Harvard, Princeton Economists Say No Fire Sale Prices, Premise of Public-Private Partnership Wrong

We have been saying for some time that the policy premise of the Fed and Treasury has been that the financial crisis is that it is a liquidity crisis, not a solvency crisis. If you are of that school, the fallen prices of various assets is due to a combination of scarcity of funding plus irrational panic. Find ways to provide liquidity and give investors that magic elixir, confidence, and voila, crisis over.

Having watched the credit markets closely before the implosion, we’ll agree there was plenty of irrationality. But it was in the gross underpricing of risk. The snapback to current pricing to us thus seems a return to rationality plus new fundamentals based on borrowers who never should have been lent money in the first place defaulting on such a scale as to damage overall economic activity. And that means, as plenty of Serious Economists (Krugman, Buiter, Stiglitz, to name a few) have warned, the Geithner cash for trash program is a huge misallocation of taxpayer dollars. Even granting that something must be done about the banking system, this is a covert and wasteful way to go about it.

That thesis has been validated by Harvard’s Joshua Coval and Erik Stafford and Princeton’s Jakub Jurek in a paper “The Pricing of Investment Grade Credit Risk during the Financial Crisis” (hat tip Bill Black). It looks at the repricing of investment grade credits, which is easier to analyze than structured credits (you have other claims, namely stocks, on the same entities, which allows for a relative analysis).

The paper starts by mentioning the public private partnership and its belief that market prices are distressed:

The government’’s view is that a disappearance of liquidity has caused credit market prices to no longer refl‡ect fundamentals:2
An initial fundamental shock associated with the bursting of the housing bubble and deteriorating economic conditions generated losses for leveraged investors including banks … The resulting need to reduce risk triggered a wide-scale deleveraging in these markets and led to …fire sales … [The Public-Private Investment Program] should facilitate price discovery and should help, over time, to reduce the excessive liquidity discounts embedded in current legacy asset prices…..

Yves here. Did you catch that? The price collapse is due mainly to “excessive liquidity discounts”. Please. Some of the exotic flavors of now junk paper (as in certain CDOs) were called “trades” because they were designed NOT to be resold. So the concept of a liquidity discount applying to paper anticipated to be illiquid from the get-go is quite a stretch.

As an aside, we have to mention the intellectual inconsistency. The logic of the PPIP is that current market prices are wrong. Yet the authorities failed to question prices (or more important, overall leverage) in the frothy days. Funny how that works. Now that prices are low, that can’t be right, since it’s way too inconvenient, so we are going to create a rigged market and claim it’s necessary to produce “better” prices. Back to the article:

Our results suggest changes in fundamentals, as re‡ected in the equity market, account for a large portion of the repricing of credit that has occurred. In particular, the dramatic increase in the price of low cash‡ow states can account for most, if not all, of the rise in credit spreads for cash bonds. The spreads on credit default swaps, which currently trade at a large and negative basis relative to the underlying bonds, appear too low relative to risk-matched alternatives in the equity

We also …nd that the repricing of the investment grade structured credit securities suggests a correction of an ex ante failure of investors to appropriately charge for systematic risk. Prior to the crisis, Coval, Jurek, and Sta¤ord (2009b) argued that investors did not appreciate the systematic risk exposures of these securities and provided evidence that credit protection on the senior tranches of the investment grade CDX was underpriced (i.e. spreads were too low), while protection on the junior most tranche was overpriced.

Did you catch that little doozy, credit default swaps now look underpriced? Lordie. And CDS are still being written, some by firms that have Federal backstops.

I’d welcome the input of any hard core quants reading the paper, but I also note the authors use a modified CAPM approach. That I presume would still use Gaussian distributions, Although most quants resist using Levy type distributions (which are brutally hard to model, the Options Clearing Corp. since 1990 has used Levy distributions for setting margin requirements, which says it is not as impossible to implement as the crowd adhering to the conventional techniques asserts). Thus I wonder if even this approach understates risk, for if so, that would argue for even lower prices.

Print Friendly
Tweet about this on Twitter0Digg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone


  1. Anonymous

    Great posting Yves. Lordie, lordie. Will those folks ever realize how foolish they are looking? Still doubling down on derivatives after they have proven to be toxic….whoocooodaanode?

