In the 1983 film Return of the Jedi, Admiral Ackbar turns to the officers on the bridge and says what everyone already knew: “It’s a trap!” It had seemed a little too easy to be able to destroy the main threat, permanently and with no risk. Of course that turned out badly for the Alliance; they shouldn’t have been fooled.
Dodd-Frank and other new regulations were supposed to have fixed the banking system, permanently and without risk. But once again that was too good to be true, and it turns out that all that new regulation did was to set another trap, though not intentionally (although the benefits to large firms are at least partly intentional). The solution to effective banking regulation is to understand the role of the “Lender of Last Resort,” and to commit to doing nothing more, no matter what. As Richard Salsman and I argued more than a decade ago, the alternative, “Too Big to Fail,” has proven disastrous.
The Way to Regulate Banks: The Lender of Last Resort
Banks, and many other financial institutions, are brokers, mediating transactions between people who have money — depositors — and people who want to secure loans to do things with the money — borrowers. Brokers generally don’t hold on to the money that is deposited with them; the value of brokerage is connecting that money with an investment. In fact, the banking business was long described as a sleepy-but-safe activity, one that followed the “3-6-3 rule”:
- 3 percent — the interest you pay on deposits
- 6 percent — the rate you charge on loans
- 3 pm — your daily tee time on the golf course, because this business runs itself
Banks package and sell a product called “liquidity.” Liquidity is a measure of how quickly and cheaply an asset can be used to buy something else. Importantly, liquidity is not money, but a measure of the demand to hold cash balances, rather than holding wealth in some other form. Still, cash is liquid. It is easy to agree on a price, and transferring ownership is cheap. Loans are (usually) illiquid. Loans (such as mortgages) are contracts that bind one party to another, requiring payments that are secured by an asset. In the case of a mortgage, for example, the loan is secured by the value of a home, meaning that it is possible to negotiate a much-lower interest rate than on an unsecured personal loan, because the risk to the lender is smaller.
It is possible to buy and sell loans, or stocks, or other equities, but it is much more expensive than paying cash. (This illiquidity was part of the reason that mortgage-backed securities seemed like such a good idea, because in theory at least those were liquid; in fact, it appears that mortgage-backed securities were pretty liquid, and held their value better than is sometimes described). Another form of loan is called a “bond,” which is a promise to make periodic payments for a term of time, and then repay the full amount of the loan, the principal, at the end of that term. Ten-year US Treasury bonds, for example, have a face value and an implied interest rate paid to the buyer of the bond.
As I said earlier, banks are brokers. They take in deposits, and then use those deposits to “buy” loans. The bank might be the originator of a loan, as in the case of many mortgages. Or the bank might literally buy bonds or other securities, financial instruments that generate a higher rate of return than just holding money.
The problem is obvious. There can be a mismatch in liquidity between the bank’s liabilities (depositors put in cash, and they want to be able to take cash out) and assets (loans, bonds, other securities of various kinds). It is easy to imagine situations where a bank will be technically solvent — the total value of all its assets exceeds the value of all its liabilities — but the bank can’t convert enough of those valuable assets into cash fast enough to let everyone pull out their money right now. And when everyone does want their cash, right now, that’s called a bank run.
A bank run is dramatic, and has been used in movies from It’s a Wonderful Life to Mary Poppins. (It can be fun to use these movies in class, as illustrations!) The reason folks hurry to get their money is that there isn’t enough, and if you snooze you lose. The policy problem is that there is enough value, there just isn’t enough cash, today. That’s why the Lender of Last Resort (LOLR) function is so crucial. All that is required is a short-term loan so that there is enough cash today.
The cool thing about the LOLR solution — and note that the LOLR could be a private central clearinghouse, or store of cash that maintains value in liquid form for immediate disbursal—is that if people believe the LOLR will act immediately and effectively, then the LOLR entity never has to act at all. If I know that I can get my money out, today, or for that matter tomorrow or the next day because the bank won’t run out of money — it cannot run out of money — then I don’t try to get my money out in the first place.
