Felix Salmon

Are restaurants which get good reviews more likely to raise their prices? That’s the question I tried to answer in my latest Gastronomics column, and I think the answer is no.

I didn’t have the ability to do very deep quantitative research on the question, partly because historical prices for restaurants are very hard to come by. But with the help of fabulous New York intern Ray Rahman, I did manage to look at what’s happened to the prices at all the restaurants Sam Sifton’s reviewed for the NYT over the past year. And here’s the result:

(Source: this spreadsheet.)

The message of the graph is pretty clear: price hikes happen all over the place, from the worst-reviewed restaurants to the best. And most restaurants are tweaking their menu prices most of the time, for many reasons. Meanwhile, restaurants with no reviews at all of late, like Gordon Ramsey at the London or Milos, often have the highest prices and the biggest price hikes.

The inspiration for the column was Bloomberg’s Ryan Sutton, whose wonderful Tumblr, The Price Hike, is a must-follow for all New Yorkers. Sutton concentrates mainly on tasting menus, partly because those are where the big money is, and partly because it’s much easier to remember how much a single tasting menu costs than it is to remember a long list of prices. But even for those of us who avoid tasting menus as much as possible, it’s always a good idea to remain conscious of prices. Because restaurants have a nasty habit of waiting until you’ve sat down to spring their latest price hike on you — even when it’s a very big one.

The chart of the day comes from Martin Wolf, who has an important column on the way that Europe’s central banks have been staving off a full-scale eurozone crisis.

What you see here is the amount of money that various individual European central banks — not the ECB — owe each other. From the creation of the eurozone in 1999 up until late 2007, such lending between European central banks was modest — but then, when the financial crisis hit, the Bundesbank in particular went on a lending spree, while the PIGS central banks all started borrowing up a storm. Since then, the Bundesbank has essentially been singlehandedly financing the PIGS central banks: it’s now owed a whopping €325 billion, and rising. (That’s about $470 billion, at today’s exchange rate; we’ll surely hit the half-trillion-dollar level soon.)

What’s being done with all this money? “Let us call a spade a spade,” says Wolf: “This is central bank finance of the state.” It’s indirect, to be sure, but this graph is a finger in the eye of Article 123 of the Lisbon Treaty, which bans monetary financing. And it means that there’s a whole new set of dominoes which could, catastrophically fall:

Government insolvencies would now also threaten the solvency of debtor country central banks. This would then impose large losses on creditor country central banks, which national taxpayers would have to make good. This would be a fiscal transfer by the back door.

Needless to say, no one envisaged, when the eurozone was created, that individual countries’ central banks could become insolvent. But that’s a real possibility, now, and no one has a clue how to address that kind of problem. What we do know, if our memories stretch back as far as September 2008, is that such an event could have enormous consequences. Here’s the ECB’s Lorenzo Bini Smaghi, in conversation with Ralph Atkins:

FT Some argue that a Greek default would be the least worst-option for the euro area. The impact on the euro area would be containable and it would reinforce euro area principles long term. Why are they wrong?

LBS It would not be the least worst option – as we can see from the reaction of financial markets, not only within the euro area but also outside. The destabilizing effect could be quite dramatic. Those who say that the impact would be contained simply do not look at the data. It reminds me of those who in mid-September 2008 were saying that the markets had been fully prepared for the failure of Lehman Brothers.

All sophisticated indicators of systemic risk, cross correlations of CDS and yield spreads show a high sensitivity to restructuring moves and are at levels higher than in September 2008. Suggesting that there are no contagion risks is naïve and entails taking a risk that no responsible policy maker can afford, if he or she has any interest in the well being of its citizens.

Tracy Alloway, today, picks up on much the same theme, quoting a research report from Merrill Lynch: while volatility is low, correlations are still very high. And that’s a combination which tends to presage nasty price crashes across many asset classes. In other words, markets are exhibiting a lot of fragility right now, and something drastic like a Greek restructuring could easily send them into a Lehman-style downward spiral, as Wolf spells out:

Debt restructuring looks inevitable. Yet it is also easy to see why it would be a nightmare, particularly if, as Mr Bini Smaghi insists, the ECB would refuse to lend against the debt of defaulting states. In the absence of ECB support, banks would collapse. Governments would surely have to freeze bank accounts and redenominate debt in a new currency. A run from the public and private debts of every other fragile country would ensue. That would drive these countries towards a similar catastrophe.

