Alas, a dinghy is what showed up — and a tiny one at that. Of hundreds of thousands of firms the Main Street program was meant to serve, only 120 or so have borrowed just $1.4 billion from it. At that rate, it would take over 70 years for the program to reach its $600 billion capacity. Instead of “levering up” its Congressional funding, Fed has levered it down — way down.
What went wrong? In a word: risk. Business lending is always risky. In bad times, it’s much riskier. Lending during the current crisis, to firms that can’t “secure adequate credit accommodations from other banking institutions,” is, in Tour de France lingo, hors catégorie — “beyond category.” And the longer the crisis lasts, the further beyond it gets.
The government understood that Main Street lending would be risky. What it failed to appreciate is just how allergic the Fed is to losing money. In designing its Main Street facilities, the Fed erred on the side of extreme caution. It chose lending terms that disqualified many businesses outright, while discouraging far more from bothering to apply. It also made commercial lenders keep a 5 percent stake in their Main Street loans. That encouraged them to watch out who they lent to. Alas, it also convinced most potential lenders, including almost all the big banks, to not participate at all. Of some 11,000 banks, credit unions, and S&Ls that might have signed up, only 575 did.
The government should have seen this coming. The Fed’s reluctance to lose money is in its DNA. For much of its history it only lent to banks on “good banking collateral.” And when, during the Great Depression, it first tried its hand at lending to ordinary businesses, it also proceeded very cautiously, with disappointing results. And that was despite being given a Treasury backstop equal to 50 percent of its business lending capacity!
Business lending is risky in part because ordinary businesses don’t usually have any “good banking collateral” on hand. Instead, loans to them are more likely to be secured by accounts receivable — that is, sales made but not yet paid for. But today’s struggling firms aren’t just waiting to get paid: they’re waiting for lockdowns and travel restrictions to end so they can start selling things again.
As a result, the Fed’s Main Street loans are mostly unsecured. Hence the Fed’s caution in granting them. Nor is the Fed’s $75 billion Treasury “backstop” all that reassuring. Were it any less diligent, there’s no telling how much it might lose on every dollar it loaned out. What does seem certain is that, if it loosened its rules enough to lend $600 billion in a hurry, it would almost certainly lose more than $75 billion.
And that’s a risk the Fed will never take. A compromise of long standing lets it use interest it earns to cover its operating expenses, so it never has to ask Congress for a budget. That gives the Fed some independence. The catch is that the Fed has to hand the rest of whatever it earns to the Treasury. If it burned through its Treasury backstop, the Fed would be tampering with that compromise.
What should Congress have done instead? The answer is simple: it should either have made no use of the Fed, or it should have used it as a mere distributor, and not as a source, of funds. The Fed and the Treasury would then have been free to design a Main Street program tailor‐made to achieve what Congress intended, instead of having to accommodate the Fed’s own strict lending rules and Fed officials’ fear of abusing their bargain with Congress.
And the extra expense? The crying shame is that there wouldn’t have been any. The assumed gain from having the Fed “lever up” Congressional funding was a financial mirage all along.
How so? Suppose Congress had appropriated not $75 billion but $600 billion to fund Main Street loans, and that the Treasury sold another $600 billion in securities to raise that amount. The burden on taxpayers would then have been proportional to the interest on those securities.
Now suppose that the Fed funded $600 billion in Main Street loans, with no Treasury support. By doing so, it would increase the quantity of interest‐bearing bank reserves by the same amount. Since the interest paid on those reserves would come out of the Fed’s Treasury remittances, the Treasury would ultimately bear that funding cost. In short, the difference in the government’s funding costs would boil down to that between the interest rate on Treasury securities and the rate paid on bank reserves.
It’s true that the interest rate on bank reserves is just ten basis points—a tenth of one percent. But the rate on one‐month Treasury bills is also ten basis points, while the yield on two year Treasury bonds is 14 basis points. Because the rate paid on reserves is adjustable, even the two year bond rate might turn out to be the better bargain. So, for that matter, might the 27 basis point rate on five year bonds. In short, there’s no such thing as a free lunch, even for Congress. The tragedy is that, in trying to treat itself to one, Congress ended up starving the nation’s businesses instead.
Thousands of those businesses might still benefit from the government’s help; and the Treasury still has over $250 billion in CARES Act funds in its coffers. To really help Main Street, it could assign that money to the Fed, not to “backstop” the Fed’s own credit, but to finance Main Street directly. Taken off the hook, the Fed can adjust its Main Street program’s terms accordingly. It needn’t dispense altogether with standards and due diligence. But it can at least set terms generous enough to allow CARES Act funds to go where they were supposed to go all along. Although even a lifeboat worth $250 billion can only rescue so many firms, it sure beats a tiny dinghy. And that makes it by far the better bargain.