Whenever there’s a surprise takeover bid at a significant premium to the (formerly) prevailing stock-market price, dozens of journalists and bloggers immediately pull up options-volume data. Much of the time, they discover a suspicious spike in options volume just before the deal was announced. The conclusion is obvious: insider trading!

So many thanks to Baruch for bringing a bit of context to bear on such exercises. His main points: is that options volumes are by their nature extremely lumpy. Just about any options contract, if you look at it, will have occasional extremely large spikes. As Baruch puts it, “the volume of a particular option resides in Extremistan”.

What’s more, the options markets are constantly awash in rumors, any one of which can easily cause on of these spikes. Baruch provides one example: last Friday, a rumor hit the market that Palm would be bought by Nokia. It wasn’t, but that didn’t stop 21,000 options being traded in one day on the $12.50 calls alone. That’s over five times the size of the option volume in 3Com before it was bought by HP.

And one other thing: if you did have inside information that 3Com was being bought by Nokia, you’d make more money simply buying the stock than you would buying the options. The only reason to buy the options is if you simply don’t have the cash to buy the stock.

None of which is to say that there wasn’t insider trading in 3Com, of course. It’s pretty hard to prove a negative. But statistically speaking, if you look at options volumes on every takeover bid which crosses the transom, eventually one of them is going to see this kind of spike, even if there’s no insider trading at all.

Peter Eavis notes something quite astonishing today:

The interest rate on [Goldman's] long-term borrowings was a minuscule 0.92% in the third quarter, down from 3.53% in the third quarter of 2008. This $203 billion of debt is Goldman’s largest single funding source, so as its cost plunges, its bottom line benefits…

Goldman has been helped by its use of interest-rate derivatives. When issuing long-term fixed-rate debt, Goldman has for years entered swaps that effectively convert nearly all of that debt to floating-rate. Thus, as interest rates plummeted, so did one of Goldman’s main expenses.

To put these numbers into perspective, a savings of 2.43 percentage points in one quarter amounts to $1.2 billion in saved interest costs on $203 billion. That’s over 40% of its third-quarter earnings.

Even so, Goldman’s hedging gains by converting fixed-rate into floating-rate debt pale in relation to $3.6 billion that Wells Fargo made on much the same trade, hedging its mortgage-servicing rights. Clearly much if not most of the US banking sector made enormous profits in Q3 on interest-rate swaps — profits which are the very definition of unsustainable.

And there’s another question, too: if the likes of Wells Fargo and Goldman Sachs are making billions on these swaps, who’s on the other side of the trade? Who lost billions of dollars by swapping floating into fixed? Call it the Summers trade, after Larry’s disastrous foray into the rates market when he was at Harvard. It didn’t work then, and it clearly isn’t working now, either.

Bloomberg’s Michael Moore has lots of detail today on the treatment of mortgage servicing rights in banks’ earnings reports. No, wait, it’s actually interesting! Especially when you look at the numbers involved.

Wells Fargo is the poster child here: it wrote down its mortgage servicing rights by a whopping $2.1 billion last quarter, but it actually made a profit of $1.5 billion on them, since the value of its hedges on those rights soared by $3.6 billion. At the end of the quarter, it valued its rights at $14.5 billion.

Moore tries to explain what’s going on here, but something smells fishy:

The value of the rights depends largely on the expected life of the mortgage, which ends when a borrower pays off the loan, refinances or defaults. When rates drop and more borrowers refinance, MSR values decline. Banks typically hedge the movements using interest-rate swaps and other derivatives…

Because there’s no active trading in the contracts, there are no reliable prices to gauge whether banks are valuing the rights accurately, analysts said.

Here’s the first thing which puzzles me: in the third quarter, mortgage rates fell by 0.26 percentage points. How could such a relatively modest decline in rates result in a plunge of $2.1 billion — more than 12.5% — in the value of a $16.6 billion portfolio? And what kind of hedges result in a $3.6 billion profit when rates decline by 26bp?

Over at JP Morgan, the numbers are a little more modest, although Jamie Dimon’s shop, too, contrived to make a profit on its hedges: the portfolio declined in value by $1.1 billion, or 7.5%, while the bank’s hedges went up in value by $1.5 billion.

I’m especially puzzled by the big writedowns because anecdotally there doesn’t seem to be a huge amount of mortgage refinancing going on, and I don’t think that expected default rates rose in the third quarter either. (If anything, judging by the prices of mortgage bonds, they fell.)

