd-Frank…

Posted by DJT85 | Report as abusive

Goldman agrees to carry on as usual

Felix Salmon
Jul 15, 2010 21:42 UTC

My favorite part of the SEC settlement with Goldman Sachs is the bit where Goldman agrees to “a permanent injunction from violations of Section 17(a) of the Securities Act of 1933″. Well, that’s reassuring, knowing that from now on Goldman has promised not to break the law. Goldman has also consented to an agreement that when it puts together new mortgage securities, it’ll run any prospectuses or term sheets by its legal or compliance departments. As if it wasn’t doing that already. And there’s lots more like that: people on the mortgage desk have to attend training seminars on disclosure! Goldman “shall provide for appropriate record keeping”! And so on and so forth.

Meanwhile, the closest thing to an admission of wrongdoing coming from Goldman is this:

It was a mistake for the Goldman marketing materials to state that the reference portfolio was “selected by” ACA Management LLC without disclosing the role of Paulson & Co. Inc in the portfolio selection process and that Paulson’s economic interests were adverse to CDO investors. Goldman regrets that the marketing materials did not contain that disclosure.

This doesn’t even rise to the level of an apology: it’s just a “regret”. Meanwhile, Goldman will be more than happy to wire $150 million to IKB and $100 million to RBS, as detailed in the proposed judgment. (Technically the judgment is contingent on being approved by the New York judge, but that’s just a formality.) As a way of getting out from under this suit, it would be cheap at twice the price.

The only bit of possible embarrassment still left for Goldman is the litigation against Fabrice Tourre, which is ongoing. I find that peculiar, since Tourre was pretty junior, and the people responsible for him behaving correctly all seem to be home safe. I’m not sure what the SEC thinks it might achieve with continued litigation against Tourre, or whether Goldman now feels that it can safely hang the poor chap out to dry. But this story has always been about Goldman’s actions, not Tourre’s. And it would be unfair, to say the least, if Tourre bore the brunt of the punishment, while Goldman gets away with little more than writing a check which barely makes a dent in its vast cash pile.


Don’t know the answer to that one, Danny_Black, but we’re running something like 1 million foreclosures per year. At last check, 10% of prime loans were delinquent and 3% were in foreclosure. I imagine the situation would be much worse if you looked at loans written between 2004-2007, since mature loans are less likely to default.

Recovery has been poor, partly because so many of the loans are heavily underwater, partly because they are dragging out the process in the hope things will improve.

Posted by TFF | Report as abusive

Goldman’s win

Felix Salmon
Jul 15, 2010 20:35 UTC

Goldman Sachs has come to a settlement with the SEC: it will pay $300 million in fines, and another $250 million in restitution, according to the NYT. There’s no indication that any Goldman executives are being forced out: this is a purely financial settlement, and does not even include Goldman admitting any wrongdoing.

This is surely a massive win for Goldman, whose entire business was at stake if it was found guilty of serious wrongdoing. The company’s shares are soaring in after-market trade, and although they won’t approach their pre-case levels any time soon, Goldman can now begin to distance itself from Abacus noise, and try to put the whole sordid tale behind it.

In financial markets, memories are short, and the effects of this case and its settlement will fade away quite quickly. Clients will probably never trust Goldman as much as they did before the crisis, but that was true even before the SEC brought its case. And Goldman is putting a lot of effort into becoming much more transparent on the conflicts-and-ethics front:

A “business standards committee” is in the middle of an internal review that will examine everything from Goldman’s management of conflicts of interest to product suitability for clients. A quarter of the firm’s 400 partners will contribute to the report, which will be handed to a board committee in December. The findings will be made public.

This settlement is surely testament to the extraordinary powers of persuasion which still exist within Goldman Sachs, but some kind of settlement was always likely: the SEC didn’t want to risk bringing a complex case like this in front of an inherently-unpredictable jury. I’m just surprised that they didn’t even get any management changes, or any kind of mea culpa. The risk, of course, is that Goldman’s victory here will only serve to exacerbate its arrogance. Could the Squids of West Street become even more insufferable, now?


When Phil Gramm and other agents of satan went about repealing Glass-Steagall and de-fanging CFTC they were essentially – no, deliberately, decriminalizing a process that would lead to mass impoverishment on an unprecedented scale. It is now but the flimsiest of clausal deniability of wrongdoing that stands between Goldman Sachs and a firing squad.

If Congress took the same “liberal” approach to decriminalizing hard narcotics, at least Americans could afford the mind-numbing substances it would take to be completely chill with the perpetrators and their unctuous apologists.

Posted by HBC | Report as abusive

Can behavioral economics cause real harm?

Felix Salmon
Jul 15, 2010 19:35 UTC

George Loewenstein and Peter Ubel are about as expert on behavioral economics as it gets, so it’s interesting that they’ve taken to the op-ed page of the NYT today to urge politicians to spend less time on nudges and more time making substantive policy changes.

