Bank Liquidity and the Solvency Illusion
Managing liquidity is an often underestimated imperative for any business (if you don’t believe me, keep your eye on the debt of energy companies as it comes due.) Liquidity concerns are especially important for highly leveraged financial firms like banks. But leverage is not a necessary condition for liquidity problems; Money Market Funds are not levered but still can suffer from runs due to their maturity mismatches.
So what is liquidity? For most businesses, liquidity is the ability to meet one’s obligations as they come due. Currency in your fist, of course, is the best source of liquidity, followed by other assets such as accounts receivable and inventory that will soon become cash or can be readily sold. Less liquid assets like plant and equipment are tertiary sources of liquidity. They can be sold to get cash only with great difficulty, or they can serve as collateral for a bank loan.
For a bank, liquidity is more problematical. Since bank assets almost always have longer maturities than bank liabilities, large banks need to replace maturing debt nearly every day. In good times, this process is automatic, as depositors and short term lenders “roll” out of maturing obligations into new ones. But if depositors suddenly sense a problem for any number of reasons real or imagined, they may decide not to roll their deposits. Then, the bank must do one of three things to come up with cash; find another creditor, sell an asset, or borrow from the central bank.
When there’s even the slightest hint that a bank might be in trouble, bank creditors have every incentive to pull their deposits. There is no upside to maintaining that deposit, only downside. To bank creditors, the bank looks like a “black box.” (This is a prime example of what economists call “information asymmetry,” a concept made famous in an article on used cars by Nobel Prizewinner George Akerloff. His spouse is also a pretty good economist.) Shoot first, ask questions later is always the default strategy. That’s why liquidity for all banks can evaporate so quickly when any individual bank is perceived to have a problem. And that’s why it has always been crucial for the Fed to stand ready to backstop any bank with funding problems — to prevent a problem from escalating into a systemic panic.
In a panic, all banks must scramble to find funding. In extreme conditions, as in 2008, a bank can be reduced to selling assets to raise cash to pay its maturing debts. The problem is that, like the plant and equipment of industrial companies, many of these assets are illiquid. Asset markets that once appeared liquid can “freeze up” suddenly when doubt arises about asset valuations or counter party strength. As a bank desperately tries to raise cash by selling assets, buyers step away, and asset prices plummet. On such occasions, panic selling can drive prices to levels that, at least on paper, can wipe out a bank’s equity. In the absence of good prices and good information, a bank’s solvency can be called into question. (I define solvency to mean simply that the firm has more assets than liabilities.) And so is the solvency of many other banks in the system.
Some folks like to pretend that there is a clear distinction between liquidity problems and solvency problems as if one could be addressed independently of the other. In the long term, there is certainly a big distinction. But in the very short term (the time frame in which these judgments must be made) it is really a distinction without a difference. There is neither adequate time nor information to conduct effective triage.
In an impressive recent analysis, Laurence Ball of the Center for Financial Economics argues that Lehman Brothers may not have been insolvent prior to its 2008 bankruptcy filing. It was, it seems, a much closer call than the authorities or the financial press admitted at the time. His paper underscores the difficulty of differentiating between a liquidity event and a solvency event.
Here’s a thought experiment that might help to illustrate the liquidity / solvency symbiosis.
Let’s say the year is 2006 and you want to buy a house down the shore. The house costs $1,000,000. (Seaside just ain’t what it used to be.) So you take out one of those interest only mortgages for $1,000,000 and buy it. You figure you’ll earn enough in summer rentals to cover the mortgage payments.
One day in 2012, Sandy comes storming through. Your house survives, but is damaged. You figure you need quarter million to pay for repairs and cover the mortgage until the tourist trade revives. Your insurer is dragging its feet and you didn’t have flood insurance (everyone with shore properties run out right now and get it. Make socialism work for you.) Plus, your house is underwater (figuratively) because home prices are down 30%. Temporarily, at least, nothing is selling.