There also seems to be a strong correlation between the size of the writedown that a bank took, on the one hand, and the degree to which its hedges made money, on the other. Is there any good reason why this should be the case? Or are banks just taking writedowns as and when they manage to pay for them out of hedging profits?

Many thanks to Bill, who read my blog entry on derivatives exposure and who is much better at navigating the OCC’s quarterly reports than I am. And it turns out that they’re rather misleading.

Consider this, from the executive summary:

Five large commercial banks represent 97% of the total banking industry notional amounts and 88% of industry net current credit exposure.

What are those five large banks? According to the OCC, they’re JP Morgan Chase, Goldman Sachs, Bank of America, Citibank, and Wells Fargo. Add up their “total derivatives” numbers in Table 1 of the latest OCC report, and you get $197 trillion, which is indeed 97% of the $203 trillion in total notionals.

But never mind that, and take a look at Table 2 instead. That shows the notional amounts at bank holding companies, rather than banks themselves. Suddenly, the size of Bank of America’s derivatives holdings spikes from $39 trillion to $75 trillion, while Morgan Stanley appears from nowhere to reveal itself as holding a more-than-healthy $41 trillion in derivatives. It seems that at Merrill Lynch and Morgan Stanley, the derivatives are generally held by the holdco rather than the bank, which allows the OCC to ignore them for the purposes of its headline calculations.

Add up the derivatives books at the holdcos, and the total isn’t $203 trillion any more: it’s $291 trillion — an increase of $88 trillion which is very hard to find in the OCC report unless you’re specifically looking for it. And never mind Wells Fargo, which was also something of an also-ran in the top five banks. The top five now comprise JP Morgan, Bank of America, Goldman Sachs, Morgan Stanley, and Citigroup, with derivatives holdings between them of $278 trillion. That’s 95% of the true total, or 137% of what the OCC would have you believe was the total.

Why does the OCC make the rather legalistic distinction between commercial banks and holding companies, and then concentrate on the former rather than the latter? If the object of the exercise is to add up total derivatives exposure, that doesn’t make any sense. Someone there should really start asking themselves why they’re publishing this report, and how they could make it most helpful to the people reading it.

The WSJ’s Eleanor Laise finds that the market in bond ETFs is rather messy, to say the least:

State Street Corp.’s SPDR Barclays Capital High Yield Bond ETF fell nearly 1% in the 12 months ending Aug. 31, even while its benchmark gained 6%. Part of the problem: The index contained some lower-quality bonds that the ETF couldn’t get, and when those bonds rallied, the fund got left behind, says Jim Ross, senior managing director at State Street.

The benchmark in question, it’s worth noting, is the Barclays Capital High Yield Very Liquid Index. That’s evidently “Very Liquid” as in “can’t find these bonds to buy no matter how hard we try”.

The lesson of this story is that ETFs don’t work when they’re investing in instruments which aren’t exchange-traded:

Keeping ETF returns in line with the indexes has proven to be tough in the murky bond market. For most bonds, there is no centralized exchange matching bond buyers and sellers, and different market players can assign very different prices to the same bonds. Many bonds don’t trade for days at a time, and when they do, they can be costly to buy and sell.

It’ll be interesting to see whether anybody tries to launch an ETF based on a liquid, exchange-traded CDS index. That might solve most of these problems, but I know a lot of people who would be dead-set against such a product, since its very existence would imply the dreaded “naked shorting” of CDS by dastardly speculators. Is a failed investment product better than a liquid derivatives market?

David Leonhardt interviewed Alan Greenspan at the Atlantic event in Washington this morning, and it was quite shocking how little Greenspan seems to have learned from the crisis. Yes, he has decided that banks need more capital: “You cannot function with a financial system with capital as low as it was,” he said, adding that “in retrospect, Basel II was clearly suboptimal. You can’t have capital at 10% or less, because human nature changes too quickly.”

But on other matters — especially when it came to systemically-important things which didn’t fail in this particular crisis, Greenspan was alarmingly sanguine.

When Leonhardt brought up the subject of derivatives, Greenspan was at pains to differentiate interest-rate and foreign-exchange derviatives, on the one hand, from credit default swaps, on the other. He explained, quite rightly, that the percentage of the notional amount which players stand to lose is much higher in the CDS market that it is in say the rates market, but he never mentioned that notional amounts outstanding in the rates market are orders of magnitude greater than the CDS market ever was.

