Felix Salmon

How blogging is like being bad at math

Felix Salmon
Jul 2, 2010 22:44 UTC

Just as I was headed down the pub to drown my sorrow at Ghana’s gut-wrenching defeat in the World Cup, the shop steward of the International Brotherhood of Econobloggers instant messaged me to remind me that I still haven’t written about Kartik Athreya. Apparently this is grounds for expulsion, and so I thank Heather Horn, with a smart little essay against the close reading technique often being found in English classes, for giving me an angle. (I’m not even going to attempt a list of everybody else who’s written great stuff contra Athreya, but for starters try Thoma, DeLong, Yglesias, Sumner, Cowen, Kling, Avent, Wilkinson, Konczal, Wade, Merkel, Harding, Evans-Pritchard, and Ritholtz.)

Anyway, Horn’s point is that any organized attempt to look deeply at something risks being self-defeating: you can end up disappearing down all manner of silly dead ends, and understanding less than you would with a more-is-more approach.

This absolutely rings true to me. For reasons which today elude me, I decided when I was doing my A-levels in England to do what they call “double maths” — essentially taking two mathematics exams (Maths and Further Maths), in the same two years you’d normally spend studying for just one. As a result, we had a highly accelerated mathematics curriculum, and there was no time to circle back and make sure the class had understood something before moving on to the next thing. It was all rather sink-or-swim.

And at any given point in time, I was sinking — along, I think, with most of the rest of my class. I was pretty fuzzy about what we’d been taught in previous weeks, and I was very unlikely to understand what the teacher was trying to say at any given time. Maths class, for me, was a combination of panic and incomprehension, combined with a desperate attempt to bluff my way through as much as I could. (Needless to say, if you’re reduced to trying to bluff, mathematics is not the best subject to choose.)

Yet somehow my classmates and I all did very well, at the end of the two years, when it came time to taking the actual exams. As I recall, nearly everybody taking double maths wound up getting an A in their Maths A-level, and most of us got an A or a B in Further Maths as well. Somehow we had managed to gain a pretty good grasp of the subject by dint of sheer velocity: the mechanism, I think, was that a desperate attempt to understand a new concept had the effect of making earlier ideas drop into place. And that the best way of mastering the Maths curriculum was not so much to study it directly, but rather to try to study the Further Maths curriculum: even getting halfway there would bring you pretty much up to speed on the stuff that went before.

Something similar, I think, happens with blogging. Bloggers tend to be foxes, rather than hedgehogs; it’s pretty clear that Athreya is an archetypal hedgehog and has a deep-seated mistrust of foxes. We skip around a lot of different things, and much of the time we don’t really understand them. But somehow the accumulated effect of all that skipping around is to make connections and develop understandings which hedgehogs often lack. What’s more, we live, as Athreya admits, in a highly complex world — one which there are serious limits to what economics can do on its own.

So I’ll continue to have a healthy skepticism when it comes to everything I read, whether it comes from people with deep immersion in economics PhD programs or whether it comes from an anonymous blog. But most of the smart and relevant insights I find will come from bloggers: they might not fully grok the mathematical underpinnings of the economics that they’re talking about, but they are useful and thought-provoking and germane in a way that economists often are not. And by dint of sheer velocity, they achieve a very modern kind of knowledge — one very well suited to the blogging platform. Maybe that’s what I really learned in those mathematics classes — the ability to synthesize bits of information that I’d picked up in the prior weeks and months. It turns out to be quite a handy skill.


I totally agree that most of the time we don’t really get the meaning of what we read. Probably because there’s just so much information everywhere and we’d like to grab most of it, if possible. More than that, we move from a thing to another because they’re so eye-catching and we sometimes forget why we actually clicked on. But you’re also right that this kind of superficial skipping around articles and posts on various blogs helps us make complex connections between very different things, so who knows what innovative idea or discovery could come up to our filled up brains just before falling asleep…

Posted by Len.Williams | Report as abusive

Fixing information asymmetries

Felix Salmon
Jul 2, 2010 17:23 UTC

In the wake of my blog entry yesterday about the way in which Goldman’s interaction with AIG helped to exacerbate the financial crisis, I got a very interesting email from a former Goldman employee:

Insofar as the law is never going to be perfect, it’s certainly good that we are nice to our neighbors even when the law doesn’t require it, but when you find yourself saying, “Yes, this action was legal and within the rights of the agent, but helped precipitate great social harm,” you’re looking at a good candidate for a change in institutions, and legal institutions, for all their shortcomings, are the easiest ones to change. Ratings-based collateral deals might be a good target for regulation; the big liquidity hit that a credit downgrade can cause feels like exactly the sort of creditor-run that bankruptcy is supposed to prevent.

