Felix Salmon

Do you want to set up a network of insider-trading tipsters? Well, Peter Lattman and Azam Ahmed are here to tell you exactly how to do it, using the secrets of the master of the art form, Raj Rajaratnam.

Different techniques work for different people, of course, but often the direct way — a lot of money, mixed with equally large amounts of flattery — is the easiest:

As Raj Rajaratnam and Anil Kumar, a McKinsey consultant, walked out of a fund-raiser in Manhattan, Mr. Rajaratnam pulled his old friend aside and made him an offer: would Mr. Kumar provide him with insights for $500,000 a year?

“You have such good knowledge that is worth a lot of money to me,” he said, according to Mr. Kumar.

Mr. Kumar faced an agonizing choice. His employer barred its executives from outside consulting, but an extra half-million dollars a year — and the chance to do business with a powerful hedge fund manager — was tantalizing.

Weeks later, Mr. Kumar accepted.

It’s worth noting that at this point neither man has done anything illegal. If found out, Kumar could lose his job — but being fired, even for cause, is not a criminal offense.

Yet the die, at this point, has been cast. Raj has both a carrot and a stick with which to control Kumar: money, and the fact that he knows that Kumar has been accepting it. The two are bound into a secret conspiracy, and once you’re in such a thing it’s impossible to get out without inflicting serious pain onto yourself.

So when Raj started asking for inside information — when he started asking Kumar to do things which were actually criminal — it was easy for Kumar to say yes, and very hard for him to say no.

With other people, Raj used different techniques. Adam Smith (yes, Raj really was getting inside information from a man named Adam Smith) was probably the easiest: Raj simply brought him into Galleon as an employee, making their interests pretty much fully aligned.

With Rajiv Goel, there was a real friendship — or at least Goel thought there was. And while money changed hands as well, it wasn’t money for tips, not directly: it was more that the rich friend, Raj, helped out with things like buying a house or caring for a sick parent, while the poorer friend, Goel, desperately tried to curry Raj’s approval in the only way that he could get it.

Incidentally, only in the world of Wall Street is it unsurprising to find mid-level Intel executives being described as “hapless” and “in need of money” — Goel had a job that most people can only dream of, but was permanently dissatisfied. Maybe if his friends had less money than he did, instead of more — if they were in the bottom 99% of the population, rather than the top 0.01% — then he would have been happier, and would have felt much richer. Lesson of the story: don’t vacation with people who are a lot richer than you are.

And then there’s Kumar, who’s the weirdest of the lot, seeing as how he was already earning several million dollars a year at McKinsey. He was set for life, yet he accepted Raj’s $500,000 a year, and also the occasional bonus:

In 2006, Mr. Kumar agreed to another compensation scheme: Mr. Rajaratnam would pay him a year-end bonus based on his annual performance. Mr. Kumar proved his worth that year, providing him with details about secret merger negotiations between Advanced Micro Devices and ATI Technologies…

In December, Mr. Rajaratnam told Mr. Kumar that Galleon was paying out big year-end bonuses. “I want to give you $1 million,” Mr. Rajaratnam said.

“I almost fell off my chair,” Mr. Kumar testified.

$1 million is a lot of money for almost anyone, but in Kumar’s case it was not enough to change his standard of living at all, and it didn’t make him significantly richer than he was before. So how come Raj’s money had so much effect on him?

Kumar sticks out here in other ways, too — he’s the only informant who could be considered even more successful than Raj was, at least professionally if not in terms of raw cash. Raj had money, more money than he really knew what to do with, but Kumar had much more societal acceptance and prestige — things you definitely need if you’re going to make it all the way to the board of Goldman Sachs.

And, of course, Kumar still hasn’t been criminally charged. But I suspect that particular shoe is going to drop at some point. Now that Raj has been convicted of all 14 counts, Kumar is surely next.

Update: As EnricoPalazzo points out in the comments, it’s Rajat Gupta, not Anil Kumar, who made it onto the board of Goldman Sachs.

NYT pageviews drop 24% in April — AdAge

HSBC still hasn’t resumed foreclosures — HuffPo

“One of the odder lots is a crystal decanter filled with a mysterious brown liquid” — Bloomberg

Pondering the ontology of the “exclusive Gaga unicorn”, and wondering what exactly “exclusive” means in this context — TBI

Only 14% of Afghan army and police recruits are literate. Not a single soldier one company needing medevac knew their coordinates — Guardian

The Gaussian Copula Function, tattooed — Tumblr

What To Do If Your Wallet is Stolen — Credit

James Stewart to take over Nocera’s Sat column at NYT — TBN

On the Treasury’s Curious Denial That Geithner Blocked Deal on Irish Debt — NC

SecondMarket put on a conference in San Francisco this morning, where I got to talk to chief strategy officer (whatever that means) Jeremy Smith. I asked him about my theory that it’s easy to make big acquisitions if you’re public, using a hypothetical Facebook-Skype deal as my example.

Jeremy pushed back a bit: anything Facebook could do as a public company, he said, it could do as a private company too. Leverage? Banks would be lining up to lend money to Facebook right now. A big capital raise? Again, there’s no shortage of people wanting to invest in Facebook, or of banks willing to give Facebook a bridge to such a raise.

