The Worrisome Analogy at the Heart of the Theory on Information-Insensitive Debt

Now for Part 4 on Gary Gorton's theory about information-insensitive debt, in which we begin by dusting off our SAT analogy skills.

Retail banking : deposit insurance :: Shadow banking : x

"X" is, of course, the kind of insurance that will save the day when there's another run on the shadow banks, as we saw during the financial crisis. Deposit insurance is a neat innovation that traces back to 1934; it eliminated runs on commercial banks in times of panic. It also made deposits at a bank "information-insensitive" debt -- the value of your $1,000 at Fidelity and Security Trust is secure, even if the CEO absconds to Tahiti with $10 million in a duffel bag.

Before solving for "x" -- or, better, asking whether we should even try to solve for "x" -- let's look at how shadow banking works.

SOME BANKING TAKES PLACE “IN THE SHADOWS”

Retail banking is for you, me, Aunt Edna. Shadow banking is for the giants in the financial system, who have large amounts of cash to park -- typically money market mutual funds, insurers, pension funds. They make “deposits” and “earn” interest through a process that involves something called a repurchase (repo) agreement.

Here's an example of how that works.

A pension fund spends $100 million to "purchase" AAA asset-backed securities from JPMorgan. As part of the deal, JPMorgan agrees to buy back these securities, after a short period of time -- overnight, or maybe a week or two. The pension fund will receive a small amount of interest (a fraction of 1%, as the lending is so short term). If JPMorgan goes insolvent, the pension fund holds those securities as collateral. They can be sold and (theoretically) the pension fund recovers all its money.

Now consider what happens with retail banking with a $100 deposit if the bank becomes insolvent. The FDIC makes the investor whole, paying the $100. Similarly, the pension fund in our example really wants its $100 million returned and doesn't want to deal with those collateral securities, which may not really fetch $100 million on the open market if they happen to be complex products, especially in times of stress.

So what happens in the repo market during a "bank run"? Nervous depositors -- like this pension fund -- demand greater and greater haircuts on securities they “purchase.” In other words, instead of “depositing” $100 million and accepting say $102 million of securities, they may demand much more collateral: $110 million, $120 million. Haircuts on asset-backed securities may go from zero to 40 percent (as they did in the crisis). This has the effect of sucking 40 percent of that $100 million out of the shadow banking market.

Spread this effect around, and the impact is similar to that of a bank run.

SO WHAT WOULD ‘DEPOSIT INSURANCE’ IN THE SHADOW BANKING SYSTEM GUARANTEE?

Here’s a big problem, for those who see insuring shadow banking "deposits" as the obvious solution to bank runs: This kind of banking has a wrinkle that's not found with its retail counterpart. In the shadow system, to guarantee a depositor’s $100 million, you essentially would have to say, “Whatever the actual value of that security you bought in a repo agreement, we’ll buy it back for $100 million.”

Think about this. If you deposit $100 in a commercial bank, the FDIC says you’ll get that $100 back -- which seems fair; you have deposited a fiat currency, and you receive the same amount of that fungible currency in return. But this differs hugely from what the shadow banking system would be guaranteeing: that you would be made whole no matter what the true value of the security that you hold as collateral.

Why is this problematic (other than for the obvious reason that the security, especially if thinly traded and "marked to model," could be mispriced -- and that this tendency to mispricing will be exacerbated because of the very existence of the insurance)?

WHY RETAIL AND SHADOW BANKING ARE WORLDS APART

Well, significant differences exist between retail and shadow banking systems.

"Deposit insurance" for commercial banking means: You're insuring that a depositor of money (common currency) will receive that money back. The bank involved is usually not too risk-loving, not too large, not too interconnected, and not too complex -- plus its commercial banking activities are regulated.

"Deposit insurance" for shadow banking means: You're insuring that a depositor of money (common currency) will receive that money back. The bank involved is usually risk-loving (often an investment bank), large, highly interconnected and complex -- plus its shadow banking activities are unregulated.

Being large and highly interconnected implies that when a bank in the shadow system gets in trouble, others will soon be at risk and the amount of "deposit insurance" ultimately needed may be very high (and the FDIC model won't work, where a team of examiners takes over the bank on Friday and sorts out things so the institution can re-open on Monday -- Lehman, which was enmeshed in the shadow banking system, is still painfully crawling through bankruptcy, almost three years later, even spawning its own periodical: Lehman Brothers Bankruptcy News).

THE CHALLENGE OF PICKING “INFORMATION-INSENSITIVE” DEBT WORTHY OF BEING INSURED

Enormous problems arise when it comes to how securities will be chosen to be insured in the shadow banking system. It's comparatively easy in retail banking. The FDIC insures dollar claims. Dollar claims are in money, or fungible currency.