The only other choice, says Wolf, is equally intolerable: Europe’s richest governments aiming a firehose of their taxpayers’ euros, in open-ended and unlimited quantities, at any country facing massive outflows.

Given these choices, it’s easy to see how Europe’s politicians and central bankers are doing everything they can to kick the can down the road and put off the moment that they have to make a big decision. But the longer they wait, of course, the more momentous and more difficult any such decision is going to be. Just how much risk are Europe’s central banks going to take on, before they draw the line and say no mas?

Amazing how a 90-second video can make you hate an entire city — Tumblr

In which Donald Boudreaux determines that “life on Earth” = “tornado fatalities in America” — GWPF

Goldman Sachs’s expensive typo — Alphaville

Louise Story and Gretchen Morgenson have a long and rambling story about the court case against Goldman’s Fabrice Tourre, which is mainly interesting for how it was sourced:

These legal replies, which are not public, were provided to The New York Times by Nancy Cohen, an artist and filmmaker in New York also known as Nancy Koan, who says she found the materials in a laptop she had been given by a friend in 2006.

The friend told her he had happened upon the laptop discarded in a garbage area in a downtown apartment building. E-mail messages for Mr. Tourre continued streaming into the device, but Ms. Cohen said she had ignored them until she heard Mr. Tourre’s name in news reports about the S.E.C. case. She then provided the material to The Times. Mr. Tourre’s lawyer did not respond to an inquiry for comment.

I’m sure this was extremely carefully formulated, but it does raise a lot of questions without answering them. Tourre’s name was splashed over the newspapers in April 2010, so it stands to reason that the NYT has had some kind of access to Tourre’s private, password-protected email account — not to mention archives going back at least to 2006 — for a good year at this point. I’d also guess that the NYT is going public with its source now because Tourre finally got around to changing his password, and the stream of emails then dried up.

Was the NYT, then, hacking into Tourre’s private emails in much the same way as the News of the World was hacking into private voicemails? The NYT certainly didn’t think much of that activity, even when it was done through an outside contractor rather than a newsroom employee. So I don’t think it makes a lot of difference whether the computer was in the possession of Cohen or of the NYT.

The NYT quotes one email from October 10*, long after Tourre was all over the news in April. So it seems reasonable to assume the NYT was accessing Tourre’s emails even after the reporters were initially approached by Nancy Cohen. In order to do access Tourre’s emails, they would need Tourre’s password, or a computer with the password on it.

I understand that the computer was found in a garbage area, and that there’s a long tradition of investigative reporters using information found in the trash. But if Tourre left a key to his apartment in the trash, that wouldn’t give reporters the right to use that key to enter his apartment and snoop around. The laptop was essentially a key to Tourre’s email account — which held highly confidential correspondence between Tourre and his lawyers. An email account, these days, is arguably more private than an apartment, and breaking into a password-protected email account is clearly wrong.

How is what the NYT did not hacking into a private email account? I can think of three defenses, only one of which has any real merit.

The first defense is that it’s only hacking if you somehow type in the password: if all you have to do is open up the email application and the emails just stream in automatically, then there’s no hacking involved. I don’t buy this defense at all. If Cohen found Tourre’s password on his computer and gave it to the NYT, they could then build an identical laptop for themselves — and doing so would clearly be hacking. It just doesn’t make sense that if you’re looking at two identical computers, using one to snoop through emails is ethically fine, but using the other is beyond the pale. The laptop is essentially the physical manifestation of the password, and using it is the same as using the password. Using someone else’s password to access their email is wrong, even if you found that password in the trash.

The second defense is that it wasn’t the NYT reporters who were snooping through Tourre’s emails, but rather Nancy Cohen. Again, I don’t buy this one — Cohen was clearly delivering to the reporters exactly what they asked her for. If she was doing their bidding and acting as their agent, then they bear responsibility for her actions.