Question. Do you have a liquidity problem or a solvency problem? Of course, the answer is it depends. We now know that property values in most beach communities have bounced back nicely. But that wouldn’t have helped in the short term. Temporarily, you are like Schrodinger’s cat; you are both solvent and insolvent. If FEMA comes through with a loan before you miss too many mortgage payments, then you discover you had a liquidity problem. If not, your bank will foreclose and then you will discover that you actually had a solvency problem. Of course, in New Jersey it will take decades for the bank to get title to your house, during which time you’ll be using it mortgage free. That might be the best outcome of all.
Extra credit question: Is the FEMA loan a bailout? If you pay it back? If you don’t? (Be clever!)
There is no question that prior to the crisis, perverse regulatory incentives motivated most European Banks and many “shadow banks” to become inexcusably illiquid. After the collapse of Northern Rock, Lehman and AIG, concerns naturally arose about the other large institutions.
Some large commercial banks encountered funding issues, and the Fed met their needs. Recognizing the better part of valor, the surviving large investment banks scuttled into commercial bank charters so they would have unfettered access to Fed borrowing. The Fed supplied generous funding through a variety of mechanisms and ultimately confidence was restored. Some observers were miffed by the Fed’s support of financial companies, but that’s how the system was designed and thanks to Bernanke et al, it worked reasonably well.
As Barney Frank once observed, it is impossible to prove a counterfactual. Still, I believe that absent near-collapse in Europe and the failure or near failure of many shadow banks, no US commercial banks (OK, perhaps ex-Citi) would have encountered any liquidity problems at all. Lots of banks – domestic and foreign – borrowed from the Fed as they were encouraged to do, but in the end among the big banks only Citicorp and Wachovia needed to be recapitalized or absorbed by another bank. (BankAmerica also received government equity, but only after being encouraged to buy Countrywide and strong armed into to absorbing Merrill Lynch.) It’s doubtful that even these two outliers actually had been insolvent.
Section 165, Title 2 of Dodd Frank directs regulators to establish “liquidity requirements” and “short term debt limits” without defining “liquidity.” In typical fashion, regulators have taken this directive and doubled down. To micromanage this effort, regulators have fashioned some of their patented sophisticated formulas – the “Liquidity Cover Ratio” (LCR) and the “Net Stable Funding Ratio” (NSFR). They force large banks into contortions to increase liquidity as the regulators define it and redefine it on a regular basis.
Within limits, trying to improve bank liquidity can be a good thing. Certainly, this should have been an imperative for regulators of European banks. But in their ardor to purge liquidity risk from the financial system, US regulators are missing something absolutely fundamental: PROVIDING LIQUIDITY IS WHAT BANKS DO. When a bank makes a loan, it creates a demand deposit that the borrower can draw on whenever he or she chooses even though the loan itself might not come due for years. So a bank’s basic function creates liquidity risk by definition, and creates money in the bargain. (The most cogent description I have found of this basic banking identity can be found, of all places, in the book Sapiens by Yuval Harari, pages 306 – 309.)
So you can see that if banks are doing their job, the banking system is illiquid, and the rest of the economy —us— have lots of cash. In Econ 101 this is known as “maturity transformation.” Liquidity-wise, the banking system is simply the mirror image of the economy. Thus, compelling banks to become more liquid inevitably drains cash from all of us who are not banks. That is, we can’t get loans. And credit restriction slows the economy.
I have a modest proposal that will be loved by liquidity freaks and equity fetishists alike. If we really hate leverage and fear illiquidity so much, we should simply require banks to hold 100% capital – no borrowing at all. Failures and runs would become a thing of the past. And so, of course, would loans. No bank could ever again make a loan, because a new loan automatically creates a deposit (debt) — verboten under our new regime. If a regulator’s prime objective is to avoid a Congressional Finance Committee grilling, this solution seems optimal.