But in shadow banking, how will securities be selected that will qualify as "information-insensitive" collateral worthy of insuring? Will government regulators be involved in picking and/or rating them? If so, why does anyone think our regulators have the expertise to assess asset-backed securities (one form of information-insensitive debt prevalent in shadow banking) that S&P and Moody's failed miserably to understand properly during the financial crisis?

Who determines how much of this insured information-insensitive debt is appropriate? Who pays for this "insurance" and how? And, if the debt is insured to market value, that will pervert the price at which it trades (Q: What would you pay for a security that is insured for however much you pay for it? A: Potentially, the sky’s the limit.).

And if the debt is insured to market value minus a haircut, who sets the haircut? How is the haircut adjusted if that debt class grows riskier? And, even with a haircut, insurance will tend to push the price higher as traders discover ways to game the system (Here's a scenario: X buys Security C for $100, its true price. Security C, which is classified as "information-insensitive" debt, is insured up to 90 percent of its market value. X sells Security A to Y for $200, who later sells it back to X for $210. Y makes $10 and X now possesses a security that's insured to $189 -- a great game for everyone but the insurer of the debt.)

Also what precautions will be taken to ensure that financial institutions don't start smuggling in junk disguised as quality securities, trying to get them classified as "information-insensitive debt" -- the designation of which will immediately boost the value of the assets?

Next: Is “sensitivity to information” really the way investors analyze debt?

The Theory of Information-Insensitive Debt Prompts Some Head-Scratching Questions

Here's Part 3 on the magic pigs of high finance, information-insensitive debt (everyone still awake?). Last time we looked at the concept using a common-sense definition of the term. Now let's try to figure out where the theory is unsatisfying on a more granular level, by using Gorton’s own words.

First, to show us what qualifies as information-insensitive debt, he offers examples: high-grade corporate debt, government bonds (presumably U.S. Treasuries, and not Greek 10-year bonds), and AAA rated asset-backed securities.

In different places, he characterizes such debt as (the bold is mine):

[Debt that] "is very liquid because its value rarely changes and so it can be traded without fear that some people have secret information about the value of the debt. If speculators can learn information that is private (only they know it), then they can take advantage of the less informed in trade. This is not a problem if the value of the security is not sensitive to such information."

Also (page 7 of the same document, Slapped In the Face by the Invisible Hand):

"Bank debt is designed to be informationally-insensitive, that is, these bonds are not subject to adverse selection when traded because it is not profitable to produce private information to speculate in these bonds."

These definitions sound impressive, in that arid academic way, but what do they mean when applied to real debt in the wild? For example, the first one doesn't really make sense. He's saying that speculators can't take advantage of the less informed while trading information-insensitive debt, even after learning private information, because "the value of the security is not sensitive to such information."

AND NOW FOR SOME HEAD-SCRATCHING QUESTIONS

What debt could Gorton possibly be thinking of here? Does he really believe that even Treasuries are immune to being profited upon, by someone who possesses private information? If I wiretap the Federal Reserve Board meeting, and learn the Fed is about to announce an operation to purchase $800 billion of Treasuries, he doesn't think that gives me an advantage trading in this market? (Note: the second definition does add "it is not profitable" to produce private information, but this opens a new can of worms, which we’ll soon see.)

Also, U.S. Treasuries are very liquid, but it’s not true that their value rarely changes -- their value changes constantly. Now, do the bonds trade without fear that some people possess secret information about that value? For the most part, yes -- but there may also be certain junk bonds that trade without fear that some people have secret information about them. If so, could these junk bonds qualify?

And what if the junk bonds are "information insensitive" for six years, then the company reveals itself to be tottering near bankruptcy and their price becomes volatile? Do they suddenly become information sensitive, or were they always information sensitive but only seemed information insensitive?

In other words, is this quality of being “information insensitive” only ascertained after empirical evidence of how the security actually behaves? Or is a security considered information insensitive only if we can’t imagine a situation in which someone could profitably produce private information to speculate in the debt? But, honestly, no security exists for which that’s an inherent quality, as the thought experiment for Treasuries shows.

PROFITABLE FOR WHO? YOU, ME? SOME GUY IN BANGLADESH?

And "it is not profitable to produce private information" raises many fresh questions. What's profitable for a trader to do at any given moment depends on many variables that seem as though they should have little to do with information sensitivity.