    Is there some way we can swap the pot smokers in jail for getting caught getting high with the greedy entitlement bastards from our financial sector who are bankrupting our country? I am sure we would be better off.


  2. Anonymous

    Yves, we should refrain from repeating the assertion that it is the government's "view" that asset prices are undervalued.

    In actuality, while it may seem to be purely semantic, it is important to recognize that it is really government "policy" that asset prices be considered undervalued.

    TaxcheatTimmy & HeliBen understand the same exact fundamentals just as well as you and any other intelligent financial blogger (perhaps more so). And that is that prices **were** irrational and they are now in the process of deflating to reach rationale levels.

    However, they do not have the luxury of speaking (or writing) the truth. There is a reason why governments lie, and that is because the general public really cannot handle the truth.

    What do imagine would happen if Barry came out and said "You know, let's dispense with this kabuki theater. Truthfully, the US is insolvent, including its major banks, your retirement savings and pension plans, etc – they are all bankrupt. Yep, worthless. We've been living on borrowed means now for 20+ years, and the jig is up."

    Chaos, that's what. Of course, it's gonna happen anyway, but at least people like Ben, et al will continue to believe in their hearts that they at least gave it the old college try.

  3. ben

    we have to mention the intellectual inconsistency. The logic of the PPIP is that current market prices are wrong. Yet the authorities failed to question prices (or more important, overall leverage) in the frothy days.

    Excellent point which is more than a matter of intellectual interest. People are entitled to change their minds, and in the light of recent events it’s reasonable to change one’s mind about the reliability of market price-setting. But if one does so, the rational action is to fire the economists who made the mistake and replace them with economists who saw the bubble. Suppose Dean Baker and Nouriel Roubini were hired at Treasury, took a hard look at bank assets, told the world they were undervalued, and recommended price discovery measures. They could not be charged with inconsistency and their policy prescriptions would deserve respectful attention.

  4. Anonymous

    I didn’t quite understand the part about CDS spreads and whether that means they are underpriced or overpriced. For example, if a bond had a spread (or difference in yield) that was “too low” compared to Treasuries given the extra credit risk, that means the bond itself is “overpriced”, right? So when the authors say that CDS spreads are too low, does that mean that CDS’s are selling for prices which are too high, rather than being underpriced? If one of the experts here can straighten me out I would appreciate it.

  5. Dave

    @Lucifer: even by my own incredibly broad standards, may I suggest that your last remark was more than a little off-colour …

    But I will happily join you in calls for “decapitation” :)

  6. Anonymous

    This is what happens when you buy a pig in a poke, only to find at home it only has 1 leg.

    Then its time to staple some thing that resembles the missing 3 legs on it and resell to the next dim whit, AKA dead parrot sketch.


  7. Anonymous

    Are CDS really underpriced? The tragedy may be that, with the government as the new backstop, their prices may have adjusted to reflect this. What’s the right price for a CDS on Citi? Considering the U.S. will let it go bankrupt this side of never, it could be pretty cheap yes? So we can thank Geithner and Obama’s crew for more market distortions … just save those banks, at whatever cost!

  8. Anonymous

    Larry Summers, Tim Geithner Owned By Wall Street!

    Financial institutions including JP Morgan Chase, Citigroup, Goldman Sachs, Lehman Brothers and Merrill Lynch paid Summers for speaking appearances in 2008. Fees ranged from $45,000 for a Nov. 12 Merrill Lynch appearance to $135,000 for an April 16 visit to Goldman Sachs, according to his disclosure form.


  9. Anonymous

    I beg of some patient soul on this forum to explain something to me:

    The argument I’ve often heard (from the banks) is that if the banks can just hold certain assets ‘to maturity’ then all will be well.

    At my amoeba-level understanding of finance/economics: The maturity date of a financial asset is the date at which an asset is converted into a specified amount of money or physical assets.

    Now, if the asset is toxic, I take this to mean it isn’t generating any cash flows at all, or very few. So, this being the case, how can it be converted to money at a particular date in the future (i.e. at maturity)? How will holding toxic assets to maturity change things? Or is my comprehension level even worse than I thought?

    — confused soul

  10. Eric L. Prentis

    Anonymous said…
    “What do imagine would happen if Barry came out and said "You know, let's dispense with this kabuki theater. Truthfully, the US is insolvent, including its major banks, your retirement savings and pension plans, etc – they are all bankrupt. Yep, worthless. We've been living on borrowed means now for 20+ years, and the jig is up."