Walter Bagehot (Lombard Street, 1873) made the very sensible argument that many financial crises are not problems of insolvency, but only of illiquidity. And illiquidity is only a problem if literally everyone wants to take their money out of the bank at the same time. That problem is that “everyone wants to pull their money out at the same time” is literally the definition of a bank run, where depositors rush to get their cash while there is still some cash left.
Bagehot (pronounced “BADGE-uht”) claimed that the LOLR must be fully committed to do three things, and never to do more than these three things:
1) Lend as much money as necessary directly to troubled (temporarily illiquid) banks; 2) At a penalty rate (far above the market interest rate) 3) But only against good collateral, as offered by a technically solvent bank.
Since there is immediate, unlimited cash available, there will be no bank runs. Since the interest rate is high, loans that are made will be very short-term. And since the bank has sufficient assets to cover its liabilities, there is no problem securing longer-term loans if that is necessary. Loaning to provide liquidity is cheap and effective, but it is not a bailout, because the bank has equity, it just lacks liquidity.
The drawback with relying solely on Bagehot’s LOLR solution is that it does nothing to address “financial contagion,” when problem banks suffer not just from a liquidity shortage but from full-on insolvency. I learned about “contagion” as part of my professor Hyman P. Minsky’s theory of “fragility” in a financial system, so I tend toward his definition of contagion as a cascade of failures, animated by one or more financial institutions failing to make good on its commitments. When these assets become worthless, other banks immediately become technically insolvent also, though they were solvent an hour ago. The failures propagate like falling dominoes, quickly causing massive financial failures.
The reader will likely notice that the US has abandoned the Bagehot rules in favor of trying to limit contagion. Our LOLR, a composite of the Federal Reserve and the Treasury Department, routinely and willfully misuses the discretion afforded central bankers. In their defense, though, the Bagehot criteria are not politically viable, because failing banks that lack good collateral are just as contagious, and maybe more contagious, than banks that have good collateral.
If the job of the LOLR is to prevent contagion — and that is how the regulatory authorities describe their job — then it is logically impossible to hold to Bagehot’s third rule, lending only to banks that are solvent but need liquidity. But that changes everything. Without the constraint of requiring good collateral, the LOLR is an insurer of last resort — a backstop for depositors who have no reason to consider risk when deciding where to place their funds. This problem has been massively exacerbated by the “deposit insurance” guarantees, which have now been extended far beyond the $250,000 limit to be essentially unlimited.
And that’s what happened for the depositors of Silicon Valley Bank, and Signature Bank (and, by the time this appears, possibly more banks). All of the deposits were guaranteed by taxpayers, even though the banks were insolvent, not illiquid. The usual story has been that the deposits were guaranteed by “the government,” but that’s nonsense. Money is being taken from taxpayers and used to support depositors who made a bad bet about where to put their money.
Since our regulatory practice has gone beyond making loans to illiquid-but-solvent banks, to paying back all the deposits of insolvent banks, the result is that there is no reason for depositors to care about whether their bank is taking excessive risks. This is called “moral hazard,” because it encourages the very risk-taking that regulators are later asking taxpayers to pay for.
The problem of moral hazard sounds arcane, but it’s a trap. In the case of Silicon Valley Bank, the risks in the bank’s weren’t even intentional, but revealed an astonishing lack of knowledge of basic financial principles regarding the capital value of bonds in times of inflation. To be fair, the stockholders of the bank itself have been punished by market forces (maybe, unless the Treasury loses its nerve, and succumbs to political pressure from union and state pension funds. Stay tuned!), because their equity is worthless. But the depositors should have been more careful. And they would have been more careful, except that deposits are insured by taxpayers who have no say in rewarding foolish risks. Worse, the fact that deposits of greater than the $250,000 statutory limit are being covered by taxpayers means that we are signaling other depositors that they should not look at their own banks, because taxpayers will cover those deposits, too.
The reason this is infuriating is that we are being told that taxpayers should be willing to double down, to reimburse even-more-careless depositors for their negligent inattention to risk. And I suppose you can see why, given that this dangerous assumption is now baked into expectations about how regulators will behave.
As Obi-Wan said to Luke, also in Return of the Jedi: “What I told you was true, from a certain point of view.” But Luke was mad that he had been lied to, and you should be mad, too.