A trade may be profitable simply because I can lay my hands on large-enough blocks of securities to make chasing a minuscule gain on each one worthwhile. Or, a narrower trading spread may allow me to turn a profit more easily. (Of course more-liquid securities tend to trade with narrower spreads, which leads to a Gortonian paradox, as being liquid is supposed to be a sign of information-insensitivity.) Profitability also hinges on what I pay my workforce -- so does a security become information sensitive simply because I’ve got six traders in Bangladesh who work at one-sixth the salary of their U.S. counterparts? Also how does "private information" factor in? What if it's profitable to ransack Company X's Dumpster for trading information. Does that make its debt information sensitive until the Dumpster is relocated to a more secure place?

Some of the above may sound a bit picayune. But here are the takeaway points: (1) If so many questions can be posed, doesn’t information-insensitive debt sound like a theory that presents a false dichotomy at best? (2) There are so many trades, and so many price movements on securities (especially liquid ones), how do you sort out evidence that proves a security is information insensitive, instead of the opposite?

Next: The troublesome analogy that Gary Gorton’s theory leads to.

Information-Insensitive Debt: An Unnatural Concept, For Starters

Now for Part 2 on Gary Gorton’s theory of "information-insensitive debt" in which we continue to study the question, "Is it a bad thing or is it a really bad thing?" :)

One big problem: the concept happens to be quite unnatural.

Fiat currency is probably the best example of information-insensitive debt, but it's essentially a trivial, artificial case. Retail banking deposits also qualify as a good example, but they're something different: a special case. Exactly how they're special is important to understand.

WHAT MAKES BANKING DEPOSITS REALLY, REALLY SAFE?

Gorton likes to illustrate the information insensitivity of retail banking deposits by using an example involving a check. Let's say I write a check for a $14 haircut. That piece of paper isn't worth $13.89 or $14.05 to my barber. It's worth exactly $14.

Likewise, if I go directly to my bank instead to withdraw that $14, I can be sure of getting the full amount, even if my bank is Lehman Brothers Savings Inc. and everyone's glumly packing their desk contents into boxes when I arrive. The FDIC insures my deposits up to $250,000. I can breathe easily.

So it doesn't behoove me, or anyone I trade with, to spend time investigating the financial soundness of my bank. No matter what terrible information surfaces about that bank, my deposits are covered.

See a problem already? The debt isn't naturally insensitive to information. It achieves this property by being insured. But the value of anything -- your collection of Pokemon cards or seashells -- can become information insensitive if insured. So, becoming information insensitive this way feels like cheating a bit.

That leaves the tantalizing question: which debt is naturally information insensitive?

None of it, really. On its face, the phrase is oxymoronic, like “jumbo shrimp.” (Note: Gorton parses the term in a special way, which we'll look at later.)

DEBT IN THE WILD IS NATURALLY SENSITIVE TO INFORMATION

Pretty much all debt in its natural state is information sensitive. Markets trade on this information. Some is public. Some is private (e.g., a stock price spikes right before a merger announcement, as the news leaks out). Much information arguably occupies a gray area between public and private. Is private analysis of public data showing that a bond is undervalued private or public information?

Even fear and wild speculation is information of a sort. Say there's a rumor that a neutron bomb will be detonated in Microsoft's main cafeteria tomorrow, based on absolutely nothing. If enough stupid investors believe it (ever hear the phrase "dumb money"?), they may sell their bond holdings in the software giant. Information about this crazy rumor will prompt a smart trader to jump in, scoop up Microsoft debt, and score a neat profit when the price rebounds.

A smart theory would posit that just about all debt is information sensitive. The theory might make an argument that there are varying degrees of sensitivity, and that a particular instance of debt lies on a continuum between very information sensitive and not-that-information sensitive. Okay, fine -- that would at least be nuanced and cautious. But instead, in Gorton's world, we get debt that is either "information insensitive" or "information sensitive" -- and of course debt that lurches from the former to the latter during a financial panic, as if undergoing a change of phase, like ice to water.

SO WHY SHOULD ANYONE CARE ABOUT ANY OF THIS IN THE FIRST PLACE?

Because there’s a shadow banking system in the U.S. that’s larger than the retail banking system. It’s where the financial crisis began in 2008.

Information-insensitive debt plays a key role in shadow banking’s repo market, according to Gorton (later, we’ll look at how repo works). Asset-backed securities, for example, are posted as collateral against repo borrowings. During the financial crisis, the securities suffered huge haircuts once they became "information sensitive" (or once investors discovered they more closely resembled magic pigs than Treasuries). At the same time, Gorton notes, other kinds of debt suffered very minor haircuts.

So here’s something to ponder: If we must have "information-insensitive" debt in our financial system, shouldn't we look to these other types for what it should look like, and not to securitizations that are opaque and become thinly traded with alarming suddenness?

