Felix Salmon

Tal Yarkoni on Dunning-Kruger; he goes on to debate Dunning in the comments — Yarkoni

A good response to the UK govt’s request for “laws or regulations you’d like us to do away with” — Gov.UK

The Volcker rule doesn’t ban all prop trading. You can still speculate in the Treasury market — WSJ

Koblin on Gaby Darbyshire — NYO

Felix Goltz and Véronique Le Sourd have an interesting paper (PDF) revisiting the question of why so many of us invest in capitalization-weighted stock indices. It turns out that the theory behind our behavior is pretty weak: first of all, you have to believe in the capital asset pricing model, or CAPM. The CAPM includes lots of assumptions which don’t hold in the real world: that all investors have the same risk appetite, for instance; that they all have the same investment horizon; that they can short securities freely; that they pay no taxes or transaction costs; and that all assets can easily be traded, including assets such as human capital and real estate.

If you make all those assumptions, then the CAPM says that the market portfolio is efficient — the market portfolio being, essentially, everything in the world: stocks, bonds, real estate, commodities, human capital, art, social security benefits, automobiles, everything. But the problem, of course, is that cap-weighted indices do a pretty bad job even of reflecting the performance of the stock market as a whole, let alone all global assets. (The world’s assets have grown a lot in the past decade; the S&P 500, not so much.)

The authors conclude:

In view of these arguments, it seems that financial theory alone does not justify the current practice of cap-weighting. In fact, from a theoretical perspective, cap-weighted stock market indices seem to offer no particular advantage.

So why are cap-weighted indices so popular? One reason might be that indices seem to outperform a simple buy-and-hold strategy. And another is that they’re easy to understand and most of them have been around for a long time. Still, it’s worth noting that the most famous stock index in the world, the Dow, isn’t an index at all, and it certainly isn’t cap-weighted. (Although with the Dow, you would have been better off just holding the original 30 stocks than following the vicissitudes of DJIA itself.)

The main reason for buying cap-weighted indices, I think, is that it’s easy and it’s cheap. (That’s probably the main reason not to buy cap-weighted indices, too, since anything easy and cheap is likely to get crowded.) Of all the assets in the world, stocks are the easiest to invest in and the easiest way to invest in stocks is to simply buy the index. I suspect there are many portfolios which do a better job of simply “investing in the world” than the S&P 500 does. But once you take into account the costs of putting them together, it’s probably not worth it.

David Goldman has reacted with a curious mixture of alarm and reassurance to Dakin Campbell’s story about U.S. bank holdings of municipal debt:

If municipal debt actually defaulted, the capital position of the banking system would be impacted, bank preferred debt might stop paying, and the holders of bank preferred debt–starting with the insurers–would be in serious trouble…

Why buy munis? For all of Warren Buffett’s dire warnings about municipal finances, the fact is that the federal government can’t let major municipal debtors (at the level of states, for example) go under without also bringing down the banking system and everything else.

If it goes, it all will go together. That’s why munis ultimately will be bailed out.

This is altogether far too sanguine. And if you look past the alarmist headline that Bloomberg has put on Campbell’s story, and the out-of-context numbers in his first few paragraphs, he eventually reveals just how much of a non-issue muni debt really is to the banks:

Lenders hold just 8 percent of the $2.8 trillion state and local government debt market, and municipal bonds are only about 2 percent of total bank assets, according to the Fed.

Muni bonds are relatively safe for two reasons. Firstly, they very rarely default; and secondly, when they do default, they generally have very high recovery values.

But let’s get ultra-pessimistic here, and say that 25% of municipal bond issuers will end up defaulting, and that recovery on those bonds will be just 50%. Then banks would have to take a hit of 12.5% on their muni bond holdings, which would correspond to a hit of about 0.25% of their total balance sheets. Needless to say, that’s not the kind of event which would precipitate a default on their preferred debt.

A widespread municipal default would be harmful to the economy more generally. With $2.8 trillion of munis outstanding, a hit of 12.5% would mean $350 billion of losses, spread across individual investors who were looking for safe, tax-free investments, and a lot of monoline insurers. That kind of thing can hurt. But investments in municipal bonds tend not to be leveraged, and for long-only investors, a drop of $350 billion is equivalent to roughly a 2.5% wiggle in the level of the U.S. stock market.

