Creeping Recognition that Regulation Has Created a Liquidity Death Star
Reason number one (by far) that I believe that clearing and collateral mandates increase systemic risk is that they transform credit risk into liquidity risk. Large price moves during stressed market situations require those with losing positions to make large variation margin payments in a very tight frame. These payments need to be funded, and funded immediately. Thus, variation margining causes spikes in the demand for liquidity. Furthermore, clearing in particular creates tight coupling because failures-or even delays-in making VM payments can put the clearinghouse into default, or force it to liquidate collateral in an illiquid market. The consequences of that, you should shudder to contemplate.
To be somewhat hyperbolic, clearing mandates create a sort of liquidity death star.
Recognition of how dangerous spikes in liquidity demand precisely when liquidity supply evaporates creates a major systemic risk is sadly insufficiently widespread, particularly among many regulators who still sing paeans to the glories of clearing. But perhaps awareness is spreading, albeit slowly. At least I hope that this Economist article indicates a greater appreciation of the collateral issue, although it fails to draw the connection to central clearing, and how clearing mandates can dramatically exacerbate collateral shortages:
WHEN the financial system teetered on the brink of collapse in 2008, the biggest problem was a lack of liquidity. Banks were unable to refinance themselves in the short-term debt markets. Central banks had to step in on a massive scale to offer support. Calm was eventually restored, but not without enormous economic damage.
But has the underlying problem of liquidity gone away? A research note from Michael Howell of Crossborder Capital argues that, in the modern financial system, central banks are no longer the only, or even the main, providers of liquidity. Instead, the system looks a lot like that of the Victorian era, with banks dependent on the wholesale markets for funding. Back then, the trade bill was the key asset for bank financing; now it is the mysteriously named “repo” market.
. . . .
Bigger haircuts mean that borrowers need more collateral than before in order to fund themselves. “When market volatility jumps, funding capacity drops in tandem and often substantially,” writes Mr Howell. The result, a liquidity squeeze at the worst possible moment, is a template of how the next crisis may occur (although regulators are trying to reduce banks’ reliance on short-term funding).
And again, it is at these times when the need to fund VM payments will kick in, exacerbating the liquidity squeeze. Moreover, clearing also ties up a lot of the assets (e.g., Treasuries, or cash) that firms could normally borrow against to raise cash. Perversely, that collateral can be accessed only if a clearing member defaults on a variation margin payment.
Just what the liquidity supply mechanism will be in the next crisis in the new cleared world is not quite, well, clear. As the Economist article (and the Crossborder Capital note upon which it is based) demonstrate, central banks lend against collateral, and the collateral constraint will already be binding in stress situation. Presumably central banks will have to be much more expansive in their definition of what constitutes “good” collateral (a la Bagehot).
It still astounds me that even though every major financial crisis in history has been at root a liquidity crisis, in their infinite wisdom the betters who presume to govern us thought they were solving systemic risk problems by imposing a mechanism that will sharply increase liquidity demand and restrict liquidity supply during periods of market stress. That should work out really, really swell.
I’ve been wondering lately whether money and collateral are closely related, maybe even parts of the same thing. Essentially, there’s an extension of credit on both that is required for the operation of the economy. This extension of credit is a private function, usually, and can fail during crises. Am I thinking correctly about this?
Comment by Highgamma — October 24, 2015 @ 3:31 pm
I think its a fair assumption that most of the goal is to prevent the trades from occurring to begin with. Collateral requirements can only become liquidity risk if the trade is transacted – no transactios, no risk. One can decide for one’s self is this a good thing, but, it is likely the overriding goal.
Comment by MS — October 24, 2015 @ 4:07 pm
@Highgamma-Your intuition is spot on. Currency retains its nominal value by fiat. A dollar bill is worth a dollar. Private monies attempt to achieve this fixed nominal value through collateral. Gorton argues that the creation of senior CDOs through collateralization and various credit enhancements was just another attempt to create a private money.