The problem here is that although behavioral economics can result in policies with positive effects, it can also mean that those policies get put into place instead of ones which really have teeth.

Prime Minister David Cameron of Britain recently promoted behavioral economics as a remedy for his country’s over-use of electricity, citing what he claimed were remarkable results from a study that reduced household electricity use by informing consumers of how their use compared to that of their neighbors.

Under closer scrutiny, however, tests of the program found that better information reduced energy use by a mere 1 percent to 2.5 percent — modest relative to the hopes being pinned on it.

Compare that with the likely results of a solution rooted in traditional economics: a carbon tax would instantly bring the price of energy into line with its true cost and would unleash the creative power of the marketplace to generate cleaner energy sources.

Behavioral economics should complement, not substitute for, more substantive economic interventions.

You can quibble with the specific example here — Barbara Kiviat notes that other studies show greater effects — but conceptually it’s easy to see that any behavioral-economics solution carries with it a potential problem, which interestingly enough might be exactly the kind of thing best studied by behavioral economists.

Consider an issue with two possible lines of attack: a cheap behavioral-economics solution, B, and a more expensive and politically-fraught substantive solution, S. Does implementing B make implementing S less likely? If B didn’t exist, would S be more likely to come about? Surely there are cases where the answer to both questions is yes — and where therefore behavioral economics is a bad thing, not a good thing. The ability to cover up issues with a behavioral band-aid is often just a way of doing as little as possible while appearing to tackle the issue at hand.

That said, in a lot of cases S would never happen anyway, and in those cases B is better than nothing. And in other cases S will happen either way, and again adding B to the mix is going to be a good thing. So the only cases we’re worried about are the ones where the existence of B significantly changes the likelihood of implementing S. I wonder how common that is.


I think this analysis is a little shallower than it seems. Most problems can be addressed in a variety of ways, and politicians will tend to follow the path of least resistance.
For example, in your formulation you could let B equal tax incentives and you’d get the same result, as tax incentives are not as powerful as more sweeping measures.

Posted by RZ0 | Report as abusive

The big problem with Europe’s stress tests

Felix Salmon
Jul 15, 2010 17:02 UTC

Can Europe’s bank stress tests do for banks over there what the US tests did on this side of the pond? Mohamed El-Erian is not optimistic:

Last year’s stress test in the US applied to institutions that were the main cause of the financial instabilities, and the government had budgetary room to support the sector. Europe’s situation is different. The concern about banks is a derived concern, reflecting worries about sovereign debt in some countries and the overall economic situation; and there are greater limits today on budgetary resources.

In other words, if bank solvency is the problem, then the government rescuing the banks — or forcing them to recapitalize — can be the solution. But if government finances are the problem, then it’s very unlikely that any kind of intervention in the banking sector can solve anything much. The one thing which no one was worried about during the financial crisis of 2008 was US banks’ exposure to the US government. But the one thing that everybody is worried about in 2010 is European banks’ exposure to European governments.

That said, I think and hope that the European stress tests, if they prove credible (and that’s a big if), will get the interbank market moving again at least between the strongest institutions. Which would be an undeniable improvement on where we’re at now. As El-Erian explains:

Europe would be in the midst of a major banking crisis if the European Central Bank had not become the major counterparty to both sides of the interbank market. Weak banks go to the ECB for liquidity. Strong banks deposit their excess funds at the ECB rather than face other banks in the interbank market.

And what’s to become of the weak banks? Well, they either recapitalize or get taken over by strong banks. Europe’s banks are too big already, and they’re about to get bigger, it seems. But that always happens in a crisis. When you’re worried about solvency, concerns over size always retreat to the back burner.


Size DOES matter… when people say ‘flight to safety’ regarding people investing in USD denominated assets, they are fleeing not to quality but to the safety of size. Something about that needs to be addressed before attacking banks for being too big.

Posted by CDNrebel | Report as abusive

Reporting and lobbying in Basel III

Felix Salmon
Jul 15, 2010 15:12 UTC

The WSJ has its own Basel III update today, which recapitulates most of what we learned yesterday. But there’s one new twist — or new emphasis on an old twist, I’m not sure which. When it comes to proposed liquidity requirements, says the accompanying graphic, “the proposal could cause huge funding shortfalls for banks world-wide.” And the article elaborates:

Something called the “net stable funding ratio,” is another sticking point. That formula would require banks to hold more long-term funding in an effort to make them less susceptible to freezes in the funding markets. Some analysts say that the requirement alone could cost banks trillions of dollars in new funds, and officials could postpone or shelve the idea, people familiar with the matter said.

It would be really helpful to know who the anonymous “analysts” are here: are they part of the IIF-based lobbying to downsize Basel III as much as possible, or are they genuinely independent? And what exactly is meant by “new funds”? Remember that we’re not talking about capital requirements here: the liquidity requirements talk about what form capital should take, rather than how much of it there should be. If you manage your liabilities so that they’re long-term rather than short-term, does that constitute raising “trillions of dollars in new funds,” or is it just a liability-management exercise?