I like the idea that contracts concerning collateral should not reference credit ratings. In general, the less standing that credit ratings have, both legally and in private bilateral contracts, the better. They’re invidious things: no one should use a triple-A rating as a business model (see GE and the monolines), and a ratings downgrade in and of itself shouldn’t be able to precipitate a liquidity crisis: such things should never be self-fulfilling, since that just introduces needless systemic dangers into the financial architecture.

The former Goldman employee continues:

Should I have tried to call Treasury/SEC/some NY insurance regulator when it was obvious from our standpoint that AIG had no idea how much trouble it was in? When reminded of my duties to Goldman clients I basically translated that in my head to “Don’t short AIG.”

If Goldman employees reckoned that shorting AIG was a breach of corporate ethics, it’s hard to justify the fact that Goldman was happily shorting AIG for its own account.

As for the first question, back in October, Lloyd Blankfein said that Goldman should have been in more contact with regulators. But he seemed to put the onus on the regulators, rather than the banks:

We have to build a culture whereby firms are required to share concerns about systemic risks with regulators…

Regulators could establish a multi-firm business practices committee to examine issues such as underwriting standards. If practices slip, regulators would be among the first to know. They should ask questions such as: “Where are policies being stretched and pressures building? Where are you seeing concentrations in risk, crowded trades or one-way bets?”

I worried at the time that there would be big conflicts between banks’ responsibility to share information with regulators, and their duty of confidentiality to their clients. But it’s also increasingly clear to me that it’s silly to expect regulators to know exactly what questions to ask. If any employee of any bank sees a systemic risk, that person should have the legal obligation to take their concerns to the systemic risk regulator.

More generally, the problem here is that Goldman makes money from knowing or seeing stuff that other people don’t know or don’t see. It saw problems in the housing market which AIG was oblivious about, and took advantage of the information asymmetry there to make money. That’s how markets work. At some point, however, it risked setting up a negative feedback loop in which AIG would lose so much money that it would get downgraded and then be forced to put up even more money and then be downgraded further, etc etc. And I think that regulators should be within their rights to tell banks not to put on systemically-dangerous trades. Bringing down a small hedge fund is OK. Bringing down AIG, by contrast, ultimately benefits no one. As Goldman is slowly finding out.


KidDynamite, I think more importantly than the narrow affect of CDSes is the general chaos there would have been in the market if suddenly none of those counterparties had CDS protection on their CDOs and suddenly had to to take that hit to their balance sheets whilst the money markets were frozen due to LEH bankruptcy.

From recollection, a significant portion of the hedging was short equity positions against AIG, which of course they are being demonised for now.

HBC, are you capable of posting links that are not clearly BS? If you had even a passing acquaintance with any of the markets you seem to pontificate on you’d know that in 2007-2008 the commodity markets were bubble due to massive stockpiling by a number of countries – Philipines in particular with regard to rice – and “green” legislation on ethanol based fuels which pushed out other crops at the time. But hey why bother with facts when you can blame Goldmans?

Posted by Danny_Black | Report as abusive

Employment falls

Felix Salmon
Jul 2, 2010 12:56 UTC

It’s never good when employment falls during what’s meant to be an economic recovery. It’s worth remembering that, if people start getting excited about today’s drop of 125,000 in the total-employment number. Yes, private-sector hiring was marginally positive, by 83,000, but we’d all like to see much bigger numbers than that.