So I don’t think that munis pose a major systemic risk in and of themselves, although a handful of them — California first and foremost — are probably too big and politically important to be allowed to fail. There will certainly be a lot of wailing and gnashing of teeth if munis do start defaulting, since they’ve long been sold as extremely safe investments, and because they’re largely held by individuals rather than institutional investors. But no one’s going to bail them out because they’re worried about the banks.

Josh Tyrangiel became arguably the most sought-after editor of his generation by boosting Time.com’s pageviews from 400 million to 1.8 billion within three years, and by successfully transforming a bunch of grizzled old magazine journalists into web-speed multimedia content producers:

“Getting Time magazine to be a daily operation? It required me to be my most charming, scheming and belligerent.”

Now that Tyrangiel has left to BusinessWeek, however, Time seems determined to roll back all of his achievements:

We’ve said for awhile that increasingly we’ll move content from the print (and now iPad) versions of our titles off of the web… Our strategy is to use the web for breaking news and ‘commodity’ type of news; (news events of any type, stock prices, sports scores) and keep (most of) the features and longer analysis for the print publication and iPad versions.

If the 1990s saw news organizations set up massive parallel online operations, then, and the 2000s saw the integration of the online operations with the legacy operations, then is this the beginning of the 2010s backswing, where the two become bifurcated again?

My guess is that the answer is no, and that this is just a case of Time making a tactical decision which makes no strategic sense. It wants to sell lots of copies of its iPad edition at $5 a pop, but it’s only putting the magazine content into the app, if all of that can be read on the iPad for free just by firing up the web browser, it fears that sensible consumers won’t bother. So rather than improve the iPad app and make it worth the money, Time is artificially crippling its website.

So long as the iPad app remains broken, however, this idea is doomed to fail. People read the magazine in one of two ways: either they subscribe, at a significant discount to the cover price, or else they buy the magazine at a newsstand, where they have the opportunity to browse through it first. Neither is possible on the iPad, which sells issues only one at a time, and which gives no tasters of what’s inside, just headlines. (The app does have some good free content, if you find the hidden button in the bottom right hand corner, but it’s exactly the same free content that’s available on the website.)

My guess is that none of this would have happened had Tyrangiel stayed at Time, but that once he left, the heart of the website that he helped to build was doomed. It won’t be long, I’m sure, before Time’s journalists settle happily back into their weekly routine, and the website is left to a very different team of people, producing very different content. Which is clearly not a sustainable model.

The problem of falling wages for people without a high-school diploma is well presented by Richard Florida, in an op-ed headlined “America needs to make its bad jobs better”:

The problem is that on average, service workers earn only half of what factory workers make – and only a third of what professional, technical and knowledge workers are paid. The key is to upgrade these jobs and turn them into adequate replacements for the higher-paying blue-collar jobs that have been destroyed.

I’m less impressed with Florida’s proposed solutions, such as they are.

He first points to a handful of companies (Whole Foods, Zappos) which pay more than average for hourly workers, although they don’t pay anything like the sort of money that blue-collar factory workers can command. But it’s simply a statistical certainty that some companies will pay high wages and be successful, just as others (like Wal-Mart or most hotels) will pay low wages and be successful, and others still will fail no matter what they pay. Demanding that the entire service sector should gravitate to one particular quadrant is, I think, unhelpful and unrealistic.

Florida also reckons we can apply some smart technology here:

Service jobs are the last frontier of inefficiency, providing abundant low-hanging fruit for the innovation and productivity improvements that can undergird higher wages.

Florida wants a service-sector equivalent to the kind of technical assistance that the government has long provided in manufacturing and agriculture. It’s not a bad idea, but it’s harder to implement in the service sector, because employers tend to be smaller and more heterogeneous, and because technical assistance aimed at a broad range of service-sector employers risks becoming a series of bland management mantras rather than anything specific and actionable.

What’s more, productivity improvements don’t necessarily result in higher wages for the less-skilled: they’re just as likely to result in greater returns to capital, as owners extract more profits from the business, or else to result in the jobs going to better-educated workers instead.

