John Dizard, in the Financial Times, tells us how the Fed is giving the wrong medicine to the colicky credit markets. But even though Dizard, in effect, argues that the Fed doesn’t get it, he isn’t taking up the verging-on-hackneyed lament, “the markets are a mess, so the Fed needs to open the floodgates.”
Dizard’s reading is more sophisticated, and pinpoints why the Fed’s efforts to revive debt markets has and will likely continue to be ineffective. All the talk of a crisis in the shadow banking system (this was a topic at the Jackson Hole conference) has somehow not led the Fed to take a new course of action. The central bank instead appears to be a walking example of the saying, “If the only tool you have is a hammer, every problem looks like a nail.”
The Fed’s conventional remedies presuppose that the problem is a lack of liquidity in the banking market, ergo, give more liquidity to banks. But Dizard points out the two failings of that approach. First, the Fed’s actions are not ameliorating conditions in the securitized credit markets because they are (by definition) not tied to bank balance sheets. Second, even measures that supposedly help stressed banks (the use of the discount window) don’t. They are designed to help in short-term liquidity crises, but are ineffective when bank are reining in lending out of current or prospective damage to their balance sheets. The discount window, as Dizard explains, does nothing to improve capital adequcy.
As often is the case, it’s easier to diagnose an ailment than to treat it. Dizard cites a 2004 legal study of permissible Fed actions that states the Fed can lend to just about anybody and accept pretty much anything as collateral, provided the situation is “unusual and exigent.”
Although Dizard doesn’t take his idea much further, it is in the territory of Wiillem Buiter’s and Anne Sibert’s recommendation that central banks act as market makers of the last resort:
The market maker of last resort function can be fulfilled in two ways. First, outright purchases and sales of a wide range of private sector securities. Second, acceptance of a wide range of private sector securities as collateral in repos, and in collateralised loans and advances at the discount window.
It isn’t clear that Dizard is advocating the first part of the market maker of the last resort concept, but he is suggesting the second may be necessary.
However, the circular problem is: how do you value the collateral when no one is making bids? If there were a functioning market, price determination would not be an issue, but then there would also be no need for the authorities to lend a hand. Because the central bank would be taking risk in its purchase and lending operations, this isn’t an academic exercise. Indeed, valuation in a vacuum is a no win situation. The same price could be attacked as being too high (and read as as effort to shore up market prices or bail out players that need to take their lumps) or too low (the central banker being too preoccupied with managing its own risks rather than salvaging the market).
From the Financial Times:
Wall Street is beginning to realise just how deep in the kimchi it is. The assumption has been that when the Fed finally came to its senses, discarded its canonic Dynamic Stochastic General Equilibrium models, and opened the floodgates, enough cash could be provided to bail out the dealers and banks.
Sure, some layers of management would be fired, some individuals prosecuted and then immolated in the customary autos-da-fe, but the system would be saved.
What, though, if the floodgates don’t work any more? Back in July, I wrote that Numerix, a big supplier of credit analytics to the Street, had never been ed by the Fed to find out how its models worked. Neither, it now seems, had anyone else. It was only in mid-August that Ben Bernanke, Fed chairman, received a briefing from the credit derivatives buy side about the effect of new market structures in deepening the crisis.
The Fed’s econometric models are based on the assumption that the conventional, on-balance-sheet banking system is an effective mechanism through which to increase or contract the availability of credit. Unfortunately, that’s not true now.
It has now dawned on the people on Independence Avenue that the connection between their open-market operations and credit going out to the private sector is not working. Two problems: first, in recent years, most of the credit growth has been provided through a securitisation mechanism that is broken. The necessary trust in ratings agencies, credit insurance providers and valuation models isn’t there any more. Second, the banks can only increase their lending based on new provision of liquidity by the Fed if they have the balance sheet capacity to do so.
They don’t, in many cases. Several of the key banks are capital constrained. They are simultaneously being forced by the accountants to take write-offs against now thin reserves, and to provide credit to customers drawing on back-up lines. There’s nothing available for net new lending to credit market makers or real economy people, though those calls on back-up lines are forcing an unsustainable short-term balance sheet growth.
“For the last couple of weeks,” says a credit hedge fund manager friend of mine, “no bank has been willing to take on a bond they can’t present at the Fed’s discount window.”
So what are the Fed and its big customers saying to each other? In the absence of intercepts, we can make use of some traffic analysis. David Small and James Clouse, as economists at the Fed, have between them published a long list of papers. But the most downloaded of those, recently, have been on a couple of seemingly obscure and hypothetical topics: monetary policy in a zero-rate environment and, lately, “The Scope of Monetary Policy Actions Authorized under the Federal Reserve Act”, published in July 2004, which was basically a legal review of what the Fed could do if conventional mechanisms weren’t working.
The key problem with cutting the Fed funds rate, as we have noted, is that at the margin, the bankers to the securitised credit markets can’t make use of the money. So the somewhat more sophisticated market people are now saying that the only way to unfreeze things is to cut the rate at which the Fed will lend directly, and destigmatise the use of the facility, so the banks can lay off some of their assets on to the Fed’s balance sheet.
The problem with this, as Small and Clouse pointed out, is that the credit risk of the paper placed with the Fed through the discount window remains with the banks. So the pressure on their balance sheets isn’t relieved. That means that the Libor rate is likely to stay up, or keep rising, even if the discount rate is cut. And Libor is the basis for most credit pricing.
There are similar restrictions on a long list of allowed “advances”, discounts and asset purchases. When banks lay off assets on the Fed’s balance sheet, that paper has a long rubber band attached representing the banks’ retention of risk.
There is, it seems, one little opening, under “Lending to Non-Depositary Institutions”. Apparently, “the Federal Reserve has the authority to lend directly to individuals, partnerships and corporations (IPCs), which could include depositary institutions, under sections 13(3) and 13(13) of the Federal Reserve Act . . . However, lending under these authorities is subject to stringent criteria in law and regulation . . . ” This lending, under which, apparently, the Fed assumes the credit risk, can only be done under “unusual and exigent circumstances” and requires the affirmative vote, most of the time, “of not less than five members” of the Federal Reserve board. But section 13(3), Small and Clouse say, “provides virtually no restrictions on the form a written credit instrument must take in order to be eligible for discount”.
This seems to me to be one of the few possible ways to jump-start the stalled securitised credit machine, since IPCs could include the key securities dealers. The Fed wouldn’t have to take on the riskiest paper. But since it’s now hard to place even the investment grade stuff, loopholes such as 13(3) could prove useful.