Next: Gorton’s own definition of information-insensitive debt comes up short.

Everything You Always Wanted to Know About Information-Insensitive Debt But Were Afraid to Ask

Over the last few months, I’ve spent a lot of time studying the idea of "information-insensitive debt" (also known less gracefully as "informationally insensitive debt"). Gary Gorton, a professor at Yale’s graduate business school, appears to be the intellectual progenitor (or one of them) of this concept. In 1990, he wrote an academic paper with George Pennacchi titled "Financial Intermediaries and Liquidity Creation."

My fascination with information-insensitive debt arose from a sneaking suspicion that it was a bad thing (except for a couple of notable exceptions). My keen interest in writing about it, after all this research, arises from a conviction that it is a bad thing, and that the theory itself isn't much good either.

A rough-and-ready definition of information-insensitive debt -- we'll return later to Gorton's own more nuanced and precise definition -- is this, by way of Felix Salmon:

Financial assets which (normally) don’t change in price when new information about them emerges.

Now if you're a markets-oriented person, this very idea should make your skin crawl from the get go. What kind of zombie asset doesn't change in price when new information about it emerges? How weird is that?

MAKING BACON OFF MAGIC PIGS

To begin this series of posts about information-insensitive debt (there’s waaay too much to fit into a single piece, unfortunately), let me introduce you to my magic pigs.

Each magic pig is worth exactly $1 million. Its astonishing value resides in something I call a noumenon. When asked what this noumenon is, which is a thing unseen, I gladly provide a 9,000-word document, with much high-level math and abstruse concepts and economic formulas, to justify its value. I trade a lot in financial markets, and whenever my counterparty demands collateral, I offer bonds entitling him to a number of my magic pigs, should I fail to deliver on whatever I have promised.

So when $10 million of collateral is requested, I hand over certificates for 10 magic pigs. My counterparty doesn't object: the whole market has accepted that these pigs are magic and worth $1 million apiece (after all, I do have documentation and the pigs have been rated top grade by Standard & Poor’s -- ignore for the moment their pro-animal bias, as one of their officials once famously observed, “It could be structured by cows and we would rate it”).

Sometimes I sell a magic pig for $1 million, and the holder of that pig then uses it for collateral, or sells it. Or whatever. Because the value of the pig lies in this complex noumenon, no market participant has any advantage in trying to profit on my pigs (through trading), by first gaining private information. And if a leg falls off a pig, that doesn't matter because its noumenon isn't affected. Even if the pig dies, its noumenon stays intact. So it's still worth $1 million.

The certificates for my magic pigs are truly information insensitive debt -- at least, until I am revealed as a fraud, at which point they very rapidly become information sensitive and start rising and falling in accordance with the market on hog futures.

Next: What do “jumbo shrimp” and “information-insensitive debt” have in common?

S&P Demonstrates Its Utter Hypocrisy

This morning, I saw S&P had "bravely" downgraded the U.S. to AA+.

What horsecrap.

If nothing else, this shows how irrelevant ratings have become.

Yields on 10-year Treasuries are about 2.5 percent, nearly at historical lows. Every time the market catches a whiff of fear, investors pile into Treasuries. U.S. debt is considered about the safest stuff out there, bar none, as indicated by the yield.

The yield is truth, the market itself speaking.

Meanwhile, S&P is still willing to rate lots of securities AAA -- if you call them CDOs and pay S&P a handsome fee for the rating, even when these securitizations are paying a yield that far exceeds that on U.S. government debt.

What we're seeing today is just rank hypocrisy from a ratings service that has the gall to claim that AAA is AAA, across asset classes.

Tip to Washington: just figure out a way to combine and tranche your Treasuries in some kind of god-awful complex structure -- make it really, really complicated -- then go back and pay S&P a fat fee to rate the mess. You'll get your AAA back. I guarantee it.

Felix Salmon: Quote of the Week

I was beginning to despair this week. It seemed like the theme was Free Market Idiots Run Amok and someone had forgotten to tell me to put on my Atlas Shrugged underwear.

First, there was Larry Summers saying don't blame financial innovation, but rather, the housing bubble, for the Financial Crisis on Steroids that we went through. Which leaves me wondering: This guy was the president of What-vard? You gotta be kidding.

Larry has nary a disparaging word to utter about any CDO, CLO, CDS, SIV, RMBS, CMBS, REMIC, re-REMIC, structured note. None of those products are bad, fee-gouging, or in any way contributed the teensiest to systemic instability?

