At VoxEU, Luigi Spaventa makes a forceful case that the remedies to the credit crisis are provisional at best and more work needs to be done:
The global financial system may be caught in a downward spiral as market and funding illiquidity reinforce each other…
Prolonged financial distress, which has now lasted for almost a year, is debilitating the financial system and risking a full-fledged crisis. Central bank interventions have thus far prevented worst-case outcomes, but they have alleviated symptoms rather than the underlying causes. Financial intermediaries are still in the process of shrinking their balance sheets, thus activating a channel of transmission of financial distress to the real economy…
The immediate problem is a spiral of forced deleveraging and illiquidity, as the link between market and funding illiquidity strains balance sheets. Proposed remedies are either insufficient or unsatisfactory…
The immediate problem is the disorderly reaction to the unprecedented growth of the financial system’s leverage and its exposure to risk. As demand for asset-backed securities has disappeared, prices have collapsed without finding a floor. Banks are reporting losses that strain their capital positions. The loss of market liquidity affecting all classes of debt securities directly or indirectly owned by intermediaries has translated into a sharp decline of funding liquidity, the more so because short-term debt issued on wholesale markets has become a major component of banks’ funding. The forced adjustment of banks’ balance sheets could, in the worst case, result in a credit crunch with painful consequences on the real economy.
Can we break the link between the illiquidity of banks’ securitised assets, which prevents their orderly liquidation, and the shortage of funding liquidity, which is the driving force of the negative feedback originating from the process of deleveraging?
For funding liquidity, emergency liquidity support from central banks has helped lower the temperature in the worst moments, but it is not a long-term solution….Capital increases are also insufficient to break the spiral, as injections of capital may prove inadequate only a few weeks after their announcement.For market liquidity, suggested remedies are equally inadequate. Mandated full disclosure of losses might reduce uncertainty, but unless market liquidity is instantly restored, full disclosure of the situation at time t offers no guarantee that it will be the same at time t+1. Similarly, retreating from marking financial products to market or model during this time of crisis would face a number of difficulties.
The feedback between market and funding liquidity problems demands more radical pre-emptive solutions. As long as “there is no immediate prospect that markets in mortgage-backed securities will operate normally”, “the situation will improve only if the overhang of illiquid assets on the banks’ balance sheets is dealt with” (Bank of England 2008). In creating its Special Liquidity Scheme, the Bank of England has moved to serve as the “market maker of last resort.”
The scheme allows banks and building societies to swap some of their illiquid assets, including debt securities rated no less than triple A, for specially issued Treasury Bills for up to three years. Eligible securities will be valued at market prices, if available, or, if not, at a price calculated by the Bank, with haircuts for private debt securities. Changes in market prices or in valuations will require re-margining. The credit risk will remain with the banks, so that there will be a loss for the lender only if the borrower defaults and the value of the collateral falls below that of the bills originally acquired in the operation.
Is the initiative bold enough? The scheme does not set a floor for assets’ market prices and uses market prices to value collateral, despite the fact that during a negative bubble they do not reflect fundamentals. Downward instability may moreover occur if haircut discounted collateral values trigger a convergence process for market prices requiring repeated re-margining.
In CEPR Policy Insight 22, I recommend the creation of a publicly sponsored entity that could issue guaranteed bonds to banks in exchange for illiquid assets, drawing on US Treasury Secretary Nicholas Brady’s solution to the Latin American sovereign debt crisis in 1989. This new entity, preferably multilateral, would value assets based on discounted cash flows and default probabilities rather than crisis-condition market prices.
As a firm floor is set to valuation and illiquid assets otherwise running to waste are replaced by eminently liquid Brady-style bonds, funding difficulties and, at the same time, the market liquidity problems besetting the banks’ balance sheets would be removed. Shielding the banks’ assets from the vagaries of disorderly markets is a necessary condition to dispel the uncertainty that prevents a proper working of credit markets.
I’m curious to get reader reactions, but I suspect that this model breaks down under further examination. First, a fair bit of paper that banks hold is sufficiently arcane that valuing it for the purposes of going into this entity is problematic. How do you value a CDO (I’m not suggesting they can’t be valued, but I suspect that valuation ranges using reasonable assumptions would be very wide). Does this mean going back to the repudiated credit models of the rating agencies?
And even if you exclude the more complex structures, coming up with a cash-flow-based value when there is still a high deal of uncertainty in the trajectory of the housing market (particularly in markets like Florida with massive inventory) involves a great deal of artwork. Moreover, the sovereign loans that were restructured in the early 1980s were pretty homogeneous, as far as terms were concerned. The mortgage assets are far more diverse, and the underlying collateral is very heterogeneous, which makes for a difficult valuation process.