Unemployment dropped sharply, to 9.5%. But why? It doesn’t seem to be thanks to people getting jobs: after all, employment fell. And the labor force participation rate — the number of employed people divided by the total number of people capable of working — hit another new low today, of 64.7%. If people are just giving up and removing themselves from the workforce, then a falling unemployment rate only serves to hide the bad news. What’s more, the only important statistical decline in the unemployment rate was among white women, who already have lower unemployment than just about anybody else. The rest of the country — including, crucially, men overall — was pretty much unchanged.

As far as markets are concerned, however, this report had better be good. Shrinking employment and 9.5% unemployment are nobody’s idea of a healthy economy, but as Barry Ritholtz notes, “whisper numbers” for today’s payrolls report have been extremely gruesome and have contributed to the big sell-off in recent days. If the bearishness continues in the wake of these payrolls numbers, the message from the markets is that there’s still really nobody out there looking for an excuse to spend money — either on equities or on employees.


Another case of lies, damned lies and statistics, I think: definitions, workforce sizes, manipulating sectors….all governments do it. They call it spin, but actually it’s lies.

There are two fascinating commonalities between the US and UK at present.
First, neither private sector can soak up the public sector jobless. And second, everyone talks about ‘double dip’, when it’s hard to see any sign in the real data that anything went up in the first place without massive QE.

There is a third area too – massive debt. So we’re both between a rock and a hard place.

This is a great blog by the way, Felix – not just well-informed, but very creative in its interpretations.
Keep up the good work.

Posted by nbywardslog | Report as abusive


Felix Salmon
Jul 2, 2010 04:32 UTC

Winemakers are nearly always white. Can South Africans step up and change things? — Atlantic

Elliott Associates has one of the “veritable flotilla of special-interest bills is quietly floating through the Legislature” in Albany — NYT

Q: So should I see the film in 3D or 2D? A: Do you have a third option? — VF

Best acquisition announcement ever — YouTube

Why Ghana can beat Uruguay — Twitvid

IBM moving to Firefox as its default browser — Sutor

“Smoking and drinking alcohol are the biggest risk factors for oesophageal cancer, particularly if they are combined” — NHS


My first lesson in the danger of statistics came about as a result of oesophageal cancer. My mother was diagnosed with the squamous cell variant, and our personal research led us to believe that she didn’t stand a good chance of making it beyond two years, as the article says. Of course, she was not over 55, a smoker, or a heavy drinker, so something seemed off. She is, however, Japanese. A good doctor parsed the information, and it turns out Asians have a much higher incidence of squamous cell carcinoma (as much as 81% more likely in Asian males than white counterparts, in some studies). If you could sound an alarm, it would be for Asian/Pacific Islanders to routinely get screened after 40 (not 55) for oesophageal cancer. In many Asian countries they routinely screen after 40, but expats (and US doctors) don’t often know this. This isn’t to say the NHS article is wrong, but correlating two factors with an increased chance of cancer without exploring racial differences might put some at (false) ease. Additionally, for squamous cell carcinoma, radiation is used quite often. Hope I don’t sound contentious, but early screening and knowledge can save lives. You’ll be happy to know the rate of survival among Asians is siginificantly higher.
Study link: http://www.ncbi.nlm.nih.gov/pubmed/16388 522

Posted by OneTime | Report as abusive

Communism and the financial crisis, cartoon edition

Felix Salmon
Jul 2, 2010 01:55 UTC

Remember David Harvey, the chap with the book-length critique of the financial crisis through an explicitly Marxist lens? I’d be surprised if the number of readers of this blog who read his book ever broke into the double digits, but hey, what if I told you that there’s a fabulous little YouTube video from RSA Animate which illustrates a lecture that he gave with inventiveness and verve? That’s more like it:

This is one of the most gripping ways I’ve yet seen of presenting complex and dry ideas: it’s Paddy Hirsch with better drawing skills, higher production values, and two months rather than two hours to put it all together. There’s more where that came from: try Barbara Ehrenreich, Dan Pink, or Jeremy Rifkin. I’m trying to think who I’d most like to see given this treatment: Matt Ridley would be very interesting, I think, or Paul Collier, or — getting very geeky and financey — how about Riccardo Rebonato? Even Nouriel Roubini might be a lot of fun. More, please, RSA!

(HT: NC)


Oh, that was just so enjoyable! Thank you so much.