The degree to which Greenspan claimed not to be worried about the vast bulk of the derivatives business was highly alarming:

Nobody hears about problems in interest-rate or fx derivatives. We’ve had the most extraordinary stress test, and they came up clean. I’m saddened by the fact that the problems in CDS have been generalized to all financial derivatives.

For one thing, we do hear about problems in interest-rate derivatives: a man appearing at the conference later this afternoon, Larry Summers, contrived to lose $1 billion of Harvard’s money in just that market. Given the extreme measures which have been taken by the world’s central banks, and given the fact that global capital imbalances remain enormous, there’s clearly a lot of tail risk in the interest-rate and fx markets. If volatility there spikes in some unprecedented way — which is entirely possible — then no one knows who could end up becoming spectacularly unstuck in those markets.

Greenspan defended the CDS market, too, in familiar terms:

We have to make adjustments in the way that market is working, but fundamentally the concept of CDS is a very desirable one. You are distributing risk to those who are more interested in holding it or more capable of holding it, and that’s desirable.

Haven’t we learned that this isn’t true? Haven’t we learned that the people who end up holding the risk, once the banks have done their derivatives-based, structured-product thing, are precisely people who are not interested in holding it but who don’t realize what risk they’re holding? After all, those buyers will always be much more attractive, to risk sellers, than the kind of risk-hungry hedge funds who demand high returns for taking on that risk. The derivatives market turns out to have been one the best mechanisms ever designed for hiding risk, and I don’t see that as desirable in the least.

Greenspan, in other words, seems to have learned almost nothing from the crisis. When Northern Rock failed, he said, “the UK authorities were the best set of regulators in the world” — er no, they weren’t, as a glance at the leverage ratios at UK banks would tell anybody. So when Greenspan talks about inflation not being a problem until 2012, as he did at the end of his session, it’s not obvious why anybody should listen to him. He simply isn’t reliable or useful any more.

James Kwak has a great response to Robert Shiller’s FT op-ed about financial innovation. But his line at the end about how “for the sake of argument, I am willing to concede that these are useful innovations that would make people better off” has been misconstrued, and it’s worth pointing out that in fact they’re not useful innovations that would make people better off.

Why not? Mainly because, at heart, they’re all variations on the theme of doing-clever-things-with-as-yet-uninvented-derivatives. But that’s a theme which really shouldn’t have survived the financial crisis.

In 2003, Alan Greenspan famously said that ““what we have found over the years in the marketplace is that derivatives have been an extraordinarily useful vehicle to transfer risk from those who shouldn’t be taking it to those who are willing to and are capable of doing so.” But, as it turned out, they weren’t. Lots of people wanted to transfer risk, and precious few were genuinely willing and able to take it on: even hedge funds generally prided themselves on being lower-risk than the stock market as a whole.

The result was a system where derivatives were used to hide risks, and shunt them off, unseen, into the tails. A system where hidden risks turned out to be much more dangerous than if they’d all been out in the open all along.

In each of Shiller’s examples, you start with a risk that an individual wants to lay off: the risk that home prices will fall, or that the market will plunge just before you retire, or that inflation or healthcare costs will kick in after you retire. And in each case, the individual tries to lay off that risk elsewhere.

Shiller simply takes it on faith, however, that a nice liquid market can and should spring up to provide two-way prices in such risks, solving lots of problems at a stroke; he doesn’t stop to consider that maybe the reason such markets haven’t sprung up is precisely that there’s no real demand from anybody wanting to take on those risks. In reality, Shiller should know that better than anyone: his much-vaunted house-price derivatives have gone nowhere, partly because no one ever really had any need or desire to go long.

Shiller’s first proposed innovation is an attempt to deal with the regrettable move, in recent years, towards a “popular reliance on housing as an investment”. Only except for suggesting the obvious — moving people away from the idea of housing as an investment, and towards good old-fashioned renting — he comes up with a complex mortgage with all manner of embedded options. As with any derivative, those options will be a zero-sum game, and you can be sure the homeowner is going to end up on the wrong side of it. But Shiller still seems to think it’s a good idea for homeowners to buy them, instead of simply getting a plain-vanilla, easily-comprehensible mortgage, with payments which are set over the life of the loan and which are entirely predictable.

Shiller’s second innovation is target-date mutual funds: I’m almost surprised that he didn’t suggest some kind of principal-protection scheme, involving derivatives, instead. (Maybe because that would have sounded too much like portfolio insurance, which caused the 1987 crash.) But target-date funds are widely misunderstood, and in 2008 some 2010 funds contrived to underperform the S&P 500. They’re a mess, and they’re expensive, and it’s not obvious that they do any good. So color me unconvinced on this front.

