The possibility that the federal government will soon be unable to finance its normal operations has become very real. As I wrote in my last column, this won’t be because investors view U.S. debt as excessive; America in 2023 isn’t Greece in 2009. If it happens, it will be because Republicans in the House are trying to use the debt ceiling to extort policy concessions they would have no chance of enacting through the normal legislative process.
In such a situation, it’s natural to consider possible end runs around the debt ceiling that the Biden administration could use to meet U.S. commitments without the cooperation of Congress. Indeed, it would be irresponsible not to consider these possibilities. It would be especially irresponsible to reject them because they sound undignified: Crashing the world economy for fear of looking silly would be unforgivable.
And while there may be legal and political obstacles to using clever budget tricks to avoid political extortion, I think it’s important to understand that the economic arguments I’ve been hearing against these tricks, sometimes from people who really should know better, are just wrong — embarrassingly so.
There are two main gimmicks that have been widely discussed: premium bonds and platinum coins. Premium bonds are harder to explain, which may make them a more likely route, simply because the platinum coin offers an easier target for false narratives. But let me start with the coin.
For those who don’t know what I’m talking about, there’s a law specifying which coins the Treasury Department may issue, and it gives the Treasury secretary essentially unlimited discretion in the design and denomination of platinum coins. Obviously, this law wasn’t intended as an answer to debt ceiling extortion; but then, the debt ceiling wasn’t intended to serve as an instrument of extortion, either.
So here’s how it would go: Treasury mints a platinum coin with a value of $1 trillion. (No, it needn’t include $1 trillion worth of platinum.) It deposits this coin with the Federal Reserve, which adds $1 trillion to Treasury’s account at the Fed. The government can then draw on this account to pay its bills without having to issue new debt.
It may sound silly, but as I said, this is no time to be worried about dignity. What I’ve been hearing are two objections that sound substantive: that minting the coin would be inflationary, and that it would amount to the Fed giving the government a zero-interest loan. Both objections are just wrong.
What is true is that as the government drew down its account, the Fed would essentially be creating money out of thin air, which sounds inflationary — and would be, if that were the end of the story. But the Fed would almost certainly “sterilize” the monetary effects of the transaction, selling off some of its immense asset holdings to remove the newly created money from the system. It would have no problem doing this, since the Fed owns $5 trillion in U.S. government securities:
The thing is, if you consider the Fed to be a branch of the federal government — which it is from a fiscal point of view, even if it has considerable policy independence — when the Fed sells off some of its bond portfolio, it’s just as if the Treasury Department were selling debt the usual way. Minting the coin is basically a way to continue normal borrowing via a backdoor route that bypasses the debt ceiling.
Once you understand this, you also realize the falsity of the second claim: that if the Fed were to accept the coin, it would be giving the government a zero-interest loan. No, the Fed wouldn’t charge interest on Treasury withdrawals, but it would sell bonds to sterilize these withdrawals, and in so doing lose the interest it would have earned on those bonds.
But here’s the thing: The money the Fed earns on its portfolio is, by law, remitted to the Treasury. So the interest lost by the Fed would, in the end, be a cost to the Treasury — exactly the same cost the Treasury would have paid in interest if it had sold those bonds itself. So, no, this wouldn’t be a zero-interest loan, not in any meaningful sense.
Bottom line: Under the surface strangeness, minting the coin is just a way to permit de facto borrowing despite the debt limit.
OK, on to premium bonds.
The U.S. government finances itself largely by selling notes and bonds (10 years or less of maturity is a note, more than that a bond). These securities combine a par value — the amount that will be paid when the note or bond matures — with an interest coupon, a sum paid twice a year. Notes and bonds are auctioned off, often for more than their par value, because sometimes market interest rates are lower than the face interest rate — the annual coupon as a percentage of par value — so investors are willing to pay a premium.
Normally this is a small factor, because interest rates on newly issued notes are set close to prevailing market rates. But that doesn’t have to be the case.
So when a $100 10-year note matures, why not issue a new note, also with a par value of $100 — so that officially we aren’t adding to the debt — but with a face interest rate of, say, 10 percent, far above market rates (which are currently 3.37 percent). This new note would sell for much more than its face value, so Treasury would in fact be raising a substantial amount of money, even though it isn’t officially increasing the debt.
And there’s nothing fundamentally wrong with selling debt instruments for more than their par value. Until 2015 part of Britain’s debt consisted of consols, bonds that pay a fixed coupon every year but never mature and therefore have no par value at all.
But, but, you splutter, that’s cheating! Shouldn’t we measure debt by its market value, not an unrealistic par value? Well, that’s not what the law says.
Also, if you want to start using market value as your debt measure, you should be aware that the market value of U.S. debt has actually declined sharply in recent years:
Why? Because the government issued a number of long-term notes and bonds back when interest rates were considerably lower than they are now, and these securities now sell at a discount. So are we supposed to use market values to measure debt when they go up, but not when they go down?
You might ask how we’re supposed to enforce a debt ceiling if the government can play games with the definition of debt. But the answer, of course, is that we shouldn’t have a debt ceiling. The government should make decisions about taxing and spending, and consider the fiscal consequences, without creating an additional choke point that extremists can weaponize.
Again, I realize that all of this can sound strange, and there may be legal or political obstacles to doing end runs around the debt ceiling. But people who say that such end runs would be unsound from an economic point of view just haven’t done their homework.