There is however a huge difference if Facebook wanted to pay in stock. For one thing, illiquid stock in a private company is much harder to sell than liquid stock in a public company. And while a VC fund might be willing to accept stock as payment on the understanding that it would sell that stock pretty quickly, it would be impossible to persuade any such fund to accept another form of illiquid equity. They’re meant to be making an exit here, not a new investment. If Facebook is buying a small company owned by its founders, they might be willing to take stock in payment. But exiting venture capitalists, not so much. (Unless the exiting fund could then sell that stock to a younger fund run by the same company; I’m not sure whether that would work.)

I learned quite a lot at the conference about the nuts and bolts of listing on what SecondMarket likes to call its “liquidity platform” — it’s not nearly as onerous as listing on the NYSE, but it’s still not easy, and it does require a fair amount of legal legwork. So far, the smallest company to use it had a valuation of about $150 million; it’s designed for companies worth $1 billion or so. That’s big money — but still small enough that we’re talking about a set of companies which are too small to go public, judging by the size of IPOs in recent years.

In any event, most of these companies won’t go public: they’re nearly all VC-backed, and 90% of VC exits are via M&A rather than via IPO. And this is where the real value of a SecondMarket listing becomes apparent: in an M&A transaction, the acquirer is always going to feel the need to offer some kind of premium over the latest value that the shares fetched on SecondMarket. Without that price being out there, negotiations can be harder, since the buyer wouldn’t be able to see the price that a significant number of buyers with ready cash are willing to pay for equity in their target.*

And one panelist, I forget who, pointed out another clever way that SecondMarket is changing the way that companies and investors interact: historically, secondary offerings, of stock held by existing shareholders, always took place after an IPO. Now, with SecondMarket, they’re taking place on a regular basis before an IPO. That removes an important incentive to go public, and will only serve to make the average age of companies at IPO even higher.

Meanwhile, all the signals on Capitol Hill seem to be pointing towards the SEC making life easier for companies like SecondMarket, and embracing the new halfway house between private and public. That would make the government far more responsive to this development than the law world, where Wilson Sonsini’s Yokum Taku said that he be “shocked” if the standard provisions in the structure of VC-backed companies, which haven’t really changed since about 1974, were changed at all in his lifetime.

The most interesting panel was moderated by Dan Primack, who asked a good question: what happens when the bubble bursts? Right now SecondMarket is riding high because pretty much all the companies using it to trade their stocks are seeing their valuations float effortlessly ever higher. But once those valuations crash, will people still be willing to sell? And who will want to buy? Will there be vulture secondary investors?

My worry is that these markets are a bit like the housing market, where people remember the valuations at the peak of the market, refuse to sell for substantially lower amounts, and the market stops clearing. After all, the primary market in private equity — the capital-raising rounds which are marked as Series A, Series B, and so on — is highly allergic to “down rounds” and does tend to seize up during market downturns. Why should the secondary market be any different?

The public markets, by contrast, go up and down all the time — they have no problems at all with stocks going down. So the open question is whether private secondary markets are more like public secondary markets, or more like private primary markets. We’ll find out, I guess, when the current dot-com bubble finally crashes.

*Update: In a classic example of the way that public shareholders get less information than anyone else looking to buy a stake in the company, it’s worth noting that this information does not make its way into the IPO prospectus. LinkedIn’s S1, for example, says that its shares have been trading on SecondMarket. But it doesn’t say for how much.

Over the weekend I watched the HBO movie version of Too Big To Fail, and I talked to Andrew Ross Sorkin about it on Monday.

As you can probably tell from the movie trailer, it’s got lots of dramatic music and equally dramatic moments — Paul Giamatti (Ben Bernanke), for instance, telling a room of assembled politicians that he’s spent his entire academic career studying the Great Depression, and that “if we do not act, boldly and immediately, we will replay the depression of the 1930s. Only this time, it will be. Far, far worse.”

It’s a curious fish, this film: there are so many white guys in suits that unless you already know the story going in it’s pretty much impossible to tell who’s who — with the exception of the lead character, Hank Paulson, played with searing intensity by William Hurt. And this isn’t the story of the crisis — for that, the Oscar-winning Inside Job will serve you much better. This is just the story of the short period of time during which Lehman collapsed, AIG was bailed out, and TARP got enacted.

What I worry about is that with the movie concentrating on a brief period of time — just one month in the course of a crisis which started much earlier and ended much later — the public will continue to think of TARP as the defining bailout and government action of the crisis. In reality, however, monetary policy was much more important than fiscal policy, and TARP wasn’t even the most effective fiscal policy — that was the Obama stimulus package which was pushed through in early 2009.

The movie is also a combination of too complicated and too simple. It’s too complicated in that there are too many characters and it’s hard to keep them all straight, and the occasional kludgy attempt at explaining, say, why AIG collapsed falls very flat. And it’s too simple because the corners that were cut turn out to create misleading impressions — that Warren Buffett was willing to buy into Lehman Brothers at $40 per share, for instance, or that the Chinese government was a significant shareholder in Frannie.

But the fact is that even though the financial crisis really did threaten the global economy, it wasn’t really a thrilling human drama. As Lloyd Blankfein says to his chief of staff Russell Horwitz, “You’re getting out of a Mercedes to go to the New York Federal Reserve, you’re not getting out of a Higgins Boat on Omaha Beach. Keep things in perspective.” If you want a thrilling drama, then Hollywood is good at making those. And if you want to understand the financial crisis, this film is too constrained in time to really explain what happened. So I’m not sure really what the film is good for, beyond providing some welcome money for financial journalists like Sorkin, Joe Nocera, and Bethany McLean. I’m all in favor of that!