The third possible defense is that Cohen waited for a good six months* after Tourre was in the news before dumping all of his emails onto some kind of drive and delivering it to the NYT in a one-shot deal. The NYT, with a trove of data in its lap, then combed through the emails and wrote its story, without asking Cohen to keep further tabs on Tourre’s email or give them any further data. That I think would come closest to the we-found-this-in-the-trash scenario, and doesn’t involve the NYT taking advantage of having password-protected access to Tourre’s email account. But it’s also pretty improbable.

It goes without saying that Tourre is extremely upset right now, and feels violated by the NYT. He’s not wrong to feel that way. The question is whether the NYT was wrong to snoop around his emails while fishing for a story — even if doing so meant hacking into his private account.

Update: The Stored Communications Act of 1986 makes it a crime to “intentionally accesses without authorization a facility through which an electronic communication service is provided”.

Update 2: The NYT has released a statement:

As we disclosed in our story, certain documents were provided to us by a named source. The Times did not “hack” any email accounts or ask anyone to do so.  We are confident that our receipt and use of those documents was in keeping with our journalistic standards and complied with the law.

Update 3: John Cook has a very smart take on all this, and some more information from the NYT:

Murphy also strongly suggested that the laptop at issue didn’t belong to Tourre: “All but one [emails referenced in the story] came from the Senate report and are public and none came from Mr. Tourre’s email system.” If it didn’t come from Tourre’s email system, it’s hard to see how it came from his laptop.

As Choire Sicha has noted, Goldman is quite the professional organization when it comes to information security, and emails don’t just “stream” unbidden to Goldman-authorized devices. So it’s likely that it was a computer belonging to one of his attorneys—if the laptop’s “Sent Items” mailbox was synced via IMAP or Microsoft Exchange, for instance, with Tourre’s attorney’s email account, then “messages for Mr. Tourre” would have continued streaming to it.

I like the idea that it wasn’t Tourre’s laptop, but rather a laptop belonging to someone who was emailing Tourre. But that would imply that Cohen was not only reading a stream of emails, but was keeping close tabs on the “sent items” folder as it updated, and noticed Tourre’s name popping up in that folder periodically. Anything’s possible, I guess.

*Update 4: Nick Rizzo points out that the October email quoted by the NYT actually looks as though it dates from October 2009, rather than October 2010. Here’s the NYT story:

In the fall of 2009, when Mr. Tourre learned that he had become a target of investigators for helping to sell a mortgage security called Abacus, he protested that he had not acted alone.

That fall, his lawyers drafted private responses to the S.E.C., maintaining that Mr. Tourre was part of a “collaborative effort” at Goldman, according to documents obtained by The New York Times…

<snip 961 words>

…In their Oct. 10 response to the S.E.C., Mr. Tourre’s lawyers, including Pamela Chepiga of Allen & Overy, made an argument that they have not emphasized publicly.

If the NYT only obtained emails to Tourre dating from before they were contacted by Cohen, then as I say, that would absolve them of hacking charges. I read the NYT’s reference to October as meaning last October, since the SEC case against Tourre only became public in 2010, and because if people mean two Octobers ago they normally say so. But revisiting the article, it seems more likely that the email in question was sent in 2009, a good six months before the SEC case against Tourre was made public. In that case, it’s no longer prima facie evidence that the NYT was hacking into anybody’s email.  Although it’s still very unclear why the NYT waited more than a year between being given the emails and publishing this story. If it’s because they were still monitoring email to Tourre up until recently, then the hacking allegations don’t go away.

Update 5: This just gets weirder. The NYT emails Rizzo to say that none of the emails cited in the NYT story “came from Mr. Tourre’s email system on the computer”. Which implies that the NYT was reading Tourre’s emails on the computer, but for some reason didn’t use any of those emails in its story. Of course, as far as hacking into email systems is concerned, I can’t see that it makes any difference whether you print them or not.

Update 6: Nancy Cohen has a spokesman! His name is Curtis Ellis, and Nick Rizzo has spoken to him. Ellis’s story goes like this: Cohen was given Tourre’s laptop in 2006, from a friend who had found it in the garbage section of the basement of a luxury downtown building. Four years later, when Tourre is in in the news, Cohen realizes it’s the same person who’s name is on the computer. She opens up Outlook, and it starts downloading a bunch of his emails, which she doesn’t read. She tries to get in touch with various media outlets, but none of them is interested, until, eventually, a couple of months ago — almost a year after first making the connection and downloading Tourre’s emails — Cohen finds someone at the NYT (neither Story nor Morgenson, it seems, but someone else) who says they’ll take a look and see if there’s anything there. She then delivered the emails on an external hard drive to the NYT.