Finally, Shiller dreams of annuities which would pay out a set amount of money per month for life, after adjusting for inflation, and which on top would include all healthcare costs. Nice dream, Bob. But there’s a good reason such things don’t exist: no insurer wants to take on a large amount of those kind of risks, and nor would any self-respecting insurance regulator allow them to do so. There isn’t a long-dated derivatives market in future health-insurance costs, but if there were, very few people would have any reason to sell protection against those costs rising, at any price. Sure, a few hedge funds might dabble in the market. But they could never satisfy demand. And in any case, as with the mortgages, the buyer of the annuity would generally lose out, just because of the zero-sum nature of derivatives. Since the cost of the embedded options would be hidden in the price of the annuity, the insurer would have a license to charge through the nose for it, and the buyer of the annuity would never know.

Simplicity and transparency are good things, and Shiller’s innovations go the other way instead. That’s nothing to get excited about.

Vincent Fernando has a blog entry today headlined “It’s Time To Stop Being Scared By Derivatives’ Trillion Dollar Notional Values”. Which is a bit like saying “It’s Time To Stop Being Scared By The $6 Billion Budget Deficit”: both of them are off by a factor of 200 or more.

A decade ago, when notional derivatives exposure started being measured in the trillions, bankers started wheeling out all of Fernando’s arguments in an attempt to reassure the public that there really wasn’t all that much risk here. And if those exposures were still only a trillion or two, I might not be all that worried either. But the likes of Fernando don’t seem to understand that when the notional exposure increases by more than two orders of magnitude, whole new systemic risks can come into play.

US banks made $15 billion trading derivatives in the first half of this year. That’s real money, by anybody’s lights. And nobody believes that you can make $15 billion in the space of six months without running any risk. Indeed, as we’ve learned the hard way, in financial markets the downside tends to dwarf the upside. If US banks can stand to make $15 billion in six months, how much can they stand to lose in the same amount of time? And who would pick up the bill if that happened?

I’m no naif when it comes to deriviatives; indeed, I’ve been on the other side of this argument, explaining how it’s possible for notional exposure to increase without net exposure increasing. But I do get the feeling that far too many people unthinkingly accept that just because such a thing is possible, it must perforce be happening. Even as the revenue figures tell a very different story. (And yes, the credit-exposure figures are falling, but that’s largely a function of the declining notional quantities in the CDS market, which makes up a very small proportion of the total derivatives market.)

Here’s a little table I put together with numbers from the OCC — page 9 of this pdf. Using second-quarter numbers for each year, I looked at the total nominal derivatives exposure of end users — the people for whom derivatives are meant to exist — and for dealers.

The results are pretty startling: while end-users have pared their derivatives exposure to a seven-year low, dealers have increased theirs to yet another all-time high. And as the OCC notes, when we say “dealers”, we really mean four banks in particular: JP Morgan Chase, Goldman Sachs, Bank of America, and Citibank.

Oh, and did I mention? The amounts here are in trillions.

What has happened in recent years that derivatives dealers now need $78 in nominal derivatives exposure for every $1 in end-user exposure? When Adair Turner talks about “profitable activities so unlikely to have a social benefit, direct or indirect, that [banks] should voluntarily walk away from them”, this is surely a prime example of what he has in mind.

When the OCC tells us that total derivative notionals are now above $200 trillion, we can’t really help but go blank: the number is so many orders of magnitude divorced from any conceivable reality that it’s almost impossible to work out what it could possibly mean. But clearly that kind of exposure wasn’t necessary a few years ago. So why is it now?

Newsweek.com is doing a big month-long series on the end of the decade, and, inevitably, it’s going to feature lots of listicles. One of them is a list of the top ten “history altering decisions” — seemingly-small moves that had massive consequences. Each one is going to be written up, and Newsweek has asked me for a very short essay (just 200 words or so) on the consequences of Bill Clinton signing the Commodity Futures Modernization Act in 2000.

Top of mind at Newsweek are the Enron collapse and the unregulated rise of the CDS market — can those fairly be ascribed to the CFMA? Are there any other key consequences of the CFMA’s passage which I should include? Newsweek has given me full freedom to crowdsource this, so if you ever thought you might have a CFMA blog entry in you somewhere, now’s the time to write the thing, or just leave a comment here. Thanks!