    Chaos, that's what. Of course, it's gonna happen anyway, but at least people like Ben, et al will continue to believe in their hearts that they at least gave it the old college try.”
    Anonymous, I fear, makes the financial crooks sound too altruistic. Americans would welcome, for a change, truth from our leaders, eight years of clumsy Bush and Republicans lies are sufficient. Geithner/Summers/Bernanke, with all their specious arguments and PPIP & TARP ledger main, have only one goal, i.e., shift all the toxic assets currently on their patron’s bank books onto the taxpayers’ backs. When the denouement finally comes, regardless who holds the bank’s worthless paper, the people, if the bankers have their way, will be screwed twice. The greedy self dealing bankers will payoff their government shills and laugh all the way to the bank while the people receive even lower pension benefits than they would have without the toxic paper unloading. The bankers tragic miscalculation is: “you can never get away from the people.”

  11. artichoke

    Having just have a moment to look at it, it’s not really CAPM like because they allow a volatility smile (see the comment after equation 1.)

    Of more concern is the ability to extrapolate from local observed prices to the tails where default would occur. They justify that the functional form has good properties for extrapolation but this is a far cry from being convincing; it just means that any show stoppers are not apparent in the data observed or assumptions made.

    I haven’t read much more than that, may have more to say later if I get time.

  12. artichoke

    Also since they reference Breeden and Litzenberger it’s probably “Consumption CAPM” which is a bit different from CAPM one-period portfolio model.

  13. Independent Accountant

    Mike Rozeff at Lew Rockwell cited this paper favorably in the last few days. When I get a chance I’ll read it. That said, I’ve made these arguments since late 2007.

  14. Anonymous

    Eric L. Prentis said…
    "Geithner/Summers/Bernanke have only one goal: shift all the toxic assets currently on their patron’s bank books onto the taxpayers’ backs."

    In some regards, it's comforting to imagine a conspiracy exists because then we can conveniently pin labels on easily identifiable "bad guys". Even better, if policy decisions are based on fraud, then the proper corrective measure would be to simply reverse those decisions.

    The much more frightening prospect is that the PTB are actually working in good faith to prevent complete system failure. There is simply no way a rationale person equipped with even basic fundamental knowledge cannot come to the same conclusions as Yves, Mish, Denninger, et al. Which means Timmy & Ben are of the same mind.

    So why are they doing what they are doing? I think people looking for the "bankster" angle are missing the hidden nod & wink from these guys. If what they are doing is honest (and I believe that is the case), then they are really giving us a secret signal to get the f*ck out of Dodge.

  15. Mike

    I’m going to go through it on the train-ride home, but I’d really appreciate some other quants tackling it as well. Economics of Contempt complains it is just using the pricing models on high-grade corporate debt, vis-a-vie the CDX.NA.IG index.

  16. Some Guy

    Personally, I don’t subscribe to the view that these guys are hopelessly corrupt. It’s likely that they care more about their own skin / political future than they do scratching their friends back.

    Nor do I subscribe to the view that they’re hopelessly incompetent. This is more of a matter of faith, but I think that interconnectedness / systemic risk of the firms along with politics have created practical constraints preventing them from taking bolder steps.

    When I’m feeling really hopeful, I wonder if they view the economy as a broken car. If you have two options as to what’s wrong and the choice is between a $5 part and a $150 repair, try the $5 part. Unless delaying the $150 repair creates systemic risks for the car. I don’t think that’ll be the case with these plans, as they’ll gerryrig something–maybe involving a coat hanger and a tarp–to prevent anything catastrophic.

  17. Yves Smith


    Thanks. For my edification, is that now the rough and ready allowance for fat tails, allowing for the vol smile?

    But to your point, still does not allow for infinite variance or tell you where the authors put the end of the distribution.

  18. Mike

    Quants: I’m calling BS on it’s relationship to the distressed ‘legacy assets.’ All I’m reading is that credit spreads on Investment Grade Bonds, reflected in the CDX.NA.IG, fits into a Merton Model (of viewing equity as a call on value with debt as a strike price).