Thus, I sentence the estimable Mr. Summers to hereafter receive all his compensation as an income stream from the bottom tier of a synthetic CDO squared.

The second forehead-slapper was Peter Wallison, who at this point isn't even interesting any longer. He's too busy riding his ideological hobbyhorse: Fannie and Freddie are the chief perps in the financial crisis, get the government out of the housing market, private enterprise can do no wrong, blah blah blah. His latest wildly inaccurate piece, on the Bloomberg Web site, was easily shredded by Daily Kos.

Which brings me to Salmon's quote of the week ... the antidote to all the uninformed opinion.

Responding to Gary Gorton's assertion that we need "informationally insensitive financial assets" (debt whose price doesn't change when new information about it emerges), Salmon is quick to respond, "No, we don't."

Informationally-insensitive debt is the best repository the world has ever constructed for housing tail risk in an invisible and impossible-to-measure manner.

That's a great line. Because "informationally insensitive debt" (except when applied to common currency, which is arguably a trivial usage) is a dangerous, oxymoronic twist of phrase. The varieties of "informationally insensitive debt" spawned in the shadow banking system, and used in the repo market before the financial crisis, didn't collapse because of a panic. Rather, the values plunged because investors realized this debt was complex, misrated and overvalued.

Gorton likes to think there was an irrational panic in the shadow banking system. He's right there was a panic, but it was more the sort of panic you experience when you realize the gold bars you're holding are bricks of horse manure spray-painted gold. It wasn't all that irrational. And to prevent that kind of panic, you need assets to be more informationally sensitive, and constantly adjusting for risk -- not less sensitive.

In Case You Were Thinking That Lawsuits Might Reform the Credit Raters ...

Well, think again.

Time to play connect the dots, with the New York Times getting us started in fine fashion. First, the article linked above notes a disturbing trend taking shape:

... since Dodd-Frank passed, Congress’s noble attempt to protect investors from misconduct by ratings agencies has been thwarted by, of all things, the Securities & Exchange Commission. The S.E.C., which calls itself “the investor’s advocate,” is quietly allowing the raters to escape this accountability.

What accountability? As Gretchen Morgenson tells us:

The Dodd-Frank financial reform law, enacted last year, imposed the same legal liabilities on Moody’s, Standard & Poor’s and other credit raters that have long applied to legal and accounting firms that attest to statements made in securities prospectuses. Investors cheered the legislation, which subjected the ratings agencies to what is known as expert liability under the securities laws.

Why would the SEC do anything that subverts Dodd-Frank so openly? Aren't these our brave regulators, our white knights (yes, with a fondness for porno, but what do you expect in the postmodern age)? Ah, but see, there was a problem when push came to shove on this "expert liability" point.

When Dodd-Frank became law last July, it required that ratings agencies assigning grades to asset-backed securities be subject to expert liability from that moment on. This opened the agencies to lawsuits from investors, a policing mechanism that law firms and accountants have contended with for years. The agencies responded by refusing to allow their ratings to be disclosed in asset-backed securities deals.

So basically, the credit rating services said, "We'll hold our breath until we turn blue."

And the SEC blinked. Actually, double-blinked. Check out this whopping beaut of negligence:

At the time, the S.E.C. said its action (i.e., the agency temporarily removed the "expert liability" threat and said it wouldn't bring enforcement actions against issuers that did not disclose ratings in prospectuses) was intended to give issuers time to adapt to the Dodd-Frank rules and would stay in place for only six months. But on Jan. 24, the S.E.C. extended its nonenforcement stance indefinitely.

Indefinitely. Golly, that has a sort of open-ended ring to it. Hey, let's face it: Like diamonds, indefinitely may be forever.

Okay, now let's do the New York Times one better and finish connecting the dots for them. Because left unanswered is a big question: why the hell are credit rating services so scared of being held responsible for how they grade asset-backed products?

Jeez, I think I know this one -- in fact I think I blogged this one -- twice over! Just go here for a full explanation:

The Ratings Charade Continues: a CLO Investigation, Part I
The Ratings Charade Continues: a CLO Investigation, Part II

You see, the credit rating companies can't afford to be held legally responsible for the grades they assign to asset-backed securities, such as collateralized loan obligations, because they know they would lose in court if these ratings were challenged! The ratings are clearly bogus. I show that above, using nothing more sophisticated than sixth-grade math.

Further, these companies must know their ratings on asset-backed securities are wrong, unless they're incompetent to a mind-blowing degree.

And now the SEC is giving Standard & Poor's and Moody's a free ride on their bogus ratings, aiding and abetting the crime ...

What a great country we live in, eh? Where does the U.S. rank on that global corruption scale again? ;)