At the Finovate conference today, Wikinvest released their latest product, SigFig, which is aimed at saving investors money. And it turns out there’s a lot of money to be saved.

A lot of people have signed up for Wikinvest and handed over access to their brokerage accounts. I spoke briefly to SigFig founder Parker Conrad, who explained that it’s incredibly easy to flick through those accounts and come up with examples like the one he pulled up, of a man with $2.3 million in his Merrill Lynch account.

This guy probably knows that he’s paying his Merrill broker an annual management fee of 1.75%, which alone is more than $40,000 a year. But he doesn’t know that other Merrill clients in his position are paying far less — that Merrill brokers basically charge as much as they can, and the average Merrill client on Wikinvest pays less than half that, just 85 basis points.

And there are other things this guy doesn’t know, as well, because they’re buried in his statements — things like the fact that Merrill charged him $5,763 to make 24 trades last year, over and above that $40,000 management fee. That’s about $240 per trade.

Other fees are even higher. The Merrill broker bought something called the Fidelity Advisor International Capital Appreciation Fund, which charges 1.45% per year on top of a 5.75% fee payable when you buy the thing in the first place. The fund is substantially identical to the Fidelity International Capital Appreciation Fund, which has a 1% management fee and no front-loading at all. Why would any advisor with his client’s best interests at heart put that client into FCPAX rather than FIVFX? He wouldn’t — FCPAX is simply a vehicle invented by Fidelity for advisors which allows them to skim off hefty commissions.

SigFig, then, looks like a good tool to use on any brokerage account which you’ve had for a while. Upload your information, and it’ll tell you immediately if you’re getting ripped off.

There are lots of other aspects to SigFig too, most of which are designed to appeal to more active investors, or people who like to keep close tabs on their money. I’m not convinced there’s a lot of value in that, but it’s clearly something which is well targeted at the existing Wikinvest user base. But at the very least, SigFig is a great way of identifying some of the billions of dollars in rents which are extracted each year by brokerages and financial advisors across the country. Mother Merrill has historically been good to its employees. But that doesn’t mean it’s good for its clients.

I’ve been playing around a bit with Bundle, a clever new tool which lets you scope out merchants — shops, bars, restaurants, hotels — based on actual aggregated consumer behavior, rather than reviews and other qualitative information.

Bundle’s based on a huge anonymized database of consumer credit-card and debit-card spending. So when you call up a merchant on Bundle, it’ll show you reviews, but it’ll also show you data like how often people who shop there go back.

The raw data is fascinating. For instance, the average spend at the Breslin is $70-80; at the Spotted Pig it’s $80-90; at Momofuku Ssam Bar it’s $90-100; at Lure Fishbar it’s $120-130, at Annisa it’s $200-210, and at Gramercy Tavern it’s $220-230. (At Per Se, it’s $890-900.)

And you can learn a lot just by seeing that Spotted Pig customers also tend to spend money at the Starbucks on the Montauk Highway in Hampton Bays.

Then there’s Bundle’s proprietary loyalty score, which is promising but which needs work. It’s based on three variables: purchase frequency, share of wallet, and popularity. And it’s meant to spit out a score in a range between 1 and 100. But I’ve been plugging in the restaurants I know with the greatest customer loyalty — the ones above, as well my own favorite, Oyster Bar — and so far the only place I’ve found with a score over 50 is Szechuan Gourmet, whose score of 60 still only counts as “average”. Poor Annisa only gets a 32, which might be related to its being closed for a large chunk of the database’s time period.

Meanwhile, the most popular restaurant in New York is a betting parlor, while #2 is a hairdresser.

So I’m definitely looking forward to the future of Bundle, where the loyalty algorithm will get improved. But the technology here is really promising: we vote with our wallets every time we spend money, and now those votes are being counted and analyzed.

How does Bundle separate out places where people tend to split the check from places where they don’t? Or places where they tend to go as couples from places where larger groups are more common? It doesn’t, yet, and I’m not sure it will ever be able to. But they do say that they’re looking to put together subsegments of the customer base — foodies, for instance, or deal-hunters.

And in the meantime, we can just marvel at some of the data which is already there. Here, for instance, is the data for Robert’s Steakhouse, an excellent restaurant which just happens to be located inside a strip club. Now that, it turns out, is a great way to build loyalty. And an average check of over $500.

If you want to see market reporting done right, I can recommend the 2,000-word Reuters special report on Thursday’s commodities crash. It doesn’t just pick a random news event or gesture vaguely at “worries about economic growth” while saying what prices did: it looks at the mechanisms behind the market moves and what might have caused them.

It’s worth underlining that Thursday’s percentage declines in commodities like silver and oil would count as a full-on disaster if they occurred in the stock market. Commodities markets are rowdier places than stock markets, however, and the only people who really got hurt are sophisticated investors who can take their medicine.

The move was certainly accelerated by the rise of algorithms and high-frequency traders, who have moved quite aggressively from stocks into commodities of late. These black boxes can go from being very long to very short in an alarmingly short space of time, and I suspect that many of them made money, rather than lost it, in the volatility.