As I understand it, the original Brady bond process arrived at structure and pricing via negotiation, not via one side putting out an offer based on its valuation (which seems to be the process here) and seeing who shows up to submit colllateral is asking for adverse selection: you’ll get offered paper when you are offering too much.
Nevertheless, this is an interesting idea, and if there was a better finesse for the valuation issue, this could be a useful remedy.
13 comments
Comments are closed.
Ultimately it’s about trust… And if you are dishonest and people sense that or know that, they won’t trust you. Why don’t we call this what it really is… an honesty crisis.
If the author is proposing a Friedmanesque “gift of money” to the banks, he should just say so. Otherwise, I have no idea what he’s talking about.
If he’s so certain that the paper (which is many kinds of paper) doesn’t reflect fundamentals, he should raise a fund to compete with Blackrock. He might first want to look up the definition of `level 3 asset’, though.
Perhaps the US could issue CNY denominated debt with some knid of pratial guarantee from the Chinese Government.
“…including debt securities rated no less than triple A, for specially issued Treasury Bills for up to three years. Eligible securities will be valued at market prices, if available, or, if not, at a price calculated by the Bank, with haircuts for private debt securities…”
Gosh, two red flags in two sentences. “Triple A” means you have to trust the untrustable credit agencies. “Calculated by the Bank” means the IBs will rip the government off.
The basic rule should be: if the government cannot value the things on its own, it should not be taking the stuff on its books.
I agree as formulated this isn’t a great concept, but there is a germ here that tweaked and used in combination with other ideas, might be useful.
I assume, as “a” says the bonds will be richly priced in order to provide a silent bail-out to banks.
Maybe I give the public too much credit, but I’m not sure they can be persuaded to buy up bonds issued by some super-SIV that buys up all the junky asset-backed securities the banks don’t want. I know the US government used words like “patriotism” and “duty” to persuade the public to buy US savings bonds during WWII, but that was to finance a popular war.
“Can we break the link between the illiquidity of banks’ securitised assets, which prevents their orderly liquidation, …”
Yes. Put the payment history, credit checks, and comps in machine-readable form and shop the database to potential investors. It cannot possibly cost more than about 5 basis points or so. That information would justify a reasonably fair sale price, and certainly avoid a panic fire sale.
Instead, the banks are choosing to pay 300 bps/year to the Fed to liquidize their assets. Clearly market liquidity is not the real problem.
*sigh* There are two (at least two) fundamental misconceptions embodied in Spaventa’s proposals as presented, to me. I’m not saying that to lay into him, it seems his goal is constructive. If we are to ‘fix a problem,’ however, it would seem best to find what the problem is, and to understand it on its own terms.
Problem No. 1 is the repeated assertion by Spaventa that we are vexed by ‘a liquidity crisis’: What we have here is a solvency crisis. If these securitized assets are valued either where the market presently assesses them (if poorly on too little information) or in relation to any historical norm for the underlying debt and properties, then the holders of these assets are, nearly all of them, insolvent. That is the real problem, and we should be talking about how to restore solvency, to whom, on what terms, and under what regulatory regimes to avoid a redo. Calling this ‘a liquidity crisis’ embeds the assumption that as liquidity ‘returns’ to the system the price of the asset-backed securities will ‘recover’ toward their ‘underlying values,’ i.e. return to near-face value. This strikes me, again by using historical norms, as an absurd assumption, as the face-values of these securitized golem corpses are far too high. Liquidity will right itself when instead the face values of punk assets adjust downward to real world values—which hasn’t happened. The reason these securities are illiquid is that anyone with a functioning olfactory systems knows they aren’t remotely worth face, and won’t be again ever.
Problem No. 2 is the logical extension, that if any public authority guarantees bonds to be swapped out for misvalued securitized debt, the result is the absolute definition of ‘a gift of money’ in the estimable phrase resurrected by Steve in his post above. Rather than face a solvency crisis, we socialize the concealed loss [and presumably then inflate it away, the logical next assumption which didn’t make it into Spaventa’s Two-Thirds Solution]. In the actual Brady bond context, the losses _were substantially factored in_ when the original debt was written down prior to being swapped out for internationally negotiable bonds. —But the whole idea of ‘swapping out illiquid securities’ is specifically designed to avoid price discovery until their present holders are relieved of the fatal inconvenience of possessing them. After that, it’s the problem of this issuing institution conveniently multi-lateral and hence not subject to the ire of any one set of voters or a direct liability on any one public fisc. This seems something like the MOLECH super-SIV concept of Paulson, just exponentially bigger. Yes, pile alllll those rotting bonds and their risk in a warehouse on Diego Garcia and then pretend that no one government has any obligation to them. . . . Right.