So while it’s undeniable that America needs to make its bad jobs better, it’s also, I fear, something which is too difficult to succeed at — certainly for any government bureaucracy. If it’s going to happen at all, it will happen from the bottom up, rather than from the top down. And so far there’s zero evidence that’s happening.

Catherine Rampell features this chart today, showing how wage inequality has increased over the past 30 years, especially for men. But in fact what we’re seeing here understates how bad things have been for most men over the past generation. If you go to the source, this chart only shows data for people working full time. And, at least when it comes to men, that’s much less common now than it was in 1979.

The labor force participation rate for men 20 years and older was 79.8% in 1979; today, it’s just 74.4%. And I don’t think that most of that drop can be explained in terms of a larger number of students: the rate was as high as 77% as recently as August 2000, and then dropped to a low of 73.9% in December 2009.

You can be sure that most of the drop in labor force participation is coming from the less well educated Americans. Which means that if you’re a man with less than a high school diploma, your real wages have fallen by 28% over the past 30 years if you’re lucky enough to have a job at all. At the same time, the number of such men without a job has been growing steadily. It’s a depressing set of data, and there’s no sign of it turning around in the foreseeable future.

The Time Mag paywall: it’s back, and higher than ever — Nieman Lab

Robert Sullivan’s fantastic story on bringing bus rapid transit to NYC. This can really happen, people! — NYMag

Record Low Yields for the 2-Year U.S. Treasury Note — Vix and more

Memo to John Ioannidis: you’re popular with some extremely rich and important people! Sebastian Mallaby reports on RenTech:

Simons’s faculty of quants does not think like the rest of the financial industry. It does not hire people from the rest of the financial industry, either, and has little in common with academic finance. For a while Simons’s scientists picked through finance journals, looking for ideas that could be traded profitably. But they soon concluded that none of the ideas worked. When I visited Simons’s Long Island campus, a star statistician had stuck an article to his office door. “Why Most Published Research Findings are False,” the title proclaimed, summing up the faculty’s disdain for the crowd’s wisdom.

The Ioannidis paper is worth reading, and not only because it’s presented, for free, in a beautiful HTML (rather than PDF) format. (If it’s all a bit wonky, here’s Alex Tabarrok’s attempt to translate it into English.) Among his headings:

• It can be proven that most claimed research findings are false
• The greater the flexibility in designs, definitions, outcomes, and analytical modes in a scientific field, the less likely the research findings are to be true
• The greater the financial and other interests and prejudices in a scientific field, the less likely the research findings are to be true
• The hotter a scientific field (with more scientific teams involved), the less likely the research findings are to be true
• Claimed Research Findings May Often Be Simply Accurate Measures of the Prevailing Bias

All of these things, of course, apply in spades when it comes to finance. What’s not at all obvious is why anybody at RenTech thinks that they’re exempt from Ioannidis’s findings — especially given that RenTech’s very own Bob Mercer told Mallaby that “the signals that we have been trading without interruption for fifteen years make no sense.”

Mallaby is convinced that this band of science-hating scientists constitute the best possible group of risk managers, and that it’s unconscionable to interfere with what they do in any way:

Hedge funds have nurtured a paranoid and contrarian culture that makes them good custodians of risk—not perfect, but certainly superior…

If lawmakers understood the virtues of hedge funds, they would have written a bill that actively promoted them. Instead, their legislation will hamper Simons and his followers in myriad small ways, forcing them into pointless registration with the Securities and Exchange Commission. Tragically, a basic truth is being missed: Investment risk is not going away, and the best way to handle it is to entrust it to hedge funds.

I fail to see why it should hurt RenTech to register with the SEC; even Economics of Contempt concludes that although the registration requirements are a bit too fuzzy and capricious, “on net, in spite of the epically bad drafting, this could end up being a net positive for the financial system.” Certainly RenTech has more than enough money to be able to hire a couple of extra compliance officers if it has to. And registration with the SEC is not “pointless”: indeed, it’s the only way in which systemic-risk regulators can try to look directly at these key market players, with an eye to seeing whether unhealthy concentrations of risk are being built up across a group of large institutions. They might not succeed, but at least it’s worth at try.