But there was also a sense that this move had to happen. Between early February and early May, the yield on the 10-year Treasury bond fell from 3.7% to less than 3.2%. That’s a massive move in Treasury yields, which indicates that market fears about inflation are abating significantly. At the same time, however, the price of silver — a classic inflation hedge — rose from $27 to $47. One of those two markets was wrong — and it was much more likely to be the thin silver market than the Treasuries. Silver was bound to fall in price; the only question was when.

When the inevitable silver crash happened, it took down other commodities like oil with it. That’s because of all the speculation in the market, and the fact that funds which speculate in silver tend to be exactly the same funds speculating in oil. When you get a big margin call in silver (and margin requirements on silver had just been raised before the crash), then you have to sell some of your other holdings to meet that call. And your most liquid holding is likely to be in oil.

At times of volatility, correlations move towards 1. We saw that in every market during the crisis, and we saw it again in commodities on Thursday. Which is why protecting yourself with diversification is so dangerous. Just when you need the protection, it disappears.

The acquisition of Skype by Microsoft comes just in time for the Capitalyze conference in San Francisco, which I’m sure will be talking about this:

The biggest winner of this deal could actually be Facebook. The Palo Alto-based social networking giant had little or no chance of buying Skype. Had it been public, it would have been a different story. With Microsoft, it gets the best of both worlds — it gets access to Skype assets (Microsoft is an investor in Facebook) and it gets to keep Skype away from Google.

If Om’s right about this, then Facebook is just plain lucky that deep-pocketed Microsoft came along to keep Skype out of Google’s hands. If Facebook were public, on the other hand, then it could have just snapped Skype up itself.

I’ve already said that Facebook will go public — but for boring technical reasons, rather than for big strategic reasons like this. And so the question arises: is Om right? Does being public give companies the ability to make large strategic acquisitions, which are impossible so long as they’re private?

This particular case, like so many other cases where Facebook is involved, is exceptional. Skype’s owners, including Silver Lake Partners and Andreessen Horowitz, might well have been quite well disposed towards a deal where they sold Skype to Facebook and got a large yet illiquid chunk of Facebook in exchange. But I’m not sure if that’s even possible, the way that those funds are set up in Silicon Valley: while Silver Lake and Andreessen Horowitz are indeed investing in the likes of Facebook, they’re investing their new funds in those companies, rather than the old funds which invested in companies like Skype and are now reaching maturity.

In any event Om’s point is a good one: if a private company wants to make a big acquisition, that’s a lot easier if your stock is public than if it’s private.

Staying private, then is something which companies might like to do for much longer than they did in the past. But if you’re extremely ambitious and want to grow through the acquisition of large companies, then you pretty much need to be public. Look at Glencore: it desperately wants to buy Xstrata, and the only way it can see of doing that is by going public first.

I’m not entirely clear on why this should be. After all, private-equity companies make enormous acquisitions all the time, and they’re not public. (At least the funds making the acquisitions aren’t public.) It makes for an interesting intellectual exercise to wonder whether Facebook could borrow $7 billion or so to buy Skype, if it were so inclined. But of course it isn’t so inclined: that kind of leveraged buyout makes no sense in Silicon Valley, and Skype would be crushed under such a debt burden. The only remotely sensible way to borrow the money would be if it were a bridge to an IPO, and then at that point you might as well just IPO first.

But the lesson of Skype is that you never know when a big strategic opportunity might arise. And when it does, there will be some part of you wishing that you were public, if only for the option value it confers.

Microsoft Will Announce Acquisition of Skype — All Things D

Dongarra estimates that a dual-core Linpack run on the iPad 2 will yield performance of between 1.5 and 1.65 gigaflops — TUAW

UltraShort 20 Yr Treasury ETF (TBT) performance simulator — Symmetric Info

“Di Tzeitung explained that it has a long-standing editorial policy to never publish photographs of women” — Telegraph

Philadelphia has, per capita, twice as many cycling commuters as any other big city in the US — CBS Local

A (hard) Greek restructuring by the numbers — Alphaville

CJR, in its worthy manner, has come out with a 146-page report entitled “The Story So Far: What We Know About the Business of Digital Journalism”. It includes the story of Gabriel Sherman’s 463-word story on nymag.com about Roger Ailes and Sarah Palin, and Jack Mirkinson’s 237-word distillation of the story for HuffPo. While the former got good traffic from the latter, the HuffPo version was still much more popular than the original.

The CJR report doesn’t examine why the HuffPo version was so relatively successful — the fact that HuffPo’s story is in its headline, for instance, while NYMag went cryptic with “Going Rogue on Ailes Could Leave Palin on Thin Ice”, and didn’t get to the actual news until the third paragraph. But the anecdote packs a punch all the same: if you blather on too long, people are liable to ignore you.

So let me try to find some of the good bits for you, since I know you’re not going to read the whole thing, and especially because the most interesting stuff — at least as far as I’m concerned — doesn’t even begin until page 110. That’s where we find this:

For decades, there has been a connection between the journalism that news organizations provide and the advertisements that generate most of their revenue. Whether it’s a glossy spread that runs before the table of contents in a fashion magazine, or the anchorman’s “more after this message” assurance on the local Eyewitness News, ads and content have always been closely linked in the stream that appears before the consumer.

That linkage is breaking down, and news organizations are scrambling to re- place it with something else. That may mean selling ads on sites they don’t own or control. “Creating content doesn’t ensure a well-sized audience,” says Chris Hendricks, vice president of interactive media at newspaper chain McClatchy Co.“We’re accepting of the fact that the two may be disengaged.”