    This is very unrelated to the question at hand; all it means is that assets in the highly liquid non-legacy market reflect fundamentals. It has no eyes to the legacy asset load.

    I want to believe :) Quants drop some knowledge.

  19. Anoninca

    Mike, I think you are missing their point. They come as close as research on real world data can to demonstrating that prices on investment grade corporate CDS and CDOs failed to take systematic risk into account prior to mid-year 2007. Thus they show that these products were mispriced during the boom and only now are close to being accurately priced.

    The analogy to CDS and CDOs based on other types of loans is simply to close to be ignored. If the market could eff up its pricing this badly with something as simple as investment grade corporate debt (remember this is the stuff that’s rated reasonably accurately and that investors actually understand), then there’s every reason to believe that CDOs on more poorly understood products were even more inaccurately priced during the boom.

    No, they can not demonstrate that mortgage related CDS and CDOs are accurately priced now, but they’ve done a fine job of demonstrating that these products should exhibit dramatic price declines to reflect systematic risk that was ignored in the past.

    In other words by demonstrating the price changes for the best understood CDOs and CDS can be explained by past pricing errors, they have shown it’s stupid to attribute the price decline in any CDO or CDS to “illiquidity”.

  20. Anonymous

    Sorry, OT – but in breaking sad news, Greg Newton of nakedshorts is no more. Devastating!

  21. Anonymous

    The investment banks “needed” to underprice these securities in order to sell them and collect their commission. In fact, they assured they would be underpriced by securing fraudulent AAA ratings.

    I see the judge appointing a special investigation into the DOJ mishandling of the Stevens prosecution. Where is the army of special invetigators into the fraud of the banking mess. How can Mozzilla, who started both Countrywide and Indybank, walk with $400 m in fraud money.

    I don’t think the administration is corrupt but scared to death of either the banking industry or an industrial run on, or by, the TBTF banks.

  22. Anonymous

    “I want to see a negative before I provide you with a positive.”
    I can’t wait to see the actual reports in April if they will be available.

  23. Anonymous

    confused soul:

    A lot of the assets in question are mortgages or securities based on pools of mortgages. Maturity in this case is, as with bonds, the point at which the debt has been paid off. In your typical home loan with amortized interest, that means it’s the point in 10 or 20 or 30 years where the final payment is made. It could be a balloon mortgage where a chunk of the principal is due at the end, but typically it’s not like a zero-coupon bond; i.e. there will be cash flows over the life of the loan.

    That the asset is ‘toxic’ means simply that the expected value of the cashflows is much much lower than it used to be, not necessarily that they are totally absent. It depends. Perhaps a bond formed from a pool of mortgages (an MBS or Mortgage Backed Security) has a face value of $100 if 1% of the underlying mortgages are expected to default. Say we enter a recession and unemployment shoots up, increasing the default rate. The MBS will be worth much less.

    Basically, the toxic assets you’re reading about are loans whose value is very uncertain right now, since it’s not clear how much worse things will get, and how many people will ultimately default. When the loans are held for their full life (to maturity), some proportion will go bad and some will not, and ex post the value will be clear. Right now, though, investors require some degree of compensation for the possibility that the default rate could be higher than anticipated, and the securities could ultimately become completely worthless in some cases (depending on how the asset is structured). Arguably, this means the market value of the assets is less than the hold-to-maturity value.

    Keep in mind though that this contradicts theory: the market price should aggregate investor expectations about the future returns from the asset. So in fact the only way the HTM value is higher is if the economy improves more than the consensus predicts. This is certainly possible, but it’s by no means certain.

    Finally, compounding the problem is the fact that banks in many cases have not marked assets down even to the optimistic HTM levels, which are well below current balance sheet levels given the severity of the real economic contraction.

  24. doc bored 2 tears holiday

    God Bless America and The FASB and Mark-to-Mark Magic:

    A unit of ratings agency Standard & Poor’s said the key indicator to watch in the upcoming round of quarterly filings from large complex banks will be expansions in the valuation of level 3, or illiquid assets, after a US accounting body eased guidelines on mark-to-market valuations last week.

    Norwalk, Conn.-based FASB yesterday approved controversial rule changes that allow banks more freedom to use their own valuation methods when markets have become illiquid.