Besides which, I’m not at all convinced that the best way to deal with investment risk is to start paying billionaires 2-and-20 to manage your money for you. For a good example, look at RenTech itself: the funds which are available to the public, RIEF and RIFF, have dropped to $6 billion of late, down from $30 billion in 2007; they might be closed down altogether. Clearly, RenTech’s management are better at enriching themselves than they are at building a long-term franchise for stewarding other people’s money.

James Saft today quotes the “six ways to dig oneself out of a debt hole” of Jeffrey Gundlach. Boiled down, they are

1. Growth
2. Lower interest rates
3. A money transfer from an outside benefactor
4. Higher revenues (taxes) and/or lower spending
5. Printing money
6. Default.

He’s too polite to mention the seventh option, which is to lie about how much debt you have and hope the markets don’t notice.

It’s worth bearing this list in mind in light of what David Merkel has to say about sub-sovereign debt:

I have long said that the health of the states is a more valid measure of the health of the nation than looking at national statistics. Why?

• The states can’t print money, or force ask allow the central bank to buy their debt.
• In general, the states must run balanced budgets. (Would that we constrained the Federal government to do the same through amending the Constitution. Somebody bring that up after the crisis is over, please?)
• State statistics are more reliable than Federal statistics, because they serve fewer political goals.

John Dizard seems to be thinking along similar lines, saying that Greece has already started restructuring its debt, on the grounds that the state hospital system is imposing haircuts on its creditors.

It’s only natural to look at sub-national defaults and near-defaults, from entities like Greek hospitals or the state of Illinois, as indicative of a broader fiscal malaise. If nothing else, it’s a sign that the sovereign is either unwilling or unable — or both — to bail out the troubled borrower in question. And as Merkel says, sub-sovereign defaults are “purer” than their sovereign counterparts, since a lot of the tricks available to the sovereign are inaccessible to anybody else.

I don’t think it’s fair to say that problems with Greek hospital debt mean that the sovereign has already defaulted, any more than non-payment from Illinois means that we’re in the middle of a US default. But I do think that default will become more common among sub-sovereign debtors, and as it does so, both creditors and debtors will start considering it seriously among the menu of options available. When nobody’s doing it, default is unthinkable. But once it starts popping up all over the place, it becomes a strategic option. That’s true of homeowners, who are more likely to default when their neighbors are doing it too, and it’s true of sovereigns as well.

CMA Datavision has released its sovereign risk report for the second quarter of 2010. It’s based on CDS prices, and it makes for fascinating reading. It was a bad quarter for sovereign credit: 93% of sovereigns widened, and they widened a lot — by 30%, on average. Spreads in France, Portugal, and Spain all more than doubled, while Greece soared from 346bp to 1,003bp, at one point becoming the riskiest sovereign in the world, trading slightly wider than Venezuela. Only one country saw its CDS spreads tighten in significantly over the second quarter: Iceland tightened in by 17% to 330bp, making it riskier than Ireland but safer than Portugal.

CMA turns all of its CDS data into a 5-year cumulative probability of default: both Venezuela and Greece are more likely than not to default at some point in the next five years, according to these numbers, while Argentina comes very close. Portugal, Latvia, Iceland, Ireland, Spain and Croatia are all in the 20%-25% range, along with Lebanon, which has long had a debt which would be unsustainable were it not for a rich, generous and patriotic diaspora.

One interesting thing is that CDS spreads don’t correspond directly with default probability: Croatia, for instance, has a 20.5% chance of default with CDS spreads at 322bp, while Spain has a higher default probability — 20.7% — while having significantly tighter CDS spread of 265bp.

The reason is that CMA is using at least two different recovery ratios. If Greece or Spain or Italy defaults (or if the US does, for that matter), then CMA assumes that creditors will get back 40% of their money. But the recovery value in countries like Croatia and El Salvador is assumed to be significantly lower, at 25%.