If you’re going to reinvent the business of journalism for the digital era, this is a really fruitful place to start — the idea that although the business and the journalism are always going to be linked, they don’t necessarily need to be linked through the slightly kludgy old-media mechanism of simple adjacency.

I’ve spent roughly 15 years, now, listening to people come up with a fantastic idea for a website which basically boils down to a single dubious business model: we’ll create great editorial content, loads of people will love that content and visit our site, and then we can sell those people to advertisers by running banner ads.

Given the astonishing number of websites out there trying to make that business model work, it’s a statistical inevitability that some of them will make money — even if it’s usually by means of paying their employees very little and their founders nothing at all. And some subset of those sites can properly be considered journalism, although maybe not journalism of the sort that CJR types generally think of when they hear the word. (I doubt Gawker or Techcrunch are going to win any Pulitzers any time soon.)

When you move away from the ad-adjacency model, however, things get a lot more interesting and exciting. Journalistic organizations have reach and credibility in their communities, whether those communities are geographical, topical, or something else. And that reach and credibility can be monetized in a number of different ways.

One way to do that is to sell ad space not just on your own site but on other sites, too. The ad-sales teams at local newspapers, for instance, tend to have good relationships with a lot of local businesses, who might well want reach beyond the paper and its website. Meanwhile, bigger websites don’t really know who those businesses are, or how to reach them. And when you have a lot of inventory to sell, you can get much more inventive and sophisticated. Consider McClatchy’s relationship with Yahoo:

Because Yahoo has such broad reach, the relationship opens a big market for lo- cal news organizations.“The typical paper has 15 percent penetration in the local market,” Hendricks says, speaking of online operations.“When we partner with Yahoo,it takes us up to 80 percent.”And because many Yahoo ads are“behavior- ally targeted”—meaning they are more closely geared to readers’ interests, based on Web usage habits, geography or demographics—the rates are much higher. But those ads need a lot of viewers to ensure that the subsections of the audience are big enough to interest advertisers.“It’s almost impossible to sell behaviorally targeted ads with 15 percent penetration,” Hendricks says. “With Yahoo’s scale you can.” McClatchy averages an $18 cost per thousand views for targeted ads, Hendricks says.That’s about twice the average for its usual display ads, though it has to share the proceeds with Yahoo.

There are other ways to leverage local relationships, too, like Groupon and its clones: the Minneapolis Star Tribune launched a coupon service, called STeals, while McClatchy both has a Groupon clone of its own and has a deal with Groupon itself.

More generally, publishers can monetize credibility in many ways: Wired has pop-up stores, New York magazine has a bridal show, the Atlantic has a big events business which already brings in $6 million a year and which is growing fast. And on top of that there are various freemium strategies, where free journalism drives demand for exclusive content or expert access.

Meanwhile, the journalism itself has value to various people and can be monetized. Publishers still have a fetish for exclusive content, often because they know a dangerously small amount about SEO. But I do think that we’re only just beginning to see sites which pay relatively modest sums to republish (as opposed to simply link to) carefully-curated content from elsewhere on the web. If publishers are willing to sell the non-exclusive right to republish their content, I think they could make some money doing that.

On top of that, of course, a handful of name-brand journalists can do very well online, monetizing their own personal brand. And much of that value goes to publishers, too. I think the CJR report is wrong about this:

Andrew Sullivan’s Daily Dish was responsible for 1 million monthly unique visitors, or about 20 percent of the traffic, at the Atlantic’s site. But it was Sullivan’s audience, not the Atlantic’s; the blogger owned the brand equity. So when he moved to The Daily Beast in April 2011, he took his unique users with him.

In fact, there’s no evidence that the number of unique visitors at the Atlantic has fallen as a result of Sullivan’s departure. In fact, the Atlantic did a very good job of leveraging Sullivan’s blog in a successful attempt to get a reputation as a respected home for smart, name-brand commentary online. People who like Andrew Sullivan also like James Fallows, or Ta-Nehisi Coates, or other Atlantic bloggers Andrew linked to on a pretty regular basis; they also followed his links to magazine articles and turned out to be pretty appreciative of the rest of the material on theatlantic.com. So while Sullivan did have a very large number of unique visitors, it’s wrong to say that he took them with him when he left. Yes, most of them now read him at the Daily Beast. But that doesn’t mean that all or even most of them have left The Atlantic.

The CJR report ends with a list of recommendations Some of them are platitudinous slogans (“leverage staffs and audience by using aggregation, curation and partnerships with audiences to provide content of genuine value”; “mobile digital devices represent a special challenge”).

Others, however, are smart and sensible. If you’re a legacy media organization, for instance, don’t just shovel your legacy-media content onto the web and expect it to take off: successful websites need to have a webby sensibility, and tend to be separated from the legacy outfit. They also need to push back against the tyranny of the CPM. And don’t expect to make lots of money from a paywall. You won’t. And if you think you will, that’s a good sign you don’t understand how your business works.

Who’s afraid of cannibals? News executives are: there’s a pervasive attitude in the media industry that if you give your content away for free in one place, then people won’t pay for it somewhere else. On this view, if a newspaper loses circulation and has a free website, then the circulation loss can be attributed not to the infinite and wonderful world of other free information available online, but just to the fact that the newspaper’s own content is available online.

