How much liquidity does a bank have to hold to protect it from a credit crisis? Enough to finance itself for six months? One month? One week?
Top marks if you answered, “a week.” That, at least, was the regime in the UK for big banks until the August credit crunch and there were similar rules in other big markets.
Think for a moment. Just five working days. No wonder the financial system came perilously close to collapse. No wonder the central banks had to flood the market with liquidity to make up for the ready cash the banks didn’t have themselves. It has since become accepted wisdom that the rules on bank liquidity need to be tightened. Indeed, regulators have been pushing banks to have a bigger liquidity cushion. The UK’s Financial Services Authority (FSA), for example, now wants big banks to have a three-week cushion. But how exactly does the current regime work? And what is the best way of fixing it?
The first thing to note about the regulation of liquidity is that it is separate from the more familiar regulation of capital adequacy. Capital adequacy regulation is designed to stop banks going bust because they incur excessive losses. Liquidity regulation seeks to make sure that banks don’t run out of cash because their sources of funding dry up.
The two types of problem can, of course, be linked—funding can dry up when a bank is losing too much money. But these problems don’t have to go in tandem. Indeed, the August credit crunch was largely one of liquidity. Most spectacularly, the UK’s Northern Rock Plc. had plenty of capital but was still in danger of running out of cash. Until a few weeks ago, liquidity regulation was considered much less important than control of capital adequacy. While the so-called Basel rules provide a global standard for bank capital, liquidity has been left, almost as an afterthought, to the national regulators.
That said, most of the big countries have gravitated towards roughly the same approach, with separate regimes for big banks, smaller banks and investment banks. All are too weak. The mechanism for ensuring banks have enough cash to finance contingent liabilities, for example providing emergency funding for the notorious conduits that caused so much havoc in the crunch, has also been too soft.
One size doesn’t fit all
Look at each of these elements in turn. Big banks are typically required to hold enough high quality liquid assets to fund their gross outflows for a specific period of time. In the UK, this regime applies to the top 12 or so banks. The definition of a high-quality liquid asset usually is anything (say a government bond) that a central bank will accept in ordinary circumstances as collateral in return for cash. In the UK, banks had to hold enough liquidity to cover just one week of outflows. That was clearly ridiculously low. The funding horizon has now been moved up to around three weeks.
Smaller banks are required to match the maturity of their liabilities and their assets. The idea is that if you have $100 million (Rs398 crore) of liabilities coming due in the next month, you also need to have $100 million of assets that mature in the next month. Otherwise, you have a mismatch. While the general idea makes sense, regulators typically focused only on matching liabilities and assets over a one-month period. The thought was that any liquidity crunch was likely to be short-lived so one didn’t need to worry about a mismatch a few months out. This proved optimistic. As the days ticked by, a maturity mismatch a few months out—of the sort Northern Rock experienced—turned into an immediate crisis.
Regulators have therefore been scurrying around to try to make sure their banks have matched assets to liabilities in the slightly longer term. FSA, for example, is now focussed on a 75-day horizon.
Investment banks are required to hold enough “collateral” to meet their short-term funding needs. The collateral can be anything from cash to subprime mortgage-backed securities, but a haircut is applied to each type of asset. Illiquid assets, such as subprime, get a bigger haircut than liquid ones, such as US treasury bills. Again, the regime has logic. The snag is that it seems like some illiquid assets were getting a trim when they needed a full short-back-and-sides. Nowadays, regulators such as the US Securities and Exchange Commission and FSA are requiring investment banks to hold more cash as a cushion.
Banks are also required to have enough liquidity to deal with contingent liabilities— such as the back-stop credit lines they provided to conduits. The snag is that regulators didn’t think these credit lines were that likely to be called upon in a crisis. The UK’s FSA, for example, thought the chance of a crisis was something between 20% and 60%. Not only is this a huge variation, it is precisely in a crisis that one might expect the credit lines to be of help. FSA now rates the chances of a crisis at close to 100%.
Charting a course
The ad hoc adjustments made by regulators during the crisis may have been beneficial. But it would have been far better to have a safer system in the first place. What’s more, the act of toughening the rules in the middle of a crisis may even have exacerbated it. Banks were being told to get more liquidity just when the markets didn’t want to supply it. That may have encouraged them to hoard cash, further gumming up the interbank market. Arguably, one should have really tough rules on liquidity in normal times and relax them in a crisis, rather than the other way around.
In any case, now that the worst of the crisis appears to be receding, it is time for a new system. Such a regime should require big banks to hold liquidity to protect against much more than five rainy days; make sure the maturity of assets are reasonably matched for a few months, not just a few weeks; apply a bigger haircut to illiquid assets when assessing their value as collateral; and take account of the fact that, in a crisis, a contingent liability is likely to become a real one.
Many banks would, no doubt, complain that such tightening would constitute a regulatory overkill. Requiring them to hold a plumper liquidity cushion would certainly push up their costs, cutting their currently very high returns on equity.
But banks cannot be allowed to operate without rules. A banking crisis would cause such massive dislocation to the broader economy that the central banks will always ride to the rescue if they can. Indeed, that’s what happened this time, when the Federal Reserve and the European Central Bank flooded the market with liquidity.
This may have saved the day. But it was not a cost-free exercise. Even if the additional funds can be withdrawn before they cause inflation, the market has learned the extremely bad lesson that it doesn’t need to provide liquidity for itself. The central banks will do it instead.
Regulation is always a second-best option. But it is a necessary correction for market failures. And in this case, the benefits would be significant. There are three big ones. First, if banks had a fatter cushion, the chance of a future liquidity crisis would be reduced. Second, if illiquid collateral needed a bigger haircut, illiquid assets would fetch lower prices. That would make it harder for investment banks to manufacture exotic structured credit products, whose main purpose seems to be to generate fees for themselves.
Finally, all those financing practices that merely rely on borrowing short, lending long and expecting the authorities to pick up the tab if things go wrong would never see the light of day. One might say good riddance.