    “Political and special-interest pressures placed on [FASB] to change fair-value-accounting standards are unacceptable and very troubling,” the Investors Working Group said in a statement issued after its inaugural meeting yesterday.

    The IWG, an independent panel set up to recommend regulatory reforms, is led by former SEC Chairmen William Donaldson and Arthur Levitt Jr.

    The group said it is “dismayed by the lack of normal due process and the accelerated timeline for commenting on FASB’s proposals” to change mark-to-market-accounting rules.

    The accounting board has not yet responded to the IWG statement, Christine Klimek, a FASB spokeswoman, said

    The IWG said the new FASB rules could reduce the flow of reliable financial information and thereby erode investor confidence and increase capital costs.

    The group is co-sponsored by the Council of Institutional Investors of Washington and the CFA Institute Centre for Financial Market Integrity in Charlottesville, Va.

    The IWG expects to issue an initial report on ways to improve regulation by May.

    Boo and hiss:

    Reminder: Fair value hierarchy The fair value hierarchy for classifying fair value measurements of assets and liabilities, as set forth in Statement 157, is as follows:•Level 1 – Observable inputs, which are quoted prices for identical assets in active markets•Level 2 – Observable inputs, including quoted prices for similar items in active markets or for identical or similar items in inactive markets•Level 3 – Significant unobservable inputs, such as models that include management’s assumptions that cannot be corroborated with observable market data.

    I don't know what the hell this is, but WTF, a pdf withaccounting crap:


  25. Anonymous

    Anonymous at 10:07 in response to confused soul.

    While I think you were very clear with the situation as far as you explained, I am concerned about the supposed derivative side of the mortgage pools. It is my understanding (someone please correct me if I am wrong)that these financial institutions have derivatives written against the toxic mortgage pool that makes the toxicity worse. The derivatives are structured such that these companies are hanging out even further as the value of the mortgage pool decreases….hence the stupid attempts to keep prices inflated.

    Just like too big to fail is the quip that the public can’t handle the truth. I think it is more that we don’t have equitable rule of law (did we ever?).


  26. d4winds

    confused soul:
    You are not confused in the least. The price of the security IS the expected discounted cash flow stream given current information. That is the HTM value also. There is absolutely no difference.[1] Just the same, there can be very different calculated prices for illiquid securities because the expectations used may vary, while nonetheless all being consistent within their modelling frameworks with observed prices of other securities wherever possible. Explication on these points follows.

    The expectations of the discounted cash flows are taken by “measuring” the events, i.e., assigning probabilities to them, using probabilities implied by market prices (the so-called “pricing measure” or the “equivalent martingale measure”). The probabilities–market assessments–depend on the current information (e.g., given, as in 2006, that housing prices had never or fallen or given, as in 2009, that have fallen quite a bit).

    However, even controlling for the current information set, there may be multiple choices of probabilities that are equally consistent with observed prices on other securities.[2] In that case, one says that “markets are incomplete,” which is argot for “there are multiple, equally defensible prices for the same, as-yet-unmarketed security.”

    When securities are priced internally using a model having incomplete markets–as do those for many toxic assets, e.g., models for CDO tranches using Gaussian copulas and models for CDS spreads on such products, like the spreads written by AIG–then other firms potentially bidding on the security may obtain very different prices for it. The security is, as Yves expressed it, “paper anticipated to be illiquid from the get-go.”

    [1] The first chapter of Duffie’s Dynamic Asset Pricing (I can’t think of a lower level reference right now) makes this clear.
    [2] The famous Black-Scholes-Merton model, on its own terms (which, alas do not jive with the real world all that well)does not have incomplete markets. There is only one price. Ditto for traditional, GARCH-inspired stochastic volatility models. Exponential Levy models occasionally used for equities and Forex (log of asset is a Levy process) are inherently very incomplete due to indeterminateness of the Levy measure (one can always cure the indeterminacy ex hypothesi, of course, or with the existence of a marketed spectrum of liquid calls and puts assumed consistent with it).

  27. artichoke

    Yves, yes a vol smile is the usual way to patch the Black Scholes model to account for fat tails.

    Two more observations that occurred to me:

    1. The extrapolation problem may not be as great as it used to be, because we are close to or in the default zone already.

    Sad but true.