This mindset has helped to hobble the online expansion of hundreds of media properties, even when there’s precious little empirical evidence to suggest that it’s based in reality. In fact, the opposite is true: Tyler Green has a fantastic post looking at museum memberships, showing that they generally increased after museums moved to free admission. Economically, that doesn’t make sense, insofar as people buy memberships to save on the cost of repeat visits to the museum. But maybe that’s not the main reason people buy memberships — or maybe people don’t always act in an economically rational way.

A similar economically-literalist mindset exists on the digital advertising side of things: Kendall Allen calls it the “math state,” where everything gets measured and auctioned and quantified, and geeks rise up the greasy pole while creatives find themselves increasingly marginalized.

Today marks an important point in the evolution of digital media: the new iPad app for the New Yorker allows for subscriptions (rather than just per-issue sales), and what’s more allows existing print subscribers to read the magazine on their iPads at no extra cost. I was doing that on the subway this morning, and from an ad-impressions standpoint it’s quite clever. You don’t just scroll a long page to read each story: instead, each piece is paginated, and occasionally the route to the next page is interrupted by an ad. The third page of Anthony Lane’s piece on Pixar, for example, is a Movado ad, while the seventh page is for Castello cheese.

The ads look great on the iPad: even if they’re not clever interactive rich-media things, they still glow in hi-res glory, filling the screen and grabbing attention in a way that banner ads on websites can’t really compete with. If I was an advertiser buying a full-page ad in the magazine, I’d be more than happy to accept this kind of thing as an alternative way of getting my message across. Especially since people who download an issue on their iPad are more likely to read it than print subscribers with a teetering bedside pile of unread New Yorkers growing by the week.

But of course advertisers and media buyers are dragging their feet on this: David Carr quotes MediaVest’s Robin Steinberg as saying that an iPad reader should not simply be considered equivalent to a reader of the paper magazine; Carr’s clear that she wants iPad pricing to be closer to bargain-basement web prices than to premium print prices.

I don’t think that’s really constructive. Magazines economics are based on the institution of the “rate base,” a circulation number guaranteed by the publisher to its advertisers. A digital edition downloaded by an iPad subscriber can and should be considered pari passu to a paper edition, for such purposes — that’s the easy and elegant solution to an otherwise very hard-to-measure problem of how to charge for access to the upscale and desirable iPad audience. iPad ads are no less glossy than their print counterparts, and are even harder to ignore. So let’s hope that advertisers will get out of their math state, and instead embrace a digital world where they can beef up their print buys with exciting and creative implementations for the iPad.

The more that both publishers and advertisers concentrate on the creative side of things, and the less they worry about the distractions of granular economics, the more successful both are likely to be. Digital display and brand advertising is still very young. Let’s nurture it without giving too much authority to the bean counters. If they stifle it now, they’ll end up suffocating the very digital publishers that they’re going to need, in future, as print slowly dies and consumption moves to tablets.

Serena Ng has been keeping an eye on AIG’s share price, which is far below where it was trading at the beginning of the year — and below even where it was in October, when Treasury’s Jim Millstein told me that Treasury was going to make a profit of roughly $13 billion on the money it used to bail out AIG. That’s looking increasingly unlikely: Treasury’s break-even price on its AIG stake is about $28.70 per share, and at current prices it’s going to have to accept less than that if it wants to sell $20 billion of stock into the market.

There are three issues at stake here. First, should Treasury have converted its AIG debt into equity just so that it could exit its position more quickly? Second, will Treasury manage to disentangle itself from AIG at a profit? And third, does that matter?

Governments care very much about the 0% return level on their investments in private companies. If they make more than that, the investment/bailout is considered a success; if they make less, it’s a failure. That’s a bit silly, but the psychology is at least easy to understand.

But if Treasury wanted to end up extracting more money from AIG than it put in, the safe and sure way of doing that would have been to keep its investment as debt, rather than converting it to volatile equity. The problem with that strategy is that the stake couldn’t be sold quite as quickly: for reasons I don’t fully understand, it’s easier to sell $20 billion of AIG stock than $20 billion of AIG bonds.

And one thing that the Obama administration shares with its predecessor is a deep disinclination to have any kind of stake — equity, debt, warrants, anything — in private companies. Treasury hates such things so much that it’s willing to take a higher risk of taking a loss, if that means it can extract itself from companies like AIG more quickly.

That’s an intellectually honest position: after all, the 0% return level is mathematically as arbitrary as any other, and shouldn’t drive government policy. A similar philosophy exists at the New York Fed, as well, which turned down AIG’s offer of a guaranteed positive return on its Maiden Lane II assets, in favor of running a slightly riskier auction which was likely to make more money, ultimately, for Treasury. (Interestingly, Treasury, as the owner of AIG, was the one pushing the Fed to just sell the assets to AIG at a modest profit.)

The big question, of course, is whether the government will really have extricated itself from AIG even once it sells all its shares in the company. One thing missing from Dodd-Frank was a proper federal insurance regulator: the insurance industry is still regulated on a state-by-state basis, and the NYT this morning has a rather alarming story of the way in which various states are competing with each other to see who can be the most lax on that front.