Both figures are low, but I don’t think that this discrepancy makes sense. It seems to me that recovery values should be inversely correlated with debt-to-GDP ratios: if you have relatively little debt, then a modest restructuring can get you back on a sustainable footing, while if you have a lot of debt — like Greece, for instance — then the haircut you’ll need to impose on your creditors is much greater. I see no good reason at all for assuming that recovery values in Croatia will be higher lower than in Greece, especially after accounting for the fact that Greece is likely to have a large number of preferred creditors in Europe who will insist on being paid back in full before private-sector creditors get anything.

I do think that in future reports CMA should make its recovery-value assumptions explicit. The CDS data is important and interesting, but the default probabilities are less so, based as they are on recovery rates which seem dubious indeed.

Adam Ozimek has a bright idea:

Allow any company to become a bank with a very minimal regulatory barrier if they do 100% reserves on deposits. This means people will know their money is there whenever they want it, which will make the “bank” safe from runs, which will eliminate the need for FDIC insurance or heavy handed regulation.

In principle, I like the idea of non-banks like Google and Wal-Mart being able to conduct banking-style services. It works fine in other countries, like the UK and Mexico, and in principle anything which increases competition in the banking sector should be good for consumers.

On the other hand, I don’t see any reason why these new banks should be less regulated than existing institutions. Indeed, in the first instance — the first two or three years of operation, say, while the model is still untested — I’d want much more regulation, even unto a complete ban on lending. And I’d certainly want these institutions to voluntarily submit to the oversight of the Consumer Financial Protection Bureau, even while they were under $10 billion in size.

It’s also worth asking exactly what Ozimek means by “100% reserves on deposits.” What counts as “reserves”? Would uninsured deposits at banks count? How about money market funds? Would they all need to be available on demand at par? And would the reserves count as the cash reserves of the parent company for accounting purposes? If so, could these banks ramp up against their current cash reserves without having to make any extra reserving actions at all, happily lending out everything they take in as deposits?

Even with strict controls, there are some interesting things which Google might be able to do in the payments space, including building payments functionality right into its Chrome browser and Android OS. Indeed, Google’s already moving in that direction without a banking license.

The reason that Ozimek wants loose regulation when it comes to these new entrants, of course, is that it’s much harder to do any kind of payments innovation when you’re part of a regulated institution. Which is why companies like Visa and Mastercard, even when they were owned by banks, weren’t actually banks themselves. But the problem is that if you want to encourage innovation, you also have to encourage failure. So the question I have for Ozimek is this: do you want to see a payments system fail? And how do you build a system which can cope with such a failure, if you’re not keeping a close regulatory eye on it?

Kimberly Kindy has an excellent and very sobering report on the monstrous discrepancies between the various numbers being bandied around when it comes to the amount of oil that BP is able to skim off the Gulf of Mexico every day.

As commenter hsvkitty points out, BP only got the permits to start drilling at the Deepwater Horizon site in the first place because the Minerals Management Service believed their statement that they “could recover 197 percent of the daily discharge from an uncontrolled blowout of 250,000 barrels per day”: a March report from BP said that it had the capacity to skim and remove 491,721 barrels of oil per day.

Even after the explosion, BP was still insisting that it had “skimming capacity of more than 171,000 barrels per day, with more available if needed.”

So far, it has managed to skim less than 900 barrels per day. Add burn-offs, and you get to just over 300,000 barrels in total, over 77 days — that’s less than 4,000 barrels per day.

BP’s reaction to being massively wrong, by a factor of over 100, is to grab onto the biggest numbers it can find — to try, in other words, to deal with the optics, rather than the reality. Take the much-vaunted super-skimmer, for instance. Some reports say that it “can collect up to half a million barrels of oil a day”, but it’s much more accurate to say that it can theoretically collect that many barrels of contaminated water, which is only about 10% oil. And, as Kindy drily notes, “thus far, it has been unable to produce those results in the gulf.”

BP knew full well that a blow-out at the Deepwater Horizon site would involve oil bubbling up to the surface from miles below sea level, rather than being spilled directly onto the surface from a tanker. But it never seemed to stop to think that much of the oil would never surface. BP also said that much of its skimming capacity would come from outsourcing skimming operations to Marine Spill Response — but BP never asked MSR whether they could hit its marks, and neither did MMS, when BP submitted its insanely overoptimistic numbers.