Posted by JohnBrookes | Report as abusive

The financial crisis as PR tool

Felix Salmon
Jul 2, 2010 00:20 UTC

There’s been an awful lot of weird and unusual PR surrounding The Zeroes, by Randall Lane. My copy has a big white sticker on it: EMBARGOED, it says, until Tuesday June 29 — as though there is hot breaking news buried somewhere in the book. (It’s definitely not in the Vanity Fair excerpt, which came and went with barely a ripple of notice.)

Lane himself maybe hit on a slightly newsier story with an excerpt at the Daily Beast, where he’s “editor-in-large”, whatever that’s supposed to mean. In an article timed to coincide with the book’s publication, Lane talks about Lenny Dykstra, who may have made some slightly sleazy deal with a small company in California:

In the late winter of 2008, an entrepreneur named Richard O’Connor, who had become Dykstra’s favored adviser, introduced him to Shannon Illingworth, the founder of a publicly traded company called Automated Vending Technologies, or AVT, and the two quickly cut a deal. O’Connor told me that on March 25, 2008, Illingworth gave Dykstra roughly $250,000 worth of AVT stock in exchange for plugging the company on Cramer’s website, TheStreet.com, and promising to provide a personal introduction to Cramer.

This is a bombshell of utterly minuscule proportions, especially since Dykstra disclosed his long position in his column. And neither Lane nor O’Connor are particularly reliable sources when it comes to these things: Lane happily admits to lots of reasons why he has an animus towards Dykstra, while O’Connor, according to Teri Buhl, owes AVT $37,569.82.

But this non-story is getting a new lease of life — even making the New York Post — thanks to a threat from AVT to sue Lane for defamation. And lo, the lawsuit has now appeared — complete with an utterly stupid claim for $100 million in damages. Here it is, from Illingworth via Buhl:

AVT vs. Lane

This is basically all bluster and bullshit, written in particularly bad legalese which doesn’t even seem to have been proofread. There’s lots of talk of “untrue and false statements” but at no point does the plaintiff, AVT, bother to actually say what those libelous statements might be, nor how they might have caused anything like $100 million in damages. (The market cap of AVT is just over $30 million, and it hasn’t moved significantly in months.)

Whenever anybody launches a substance-free lawsuit with a stupidly huge number like $100 million attached, you know that they’re mainly interested in publicity. But this is where things get weird: why would Illingworth and AVT want to give free publicity to the book they claim is libeling them? And why would they carefully time the lawsuit so that it coincides exactly with the publication of the book?

I’m sure that both sides are loving whatever extra publicity they can extract from this lawsuit and I feel a little bit sleazy adding to that publicity myself. This is one of those spats you don’t particularly want either side to win. But I guess the good news is that we’ve come through the first tranche of serious and important crisis books and now we’re seeing the second tranche, where the crisis is used as an excuse for failed hype merchants to try yet another desperate attempt at self-promotion. The sleazy and litigious side of the American Dream, it seems, is impervious to crisis. Maybe in fact it thrives on it.


“editor in large” sounds like a fat joke.

Posted by drewbie | Report as abusive

Goldman’s collateral demands and the financial crisis

Felix Salmon
Jul 1, 2010 19:31 UTC

Daniel Indiviglio is coming over all faux-naive with respect to the FCIC’s grilling of Goldman executives over its aggressive demands for collateral from AIG:

The FCIC’s decision to waste its time investigating this question is troubling, because it has nothing to do with its mission of determining what caused the financial crisis. Goldman demanded collateral from AIG because it’s the job of an investment bank to demand cash from a derivative’s counterparty when asset deterioration occurs. The only explanation for spending two days questioning the bank about this issue is a desire to mimic other politicized pursuits of using Goldman as Washington’s favorite punching bag. The commission should focus on real causes of the crisis, not convenient targets.

The fact is that the financial crisis had many moving parts, all of which were interwoven with each other, and one of the central ones was the collapse of AIG. It’s reasonable to say that if AIG had not collapsed, then the crisis would not have been as bad as it was — and it certainly wouldn’t have been nearly as expensive as it turned out to be for the US government.