Insurance companies in general, and AIG in particular, are still too big to fail: no government is likely to turn around and tell policyholders that they’re simply unlucky that their insurer ran out of money and went bust. So AIG, along with all other insurers, represents a significant contingent liability for the government. Treasury might be trying to get out of its formal stake in the company as fast as possible. But it can’t get out of its informal links.

Should we be worried about ETFs failing to settle? — CNBC

Great Donald MacKenzie essay on high-frequency trading, algos, and the flash crash — LRB

Greg Mortenson and CAI Roll Out a Defense — Outside

Is Going Public Going Out of Style? — CFO

Hasidic Paper Removes Hillary Clinton From Osama Picture — Failed Messiah

Masnick on Groupon: “claiming that the company is overvalued because the ‘idea’ is too easy makes little sense” — Techdirt

Intriguing pages from Nassim Taleb’s new book — Facebook

Carlos Slim, push-poller — Reuters

Alex Leo on The Four Kinds of NYT Headlines — Tumblr

“Which of the two numbers should you believe? The short answer is the job-growth number” — NYT

Don’t Leak to the Wall Street Journal’s New Wikileaks Knockoff — Gawker

The NYT employs some of the smartest and ablest users and analysts of social media: it’s probably the most sophisticated newspaper in America on that front. And then it has dinosaurs like Bill Keller and Arthur Brisbane, whose respective columns this weekend betray the fact that the people with the bully pulpits are stuck in a completely different world, seemingly ignorant of some of the biggest stories in social media.

Brisbane is the NYT’s ombudsman, and today he describes the way that the paper broke the news of Osama Bin Laden’s death. Well, he can’t do that, because the NYT didn’t break the news of Osama Bin Laden’s death. But he ignores the people who did break the news, and just tells the story of how the official NYT machine worked. His story starts at 10:34 last Sunday night, when a source told NYT reporter Helene Cooper that Osama had been killed. By 10:40, an alert was up on nytimes.com. Then, by Brisbane’s account, Twitter got involved:

One minute after Ms. Cooper’s news alert was posted on the Web, Jeff Zeleny, The Times’s national political correspondent, posted on Twitter: “NYT’s Helene Cooper confirming that Osama Bin Laden has been killed. President to announce shortly from the White House.”

At virtually the same time, Jim Roberts, an assistant managing editor, sent a similar Twitter message. Next to come was an automated Twitter post generated by NYTimes.com, regurgitating the original news alert.

Those links are all Brisbane’s, by the way, including the rather hilarious link to the homepage of the very site his column is on. All of the links are internal; none are to the actual tweets in question. But here’s the first tweet that Brisbane mentions, from Zeleny. As Brisbane says, it was posted at 10:41pm.

For a very different look at how the Osama news broke check out SocialFlow’s exhaustive analysis of 14.8 million tweets on Sunday night. As far as Twitter is concerned, the news was broken by Keith Urbahn at 10:24pm. But it really got momentum from being retweeted at 10:25pm by NYT media reporter Brian Stelter, who added the crucial information that Urbahn is Donald Rumsfeld’s chief of staff. Urbahn, here, gets the goal, but Stelter absolutely gets the assist:

Stelter’s 55,000 followers are extremely influential people in the US media scene, and until Monday’s physical newspaper started landing on subscribers’ doorsteps, Stelter’s tweets were the single most important thing that the NYT published on Osama. Note the timing here: at 10:34pm, when Cooper still thought that Osama had been captured, Stelter had already retweeted Urbahn; had then tweeted that “the whispers about bin Laden are getting louder in Washington circles”; and had then come out with a pretty definitive third tweet, at 10:33pm:

How come Brisbane is ignoring all this? Stelter was way ahead of the rest of the NYT, but Brisbane incomprehensibly discounts his excellent work. That might be because he doesn’t consider tweeting to be part of a NYT reporter’s job; it might be because he doesn’t consider retweeting to be reporting. But Brian Stelter is a prime example proving that neither is true. Brisbane should have taken this opportunity to congratulate Stelter on a job extremely well done. Instead, he is completely overlooked, in favor of tweets from Zeleny and Roberts which came out more than a quarter of an hour after Stelter had publicly jumped onto the case. Which, of course, is an eternity in the twittersphere.

Meanwhile, Bill Keller, the NYT’s editor, has devoted his magazine column to the subject of the newspaper’s war reporters, both staffers and freelance, who are killed or injured in combat zones. Again, anybody conversant with social media knows that there’s an important debate going on around precisely this subject — and that if the NYT doesn’t handle it well, then, in the words of Paddy Hirsch, it “could threaten the company’s brand”.

The debate started on Facebook, between war photographers Teru Kuwayama and Mike Kamber, who wrote a “muted eulogy” in the NYT for photographers Tim Hetherington and Chris Hondros; it then moved on to a discussion board called Lightstalkers, and although that debate seems to have disappeared for some reason, it lives on, for the time being, in Google’s cache. Meanwhile, Teru has put his side of the debate here, at Gizmodo.

Teru’s point is that the NYT spends vastly more money, effort, and resources on Americans and Europeans with names like Tim and Chris than it does on locals with names like Mohammed or Ali or Raza. Paddy puts it very well:

For years there’s been rumbling discontent among journalists about the way media organizations take pains to look after their staffers when they’re caught in the line of fire, but often fail to provide support to the locals who make it possible for those staffers to get the story.