All of which is a sign of the knee-jerk credulity that most of us exhibit when faced with a large and seemingly highly accurate number. 491,721, you say? Well, that must be true — or at least in the ballpark.

It would be interesting to compare that number with other oil companies’ projections of their oil-skimming capacity in the Gulf. Their rigs haven’t exploded, of course, but what did they say they could skim off in the event that the unthinkable were to happen? In hindsight, it seems that any number over a few thousand barrels per day would clearly have been a massive overestimate. But of course the bigger the number submitted, the easier it was to get the necessary approvals. You can see why BP exaggerated so much, and why there would have been enormous incentives for its rivals to do likewise.

Quiggin’s Zombie Economics is available for pre-order — Amazon

How Citi “structured their business so that no agency knew what they needed to effectively regulate the company” — American Banker

“As any competent designer knows, the Futura uppercase ‘N’ needs to be treated with care, as the pointy corners can be treacherous” — Brand New

Surowiecki on how the auto-dealer exemption came into the financial-reform bill — TNY

The WSJ covers Middle School 223, the Lab School of Finance and Technology — WSJ

You’re more than twice as likely to be Lady Gaga’s fan on Facebook as you are to follow her on Twitter — Reuters

Kinsley with a sensible deficit column — Atlantic Wire

WizardRSS will convert any partial rss feed into a full feed — Wizard RSS

Patrick LaForge was underwhelmed by his visit to McNulty’s Tea & Coffee:

I inquired about the roaster and was told with a shrug that the shop used an unnamed roaster in Long Island City, Queens. Presumably the beans had been roasted recently.

Many coffee sellers now offer tasting notes as florid and adjective-rich as wine descriptions, but there was none of that at McNulty’s. The country of origin was listed and in some cases beans were described as organic or free trade. No details were offered about the specific growers. I didn’t realize how hooked I have become on knowing this information, even though I am not an expert who can make useful judgments based on it.

This is in some respects just a difference in marketing. A place like McNulty’s relies on the mystery and mystique of foreign lands. A roaster like Intelligentsia and shops Stumptown and Cafe Grumpy appeal to a different type of consumer.

This type of customer is obsessed — perhaps too much so — with authenticity. For these consumers, coffee is no longer an exotic product arriving by ship from third-world places with unusual names. Knowing the details of origin improves the taste.

LaForge has hit on something important here, which is clearly making its way into the world of coffee from the world of wine, where it has been going strong for decades. The more you know about your beverage, the better it tastes. That’s why so many wineries put so much effort into wine tours and that’s why you’re much more likely to enjoy your bottle of pinot noir if it has been preceded by a short explanation from the sommelier of who the winemaker is, where they’re from and what exactly they’re doing. There’s really no way of telling how or whether any particular part of the story affects the taste, but the simple telling of the story makes an enormous difference.

And so when you go to the Intelligentsia website, you’ll find them featuring specific coffees like the one from Edelweiss Finagro Estate, in Tanzania. You’ll learn what to look for when you taste it: “Toasted marshmallow and sandalwood greet you in the nose while saturated notes of pomelo and red wine appear immediately on the palate.” You don’t have to have any clue what a “saturated note of pomelo” is in order to get the message.

But that’s just the beginning of what Intelligentsia serves you with your pound of joe. They’ll also tell you who’s growing the coffee (Neel and Kavita Vohora), exactly where the farm is, what varietals are grown (Bourbon, Kent, SL-28, Tacri), what altitude they’re grown at (1700 – 1800 m) and what months they’re harvested (July – November). They’ll then add some color:

These are the only farms I’ve been to in the world where the biggest source of worry is not fungus or insect damage but invasion during the night by marauding herds of elephants, buffalo and even lions! They pass through from time to time looking for water and elephants will actually locate underground pipes and dig them up with their tusks. When they walk through the farm they trample everything in their path, leaving a big swath of razed land.

People like LaForge don’t want altitude information on their coffee because they prefer 1700m coffee to 1400m coffee. Instead, Intelligentsia is supplying something much more important and valuable: a unique narrative. It’s the same thing that’s going on in the wine world:

Unlike Bordeaux, where many of the best-known chateaus are run by corporations or wealthy absentee owners, Burgundy is full of estates, including many of the leading ones, that are essentially small businesses. Dealing with Bordeaux often requires working with middle management and marketing specialists. It’s much easier to visit a Burgundian estate and find the one person who has dirt on the boots, wine on the hands and a name on the bottle.