    2. The “Merton model” (I think of it as Modigliani-Miller), that equity is a call on the firm’s assets, undervalues stock when the firm does not wipe out equity at the time that net assets go negative. Citi’s stock price should be zero, but it’s positive because stock in an insolvent firm still has option value when government is propping the operation up.

    Following up d4winds’ comment above, yes the market for exotics, and thing one might not even consider exotic, is significantly incomplete. Thus plug-numbers are needed to mark to model. Correlation is a typical plug number that frequently is not well observable in the market. “Correlation traders” are trading things that depend on lots of more observable stuff (prices, volatilities) and correlation. Since correlation is the big unknown, that’s the essence of their trading. Given that you know all the other stuff, knowing a theoretical asset price is equivalent to knowing where to mark correlation, or observing an asset price is equivalent to observing implied correlation.

    The new FASB rule will let them put correlation where they think “reasonable”. A big invitation to what most would consider fraud, most likely.

  28. Anonymous

    Anon at 10:07 7 April;

    anon at 12:20 Apr 8 (psychohistorian);

    and d4winds at 06:46 8 April:

    I thank you all for the interest you displayed in my query and most especially for the effort expended to explain things for me.

    Anon at 10:07, your answer was the easiest to understand, and helped immensely, thank you.

    d4winds, I did appreciate what was obviously a reply entailing considerable effort on your part. However, as I am a mere amoeba in the evolutionary hierarchy of things financial, I fear I cannot say with certainty that I understood your answer. I am reminded of Heisenberg’s uncertainty principle, about which the only thing I ever really grasped during my undergrad chemistry was that electrons do not travel on sharply defined orbits. Similarly, to the degree I understand you ( a big if!) you are telling me that prices for certain assets under certain circumstances are also not sharply defined. (From the sound of things, however, prices are still easier to estimate than the position of an electron, which is comforting to some degree I suppose…. ; – ) )

    I must confess your notation re Levy models sailed right over my head, but I have taken copious notes, and shall apply myself diligently in an attempt to follow at least the gist.

    Truly, I am glad of the effort everyone put forth, ta for now,

    ‘confused soul.’

  29. Anonymous

    Negative basis doesn’t mean that CDS is “underpriced”. See many of the 2007-2008 CDS basis “arbs” that blew up.

    In short: CDS ~= (risk free rate)* + (credit spread) + [L]

    [L] can be interepreted in many ways. Sometimes it is just bid/ask noise, sometimes it is the risk premium that CDS writers demand for writing a derivative, sometimes it is just the way that the market is leaning temporarily. I personally appreciate that a large portion of [L] can be thought of as a liquidity premium (or discount).

    These days, CDS (and most derivatives in general) are actually much more liquid than cash bonds, hence they trade at negative basis in distressed environments.

    The assumption that [L] ought to converge close to zero, or that liquidity would almost always return to the market despite short run stresses led to spectacular blowups at various CDS desks (ML and DB are rumoured among the casualties). The idea was that [L] should be close to zero when [L] is negative, buy the (implied cheap) cash bond, sell the CDS, hedge your risk free rate, and you collect [L]. If, however, [L] doesn’t converge to zero, and actually blows out because people want the liquidity (and seniority in distressed situations, which I would roughly equate to liquidity in the event of shitstorm — thank the ISDA, this is a Good Thing, IMHO), then people get margined called, for which they at some point if sufficiently underwater will unwind the trade by buying back the CDS, selling the bond… further blowing out [L] which sucks for every other desk that has the same trade on, etc.

    *now that US Treasuries and other sovereigns trade at non-marginal CDS spreads, one might infer from the classical assumptions that govies are no longer “risk free”. This would be consistent with recent (see Russia, LatAm) and historical (US default via currency devaluation in 1930s and arguably end of Bretton Woods 2) observations that despite their vaunted “ability to always print more money”, sovereigns do in fact, sometimes default on their payment obligations.

    **web searching for “negative basis CDS” leads to better and more cogent explanations.

    *** big ups to the guy who taught me about [L] during my boot camp training on wall street. He was a pioneer in index arb, and had a fantastic story about who stole the idea for CAPM. Unfortunately, I forgot his name, and I think he might have recently been indicted for something :/

    ****also big ups to YS, whom I do not always agree with but is more often than not trenchant, insightful, and willing to take the establishment to the woodshed.


Comments are closed.