The support those locals give is considerable and invaluable. Anyone who has ever reported overseas knows this. Most reporters who arrive in a conflict zones are like newborn babies. They can’t speak the language, they don’t know what to eat, how to find shelter, or how to get around. They are utterly vulnerable. If they’re lucky, and news people have been in-country before, they’ll have a network of support on the ground: so-called fixers, whose job, on the face of it, is to arrange interviews and get the reporter to the story.

But fixers do a lot more than that. They translate, they find safe accommodations, they know where to find gear. And batteries to power that gear. They find the least dangerous routes to drive, and then they often drive those routes. They know who can help and how to get them to provide that help. They are, in short, architects of an entire network of support for the reporter.

And providing that support is dangerous. Not just because they’re often in the line of fire with the reporter, but because they have to live in the country when the reporter’s job is over. That makes them uniquely vulnerable: if the story the reporter files is unpopular, the local will go after the fixer. If the country the reporter comes from is unpopular, the fixer is regarded as giving help to the enemy.

Fixers are vital to the creation of a good story, and therefore essential to a news organization’s coverage. Shouldn’t the news company therefore treat fixers and their ilk with the same care and attention that they provide the company’ support staff at home? That’s the argument that’s going play out in blogs and stories over the next few months. My question is, as the debate plays out in public, what should news companies do about it?

Keller, in his NYT column, wades into this debate in the most high-handed way possible. He talks at length about Hetherington and Hondros, and about other photographers, like Joao Silva, who parachute in to war zones, meet fixers, get their shots, and then move on to the next job — if they don’t get their legs blown off in the process. He writes movingly about NYT photographers Tyler Hicks and Lynsey Addario, who were brutally treated in Libya but survived; he doesn’t mention their driver, Mohamed Shaglouf, who almost certainly didn’t.

Keller quotes his colleague Greg Marinovich as saying that “sometimes we fail our own moral compass, our own emotional compass.” It’s a resonant quote, for people following Kuwayama’s accusations that the NYT has short-changed the families of people like Raza Khan, who was killed driving Kuwayama and Lynsey Addario in Pakistan:

Raza’s family had modest hopes for compensation or support—they were hoping to get enough money to replace the used Toyota he died in so that his oldest son could carry the family business of driving for foreign journalists. In a single sentence mention, in a blog post about Addario’s recovery, the NY Times mentioned that it was “gathering a fund to give to the six children of the driver, Raza Khan, for whom he was the sole provider”. That fund seems to have amounted to about a thousand dollars, which probably as much was being spent on an hourly basis to provide red-carpet medical treatment to their American photographer, who’d broken a collarbone.

As the debate about the NYT’s responsibility to these fixers rages, Keller’s response is to ignore both it and them entirely, as though neither the debate nor the fixers even exist. Just like Brisbane, Keller makes sure that every single link in his column is an internal one, to some other NYT web page — I count 26 different links between the two columns, which implies that in the eyes of the New York Times, the 26 most important online resources to link to when writing those columns are all NYT stories or pages. It’s as arrogant as it is hermetic.

All of this has to be extremely demoralizing for people like Brian Stelter, who do great work on and with social media and who take pride in linking to news and information wherever it can be found. They’re greatly appreciated outside the company, by people like me. But inside the company, it seems, at least when it comes to the important and visible weekend columnists charged with writing about the NYT itself, anything discussed or reported outside the NYT’s own hallowed pages is probably best ignored.

I had a long discussion at lunch today talking about my theory that it’s just as well the Basel III process was ill-publicized and depoliticized. Because when issues get onto Congress’s radar, the quality of debate can be low indeed. Take this debate between two Democrats on the question of whether private-equity funds should register themselves with the SEC:

Rep. Jim Himes, D-Conn., said that private equity funds act a lot like venture capital funds, which are exempt from SEC registration under Dodd-Frank.

“Private equity entities do not employ leverage any more than venture capitalists do,” he said.

Mr. Himes praised private equity funds for staying in falling markets when others are fleeing.

“They are countercyclical investors,” he said.

Despite Mr. Himes’ support, the ranking Democrat on the subcommittee opposes Mr. Hurt’s bill. Rep. Maxine Waters, D-Calif., said that putting private equity under the SEC’s aegis would better protect public pensions in part because it would subject private equity advisers to a fiduciary duty.

Even by Congressional standards, “private equity entities do not employ leverage any more than venture capitalists do” is pretty spectacularly wrong. “Private equity” is, after all, the polite way of saying “leveraged buy-out”, while venture capitalists don’t use debt at all.

As for the idea that PE shops are “countercyclical investors,” wouldn’t that mean that they did more deals when markets plunged during the crisis? As opposed to, say, this?

On the other hand, it’s a rare argument where you end up siding with Maxine Waters, and I’m underwhelmed with her idea that the purpose of the bill is to protect GPs by ensuring that LPs have a fiduciary responsibility.

The main reason for PE shops to be regulated, of course, has very little to do with fiduciary responsibility, and everything to do with the fact that leverage is a systemically-dangerous thing, and regulators need to know where it is and how it’s being put to use. But it can be hard to explain systemic tail risk to the kind of people who only really understand the meaning of a pie chart when they bake an actual pie.

And remember — this is the Democrats, who tend to be slightly — slightly — more sophisticated about such matters than the Republicans. People like Brad Miller and Barney Frank really do know what they’re talking about. But it’s hard for them to compete with armies of lobbyists intent on dumbing everything down to the point of utter nonsense.