“For people of my generation, 30 to 50, I don’t think we’ve had the same magical Bordeaux moments, not in the same way we’ve connected to Burgundy or even the Rhone,” said Laura Maniec, who runs the wine programs for more than 15 restaurants in the B. R. Guest group.

She still buys a lot of Bordeaux for restaurants like Primehouse, a Manhattan steakhouse, and Blue Water Grill, a Manhattan seafood restaurant that hosts plenty of corporate parties where Bordeaux is nearly obligatory. “But there’s a passion and a spark and a personal connection that are missing,” she said.

What you get in Burgundy is a story and that personal connection, which is impossible to find in Bordeaux. And you’re increasingly finding the same thing in the new school of coffee roasters and importers. It’ll be interesting to see where it turns up next: tea? Truffles? Tofu?

Value investor Whitney Tilson is long BP, and answered my ethics question in a Q&A sent to his investors:

Q: Regardless of how cheap BP’s stock is, is it immoral to try to profit from owning it, in light of the company’s bad behavior?

A: As noted earlier, BP appears to have an atrocious safety record. In owning the stock, we are not endorsing its behavior, either before or after the Deepwater Horizon accident. But as value investors, we sometimes have to hold our noses when we invest because the cheapest stocks are often the ones of companies that have behaved badly or are otherwise tainted. Example include McDonald’s, which many believe bears responsibility for the obesity epidemic in this country (see Fast Food Nation and Super Size Me), and Goldman Sachs, which many blame for the global financial crisis (see The Great American Bubble Machine).

That said, we would have a problem owning stock in a company if we believed that its core business harmed people – most subprime lenders at the peak of the housing bubble, certain multi-level marketing firms and tobacco companies come to mind. BP certainly doesn’t fall into this category.

As for BP’s safety record, we don’t defend it, but we don’t think BP is deliberately blowing up its own rigs and refineries and killing its employees. If an email emerged that the CEO or board of BP were warned that the Deepwater Horizon rig was likely to explode and failed to act, we would certainly rethink the morality of holding the stock.

I don’t find this answer compelling at all. First is the language in which Tilson talks about his comparables, McDonald’s and Goldman Sachs. He writes about what “many believe” and what “many blame”, and cites the most shrill and stringent critics in both cases. Being a contrarian value investor is all about making your own mind up, and what’s germane here is what (and whether) the investor thinks about the ethics of the investment, rather than what someone like Morgan Spurlock or Matt Taibbi thinks.

Tilson then says there are companies he’d have a problem investing in, if they make harmful products. That seems to imply that it’s worth taking a serious look at the ethics of owning stock in BP. But his conclusion is trite, setting up a straw man of BP deliberately killing its employees, and saying that he’d only have a serious ethical problem with BP if it knew the explosion was likely.

Note the definite article here: Tilson is saying that he’d only have qualms if BP knew this particular explosion was likely. But the ethical case against BP is that it acted with reckless indifference towards safety standards in general, that it cut corners knowing that doing so increased the likelihood of disaster, and that it should have known that an explosion was likely, at some point, and that the chances of this explosion happening at a BP rig were significantly higher than the equivalent probability at other big oil companies.

This has important implications for the stock, of course. BP has thousands of oil rigs; the chances of one of them exploding are not much smaller today than they were a few months ago. The clean-up and other costs associated with the Deepwater Horizon are one thing, but how much will BP be forced to spend on upgrading the safety systems at all of its other rigs, now? We’ve learned our lesson, and surely all want to ensure that this kind of thing doesn’t happen again. But we’ve barely started to think about what that kind of root-and-branch revamp of BP’s physical and managerial safety systems might cost, both in terms of cash and in terms of opportunity cost. I’d be interested in what Paul O’Neill thinks — before he was Treasury secretary, he did amazing things for Alcoa’s safety record. If Tony Hayward’s successor wants to do something similar, it won’t be easy, and it won’t be cheap.