If this story is true, then it does pretty much exonerate the NYT from allegations of hacking — Cohen really did simply deliver all of Tourre’s emails to the NYT, long after he was in the news, in a one-shot deal. It sounds pretty crazy, to be sure. But weirder things happen in New York every day.

The U.S. Congress only has two choices. It can raise the debt ceiling, or it can abolish the debt ceiling. The latter option isn’t realistic, sadly. Which means that the debt ceiling is going to be raised. Yet somehow the House of Representatives contrived today to vote 318-97 against raising the ceiling, in an idiotic political stunt which makes all 318 Nay votes look like complete morons. Not a single Republican voted in favor of the bill, while the Democrats were pretty evenly divided between the sane and the bonkers; even the likes of Nancy Pelosi and Charlie Rangel voted nay.

Faced with a Congress of such monumental doltishness, what is the White House to do? It can stick to its guns and try to put in place the very best policies for America that it can. Or it can randomly detonate various such policies, even if doing so would achieve precisely nothing, on some kind of inchoate principle that the occasional public sacrifice might somehow mollify an unknown number of lamebrained legislators.

That seems to be the philosophy of Bill Cohan, who has decided that Elizabeth Warren must resign from the nascent Consumer Financial Protection Bureau, on the grounds that Congress won’t confirm her. But Cohan’s argument doesn’t even make internal sense:

The inconvenient truth facing Elizabeth Warren, the controversial Harvard Law School professor President Obama would like to run the newly created Consumer Financial Protection Bureau, is that she has made herself so bloody disagreeable on Capitol Hill that she has obliterated her chance of winning the Senate votes she needs to be confirmed…

In the Senate — the body that actually confirms appointees — Warren is faring little better. In early May, 44 Republican senators sent the president a letter saying they would oppose any nominee of either party to head the bureau until “the lack of accountability in the structure” of it is “reformed.”…

This reeks of a political ploy by the Republican senators to gut an agency despised by their financial backers on Wall Street.

By Cohan’s own account, then, any nominee would face exactly the same fate as Warren — even a Republican. It’s the bureau that these senators don’t like, and they’re not going to confirm anybody to run it unless and until they also get the authority to hobble it. Throwing Warren to the sharks would just give them a taste for blood, and make them even more optimistic about their chances of getting exactly what they want.

Cohan says that the Warren nomination fight “is not a fight that Obama can win”. But it’s the first principle of any game that you don’t give something up without getting something in return — especially when your opponent is crazy. Unless and until the Republicans show Obama a way that he can get a nominee approved in line with Dodd-Frank’s vision for the agency, it’s a no-brainer that Warren must stay. Otherwise Obama would just be sacrificing a great public servant for no purpose at all.

Back in December 2007, before Bear Stearns went bust but just as the country was tipping into recession, a California lawyer named Sean Olender was worrying about the putbackpocalypse — the idea that holders of mortgage-backed bonds would force banks to buy them back at par, thereby wiping out the capital in the banking system. To his credit, Olender was one of the first people to latch onto this worry, which still hasn’t gone away.

But Olender went much further than worrying about putbacks; he was also convinced that there was a conspiracy afoot to prevent them from ever happening. The idea — which he outlined in an op-ed article on the front page of the C section of the San Francisco Chronicle — was, essentially, that Hank Paulson knew all about fraud in the mortgage origination process from his time as CEO of Goldman Sachs. And that therefore he knew all about putback risk. And that therefore he was deathly worried about putback risk. And that therefore when Paulson proposed that subprime adjustable-rate mortgages be frozen at their “teaser” rates for five years, what he was really doing was trying to put into place a cunning plan which would prevent any putbacks.

With hindsight, this credits Paulson with far too much prescience. As we now know, Paulson was taken by surprise by the events of September 2008, and wasn’t remotely prepared for what happened, responding in a panicked and ad hoc manner. His long-forgotten mortgage-freeze proposal was just a mortgage-freeze proposal: it wasn’t an attempt to prevent putbacks, largely because there’s no real reason to believe that a frozen mortgage would be harder to put back to the originator than an unfrozen mortgage.