Was Goldman only doing its job when it started demanding ever-increasing amounts of collateral from AIG? Certainly, yes. But in doing so, it helped to bring down a venerable insurance company, and cost the US taxpayer the best part of $100 billion in bailout funds. Some of that money will end up being repaid, with interest; the chances are that much of it won’t be. So this is a perfectly legitimate subject for the FCIC to investigate.

Essentially what happened at AIG was that it bought subprime assets high and was forced, by the government, to sell those same assets low. In doing so, it lost so much money that it had to get bailed out by the government.

Now if the government was going to put up all that money anyway, why not simply lend it to AIG and then let AIG post it as collateral? Or better yet, just provide a government guarantee on AIG’s CDS exposures, in return for a fee, which in turn would keep AIG’s counterparty risk at AAA levels, and would mean that AIG didn’t need to post any collateral at all.

Eventually, AIG would probably lose money on its CDS exposure — but as we’ve seen, those positions haven’t lost money yet. And losing money at some point in 2011 or 2012 or later is vastly preferable to panic-selling at the height of the crisis, which is what we ended up doing.

Goldman Sachs, of course, was on the opposite side of these trades to AIG, and if AIG bought high and sold low, then Goldman bought low and sold high. The credit protection Goldman bought from AIG was dirt cheap, when the deals were initially entered into. And then, at the height of the crisis, Goldman unwound all of those CDS deals for cash, at valuations which were extremely attractive to Goldman Sachs and extremely unpleasant for AIG. Absent that deal, Goldman could have gone to the open market and tried to sell credit protection on those assets in order to lock in its gains, but it’s far from clear that it could have found enough willing buyers to be able to do so.

Meanwhile, of course, a key Goldman alumnus, Dan Jester, was in charge of the AIG problem at Treasury, while his boss, Hank Paulson, was a former Goldman CEO. As they watched Goldman apply ever-increasing amounts of pressure to AIG, in the form of collateral demands, their solution was for AIG to give in to all of those demands, and eventually for the US government to end up writing an enormous check to Goldman Sachs, in the knowledge that AIG would be unlikely to repay the money.

Yes, in a capitalist system it makes sense that Goldman would push hard for as much money as it could get its hands on. But its actions undoubtedly exacerbated both the financial crisis generally and the cost to taxpayers specifically. If the AIG insurance contracts had simply been allowed to stand and slowly unwind, rather than being torn up at great expense to the government and concomitant benefit to Goldman Sachs, the amount of damage done would have been much smaller.

And indeed, the biggest weakness of AIGFP’s CDS operation, as explained by Joe Cassano, was that it only ever concerned itself with total eventual possible losses, rather than with the path that asset values might take along the way. If there was a panic big enough to justify a downgrade of AIG’s triple-A credit rating, that could drain the company of vital liquidity even if the CDS insurance policies weren’t paying out any money at all. Cassano’s failure to model or worry about such matters is inexcusable. But once the government ended up being responsible for AIG, it had every opportunity to help AIG retain its triple-A rating and thereby deny Goldman its collateral, putting off any possible losses until years into the future.

So it does make sense to say that Goldman’s demands for collateral — along with the government’s acquiescence to those demands — were a central part of the narrative of the financial crisis. I’m glad that the FCIC is looking into them, and I trust that it’s thinking carefully about ways in which they could have been dealt with better.

Update: Indiviglio responds, and as far as I can tell essentially concedes the point, retreating to a position which involves saying no more than that Goldman didn’t “single-handedly bring down AIG.” He seems to think that if Goldman didn’t single-handedly bring down AIG, and if it acted legally, then the FCIC has no business looking at what it did. Which is ridiculous on its face.


HBC, GS never got even close to 100 billion from the government via AIG. I know facts mean zip to you but on the off hand chance you feel like not embarassing yourself in the future:

Par value of CDOs insured by AIG for GS and GS clients:
roughly 20billionUSD

Market Value of CDOs at time of AIG collapse:
roughly 10billion USD

Collateral held against the CDSes before the collapse:

Direct counterparty exposure to AIG:
2.5billion USD

At the most aggressive valuation GS got a grand total of 12.5billion – assuming their CDOs went to zero which they didn’t – which whilst it is not peanuts is nowhere near 100 billion but why confuse a rant with facts?