The problem is that these private monies work only if the value of the underlying collateral itself is at low risk of falling below the face amount of the claim. Despite all of the clever attempts to do so, no one has come up with the magic formula for doing this, and the periodic failures are associated with crises. AAA CDOs backed by subprime is just the latest example.
Variants on narrow banking proposals, which have been around forever, are an attempt to get around this problem by sharply limiting the risk on the asset side of the balance sheet of banks who can issue fixed nominal claims. Maybe they would work if banks were the only game in town. But there are likely to be pressures to create private monies outside the banking system using other forms of collateral.
But even narrow banking proposals might not work, because they usually would allow investment in government securities, and those can be problematic too, as the European crisis shows.
Maturity transformation is dangerous and is a form of scam. Requiring all deposits to be redeemable at fixed times would go a long way toward fixing this (and no, government bonds are not risk free).
If you simply take away the dollar printing press lots of these kinds of issues become obviously bad.
Comment by Dave — October 25, 2015 @ 5:11 pm
[…] Professor warns that clearing and collateral mandates create a ‘sort of liquidity death star’ because […]
Pingback by Central banks have to get off the liquidity death star before they can destroy it | Critical Macro Finance — October 26, 2015 @ 2:33 am
Credit to the pension funds in Europe for getting themselves organised and lobbying hard in the European Parliament. They gained a temporary exemption from the clearing mandate for this very reason – although their exemption might expire before the mandate actually starts!
I’m certain the regulators understand the clearing mandate will not reduce overall risk (and will probably increase it) but their political masters cannot ever be allowed to bail out a bank again if they want to be re-elected. Therefore they had to take some superficial action to say “look, Counterparty risk has gone so we won’t have to bail out the banks. Please don’t look over there at Liquidity risk, there is nothing to see”.
This same attitude to Counterparty risk is being taken towards non-cleared derivatives with bilateral IM and VM requirements. At least the regulators can pat themselves on the back when the next crisis is caused by one-way CSA’s or other liquidity requirements – they will have cured the patient from the curse of counterparty risk.
Comment by GreenwichMeanTiger — October 26, 2015 @ 10:38 am
Question: in what way is central clearing worse than the old bilateral clearing system? That’s always the objection people raise when I bring this up.
Comment by Rob — October 26, 2015 @ 4:54 pm
@Rob. Like Tolstoy’s remark about unhappy families, cleared and uncleared markets are flawed in their own ways.
The biggest advantage of a cleared market is that management of a defaulted position is less chaotic than in a bilateral market. Though as I’ve pointed out in stuff I wrote in the 2009-2011 time frame, it would be possible to get this benefit without mutualizing credit risk, and creating the kinds of liquidity strains that clearing does.
Derivatives credit risk is lower in a cleared market. But as I’ve pointed out repeatedly, that’s probably a push because reducing derivatives credit risk increases other credit risks.
Cleared markets pose greater liquidity risks. In my view, that consideration is decisive.
Finally, it shouldn’t be either-or. It is possible to clear some things, and not clear others, and it makes sense to clear some things and not others. In work I did in the 2000s, I argued that the pre-crisis division made economic sense. My problem is mandating the movement of everything to a cleared (or collateralized) system. Not only are the liquidity problems exacerbated greatly thereby, but there is no guarantee that a clearing system that works well for listed products will scale up and work even remotely as well when extended to the larger, and more complicated, swaps market. As the expression goes, the dose makes the poison. What is medicine in moderate doses can be fatal in large ones. Furthermore, there is a great risk of increasing risks by clearing products that are not suitable for clearing.
Interesting post, I agree. Many are wondering how CCPs came to be the holy grail for regulators. May have to rely on central bank LLR, which would be an irony, now that bail-in capital has supposedly cured TBTF institutions. See more elaborate comments on this in my article Shadow banking: policy challenges for central banks https://www.gfsi.ey.com/the-journal-x.php?pid=13&id=94
Comment by thorvald — November 4, 2015 @ 7:16 am