I’ll know that we are getting serious in suggestions to end our present crisis when we stop talking about liquidity and start talking about solvency.
“Can we break the link between the illiquidity of banks’ securitised assets, which prevents their orderly liquidation, and the shortage of funding liquidity, which is the driving force of the negative feedback originating from the process of deleveraging?”
In other words, “can we prop up the value of the underlying assets to something near what the banks paid for them?” No thanks. Houses have fallen and foreclosures have increased dramatically, thus MBS are worth less. This is the reality of the situation. Let’s not conflate an asset price decline with illiquidity.
I completely agree with Richard. Thats why I still think we’re in the first inning of this situation. No one wants to look down the black hole of what has happened. This is going to be a global disaster.
I’ll just ask some questions referencing the points that struck me. Spaventa wrote: “value assets based on discounted cash flows and default probabilities rather than crisis-condition market prices.” Yves Smith countered: “coming up with a cash-flow-based value when there is still a high deal of uncertainty in the trajectory of the housing market (particularly in markets like Florida with massive inventory) involves a great deal of artwork. Moreover, the sovereign loans that were restructured … were pretty homogeneous…” Is the problem “epistemic” – we cannot get the data to figure – or is it “ontological” – the stuff keeps changing? One of the recurring explanations I cross for why the mortgage/housing/CMO problem cannot be fixed is that as collections of inputs (mortgage), the holders of the CMO’s don’t know their ultimate payers (the home purchasers). But surely a paper trail has to exist – some specific mortgages had to be identified and collected as the source of the funds supporting the CMO. I don’t suppose that these records have been obliterated.
It seems to me that more of our problems have to do with whose interest is to be served rather than what options are availabe – decomposing complex paper is probably labor intensive, but there ya go.
I have no more interest in giving money gifts to the people who caused this than anyone else, but there is a different way of avoiding that: Fix the problem, structurally, and resort to confiscatory taxes to insure that no moral hazard is encouraged. Besides, people who cannot convince themselves to get up and go to work for less than a few hundred thousand a year are people who probably need to learn something about work ethic and, as someone above noted, honesty.
dave,
The problem is that the value of this paper depends on future events. We used to think that those future events were unknown, but predictable. We have already reached territory outside the realm of our predictions.
Now any serious prediction of the true value of that paper has to include a wide range of possibilities — some of which imply that our financial institutions are solvent and some of which imply that they are not. Now add in what Soros calls reflexivity: our actions today may well have a determining effect on what the value of the paper is in the future. Some people like to call this fact “liquidity” (in other words, if we do the right thing today, we will find that the problem is just one of liquidity), but like most of the other commentators on this blog, I find that point of view optimistic.
To Dave Raithel, a core aspect of the problem you raise in your post is that the ‘basic data matrices’—for example mortgages—in these CDO intstruments are highly disparate in form and comparatively immobile. Individual mortgates vary widely in their variables: first mortgage; re-fi; seconds; thirds; typically these loans were issued by different lenders; local property values; regional asset values; income stream of mortgage payer; behind in payments or stopped paying; viable with a mod or not viable in any form; etc. These mortgages are held in different legal jurisdictions with _substantial_ variations in how those variables may be manipulated or re-indexed. Furthermore, these individual matrices are super-glued into their constitutent [in]securities such that they cannot be stripped out, mortgage/matrix by mortgage/matrix, if they sour and crumble. Yes, somewhere in the provenance chain, someone actually holds the note, and legally could negotiate a mod—in theory. However, the authority to renegotiate in many of these CDOs is delegated to the _payment servicers_ who by no means always have possession of the actual notes, and in no way have the staff, the experience, and some would say the competence to offer mods en masse. BTW, these servicers, structurally essential to the internal functioning of these CDOs, are bleeding money and (though we don’t speak of it yet) likely to bankrupt out _themselves_ long before the underlying mortgage payment streams in these instruments dry up: ‘blood’ to the ‘head’ of these instruments will be gone in all probability while their ‘body’ is still alive if missing limbs, killing the securities even while the underlying ‘cells’ are many of them still quite functional. The securitization process was designed to make it easy to get the mortgages in—that’s where the fee-profit was—and difficult or not designed AT ALL to get the mortgages out—that’s where the cost-loss risk is.
Oh and BTW, in more cases than you would think the actual paper trail for individual mortgage/matrices is missing links or entire files. Folks kept passing the mortgages around with little interest in keeping the backoffices current and matched—so papers dropped between the cracks too often.
CDOs are the Devil’s Own Cat’s Cradle: stick your hands in for the money and you can’t get the money, your hands, or your life disentangled again. Is financial innovation a laugh riot, or what? It is until that financier with his hands in the poison strings has his arms locked around our head in the process; now, we’re not breathing, either . . . .