Posted by Danny_Black | Report as abusive

Adventures in financial literacy, Mint.com edition

Felix Salmon
Jul 1, 2010 16:10 UTC

Aaron Patzer, the founder of Mint.com is pretty economically illiterate, so maybe it’s not so surprising that his site is running the headline “What Inflation? Products That Cost Less Today Than in 2000″ over a chart of things which cost more today than in 2000. If there was a prize for the most incoherent piece of consumer financial journalism, Ross Crooks would surely be a finalist for this:

Not all prices have increased over the past ten years. As WalletPop.com recently found, certain items or services are actually cheaper today compared with their 2000 price adjusted for inflation through 2010.

Yes, Mint.com really is comparing 2010 prices to 2000 prices after adjusting the 2000 prices for inflation, and finding that some prices have gone down. Which of course is true by definition, since inflation is the average rate at which prices have been rising, and some items will rise in price more slowly than others.

WalletPop itself does a slightly better job of presenting its data: at least it shows, in a table, the handful of items which have fallen in price in nominal terms. (Hummel figurines, Martini and Rossi Asti Spumante, and 100 Bayer aspirin, you win!) But the illustration the two sites collaborated on is pretty bad. Not only do they adjust the 2000 prices for inflation, which largely defeats the purpose of the exercise, but they also exaggerate the difference in real price by measuring on one dimension and illustrating on two. For instance, here’s how they show how the price of CDs has fallen from $14.04 to $13.02, a drop of just over 7%:


Is it interesting that the price of CDs has fallen by 27% in real terms? Maybe. I don’t know. But if you compare the size of the green 2000 CD and compare it to the size of the brown 2010 CD, you’ll see that visually the decrease in size is not 27% but 47%.

In other words, to illustrate a price drop of 7% over 10 years, Mint and WalletPop first exaggerate the fall by putting it in real terms and making it 27%, and then they exaggerate it again by illustrating it with a CD which is 47% smaller than the one representing the price 10 years ago. And these are sites which are supposed to help in the quest for improving financial literacy. Sad.


Prof. Mark Perry at the “Carpe Diem” econo-blog has some good posts on the topic – he looks at things like number of U.S. labor hours required to purchase semi-equivalent items over large time frames, or the percentage of income spent on food, clothing and housing over time.

His very interesting point, obvious once pointed out, is that while many (including me) complain about stagnant wages, declining middle class, etc., we fail to remember that lower income folks this country today have luxuries that the rich didn’t have just a generation or two ago. That doesn’t mean that growing income disparities are a good thing, but does remind that a rising tide floats most boats.

http://mjperry.blogspot.com/2010/06/good -old-days-are-now_22.html

http://mjperry.blogspot.com/2010/06/good -old-days-are-now.html

http://mjperry.blogspot.com/2009/11/good -old-day-are-now-its-getting-better.html

Posted by SteveHamlin | Report as abusive

The Dow yields more than 10-year Treasuries

Felix Salmon
Jul 1, 2010 15:37 UTC

Calculating the dividend yield on stock indices is more of an art than a science, since no one knows for sure exactly what dividend the stocks in any given index are going to pay over the next year. But one good estimate puts the dividend yield on the Dow at 3.1%, while Eddy Elfenbein has calculated it to be 2.9%. Either way, it’s higher than the current yield on the benchmark U.S. Treasury bond, which was last seen at 2.89% and falling.

What this means is that if the Dow’s stocks, and their dividends, go absolutely nowhere over the next 10 years, they will still outperform Treasury bonds. Which doesn’t necessarily make either asset class a good investment: it’s entirely possible that both stocks and bonds are going to go down rather than up over the next decade. But it does say to me that stocks are increasingly attractive, on a relative basis, when compared to bonds. And if you’re valuing stocks on some kind of discounted-cash-flow basis, then your valuations should be soaring right now, as long-term interest rates continue to fall. Which probably just demonstrates the limitations of DCF analysis more than anything else.


If you like the dividend yield now, just wait six months. It’s about to become _awesome_. Or at least that’s my bet.

Posted by ckbryant | Report as abusive