Olender, by contrast, painted Paulson as a criminal fraudster, taking advantage of his position at Treasury to avoid lawsuits and even prison. I was not impressed, and responded by calling his column “incendiary and meretricious”.

With the benefit of hindsight, I went too far in my post. I said that there was nothing to Olender’s accusations of fraud, when in fact there was a lot of fraud in the mortgage-origination process, even if there wasn’t a criminal conspiracy reaching up to the Treasury secretary to cover it up.

I’m pretty sure I was right about our fundamental disagreement: the mortgage freeze proposal was not a fraud, it was just a well-intentioned attempt (which failed) to freeze mortgage rates. (If it had been driven by bank fears of putbacks, you would think the banks would have embraced the proposal, rather than successfully shooting it down.)

But I did attack Olender personally — I called his piece “grossly irresponsible”, and lumped him in with “nutcases” who are “rather out to lunch”, including Ben Stein. All of which has prompted Martha Hamilton to declare that I owe Olender an apology.

Hamilton didn’t talk to me before writing her piece, and I’m pretty sure she didn’t talk to Olender either. So she spends a lot of time on what I think is a side issue here — the culpability of the investment banks in the mortgage crisis. Yes, they deserve prosecution, as Carl Levin and others have shown. But I never said they were white as snow; I just said that they were generally not at risk of having bonds put back to them, if they didn’t originate those bonds. The banks did buy a handful of subprime mortgage originators, but most subprime loans were originated by independent shops, at least until Countrywide got bought by Bank of America — something which happened long after Olender’s column was published. Again, if the banks were very scared of putbacks, BofA would never have made that acquisition.

And Hamilton’s simply wrong about this:

I remember, in the early days of blogging, questions raised about how far journalists could go in this different form of writing. There was pressure to be over the top, to be outrageous, to turn yourself into an attitude-propelled superbrand. And some journalists went too far, as I think Salmon did here.

I was hired, a few months before this post was written (but still, long after “the early days of blogging”), to be the finance blogger for a brand-new and meticulously fact-checked glossy monthly called Condé Nast Portfolio. I can assure Hamilton that never once was there any pressure at all to be over the top, to be outrageous, or for me to to turn myself into an attitude-propelled superbrand. Quite the opposite: all the tensions I had with my editors were a function of me having too much attitude, rather than too little, especially when the subject of my tirades was an article in Condé Nast Portfolio itself.

Insofar as there was attitude and outrage in this piece, it was simply me, saying what I thought of Olender’s piece in one of my many siwoti moments. Yes, I was shrill, and I make no apologies for that. Hamilton reckons I “would have benefited from an editor who might have helped me walk it back”, as though shrillness is a bad thing, in blogging; in fact, it’s one of the things which makes blogs compelling and readable in contrast to long and dutiful pieces of classical journalism.

But still, I was directionally wrong in December of 2007, and Olender was directionally right. Because Olender was ascribing so much bad faith to so many players in the market, up to and including the secretary of the Treasury, I missed an important point about the degree to which the whole mortgage-bond edifice had been built on fraudulent foundations. And in missing that point, I found it far too easy to dismiss everything that Olender had to say on the grounds that some of it was clearly going too far. Having done that, I just piled on, making my share of bad points along with my good ones.

In the normal course of things, my post would be long forgotten today, just one of ten different blog entries I posted on December 11, 2007. Certainly I would have forgotten it. But then there’s the matter of Google: my post remains stubbornly high in the Google search results for anybody looking up Olender’s name.

Olender is upset about that, and has periodically sent me dyspeptic emails asking for an apology in between calling me “such an idiot”, or “the foppish Felix”, or a “pompous Metrosexual” lacking “moral sensibilities” who hasn’t “been adequately trained as a journalist”. Since he was pretty much right about all that, I forwarded his latest email on to CJR for a third-party verdict, and that’s why Hamilton’s weighing in now, with her verdict that I owe Olender an apology.

So: well done, Sean, for being very early to the problem of putbacks. You were much more right than I gave you any credit for, or than I thought possible at the time, and you certainly didn’t deserve to be lumped in with Ben Stein. Sorry.