The next paragraph of the article suggests that it might be the latter, and that the anonymous analysts do indeed have a dog in the fight:

Delegates at this week’s meetings discussed current studies by Basel staff and others estimating the likely impact of the proposed rules. Those studies in general show that the version of the rules outlined in December could require banks world-wide to raise nearly $1 trillion in new capital, according to people briefed on the process. That’s considerably less than the multi-trillion-dollar estimates published by some industry groups.

The debate over Basel III is by its nature tough to report on, because financial journalists’ sources tend to be in the financial-services industry, and you really can’t take anything that anybody in that industry says at face value on this subject. Meanwhile, central bankers and other technocrats deeply involved in the negotiations tend not to want to say anything on the record, which means that these stories are full of anonymous sources saying things which may or may not be trustworthy, and which they certainly can’t be held accountable for.

Is it too much to ask that when anonymous sources are used, they can at least be characterized broadly by whether they’re public-sector technocrats or private-sector financial-services professionals with an entrenched interest in weakening the Basel III regulations as much as possible? The standard WSJ “people familiar with the matter” formulation simply isn’t good enough here, especially when it’s used ambiguously: are the “people familiar with the matter” saying just that the liquidity requirements could be shelved, or are they also the source for the multi-trillion-dollar scare estimates?

Basel III is not going to be decided by a legislature susceptible to traditional lobbying techniques. So banks seeking to influence the outcome are naturally going to try to set the tone of the debate by talking strategically to members of the press. Anybody reporting this story should assume that they’re being used, somehow. And be very careful in what they say and how they say it.


“….are they genuinely independent?”

Thank you. Lawling hard to the port side.

Posted by Uncle_Billy | Report as abusive

How can Benmosche be tamed?

Felix Salmon
Jul 15, 2010 14:02 UTC

Robert Benmosche is probably the highest-paid public-sector employee in America: his $10.5 million salary means that he takes home Barack Obama’s $400,000 annual pay every two weeks. And yes, Obama is his ultimate boss.

Along with his outsize remuneration, Benmosche gets lots of other privileges normally denied to civil servants, especially in this administration. He’s allowed to throw regular diva fits, he’s allowed to bully his own board into submission, and he’s even, it seems, allowed to fire his boss against the board’s wishes. In other words, AIG is not controlled by its 80% shareholder: instead, it’s the Benmosche show, with the rest of us just spectators.

This is, needless to say, a novel way of running a state-owned company. Historically the problem with such companies has been that they were hobbled by interference from above; in this case, the problem is exactly the opposite, that the government seems to have no control over the company at all. Yes, it can appoint board members, but the board members wanted Golub to stay on as chairman and he resigned anyway, and they also seem to be incapable of giving Benmosche any kind of strategic guidance. He does what he wants, and if there’s any pushback from the board, he simply threatens to quit.

All of this is complicated by the fact that you can’t look at AIG’s share price to give any kind of indication of how Benmosche is doing: it’s something out of a Borges novel, a security which gets traded at $37.50 per share but which doesn’t really have any value. The important part of AIG’s capital structure is the billions of dollars in debt owed to Treasury: the question there is not whether it will be repaid (it won’t) but how much of it will be repaid. If that number is going up — if the expected amount that the Treasury is going to get back from AIG is rising — then Benmosche is ultimately doing a good job. If it’s going down, then Benmosche is doing a bad job.

So I’m thinking that Treasury should securitize some small portion of what it’s owed by AIG — just a couple of billion dollars, not so much as to make a big difference, but enough to create a liquid market. Then we can look at the market price of those securities, and judge Benmosche much more easily. Without such an indication, he’ll continue to make a mockery of AIG’s corporate governance, and his bosses, American taxpayers.


Greycap, good point.

Posted by FelixSalmon | Report as abusive


Felix Salmon
Jul 15, 2010 03:54 UTC

Steinbrenner pays no estate tax; this is “a social scientist’s fondest dream” — Improbable Research

NYC Man Fined $2,000 For Taking Discarded Garbage — WCBS

Having small investors participate in the market only undermines markets’ role in governance and capital allocation — Yglesias

This TBI parody is absolutely priceless — Photobucket

Possibly the greatest media prank ever: Rupert Murdoch, Neel Shah, and the Short Pants — Get Excited

Khoi Vinh is leaving the NYT. A huge loss for them, very sad. He’ll do great in his new projects — Subtraction

Someone tell Oyl Miller that a tweet can’t be 503 words long — McSweeney’s

Cardiff Garcia on Sebastian Mallaby on hedge funds — Garcia


Oy! Felix! “Tweet” is this generation’s “Howl” (or at least an homage to it), not an actual tweet.

Posted by Citoyen | Report as abusive