Your Central Bank is Insolvent. Now What?

Does central bank insolvency matter?

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of over 11,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

---

I tend to write a lot about how banks work. To summarize things simply, banks do two things. One thing that banks do is serve as intermediaries between savers and borrowers. The asset side of their balance sheet includes things like loans and bonds (which are also loans). The liability side includes things like deposits, which are basically loans to the bank that can be called without notice. That latter characteristic is the second thing that banks do. What makes banks different from other intermediaries, like say insurance companies, is that the bank’s liabilities can be redeemed on demand at any time. In fact, the customer does not even have to come to the bank to do so. The customer can spend these funds with the swipe of card.

This characteristic of bank liabilities means that the bank’s assets also have to include bank reserves, either in terms of vault cash or balances held at the central bank. That way, if a customer shows up and says “I want my money”, the bank can give them cash. If the customer shows up at a store and spends those liabilities, the bank can transfer reserves to the bank used by the seller in that transaction.

One problem banks can have is what is known as a “liquidity” problem. This occurs when the bank does not have sufficient reserves to meet these payment and withdrawal requests and/or when the bank does not have enough assets that it can sell quickly when reserves are low.

A much more serious problem that a bank can have is an “insolvency” problem. This occurs when a bank has more liabilities than it has assets. This is a big problem. When a bank is insolvent, it would not be able to pay all of its liability holders what they are owed even if it sold off all of its assets. That’s not great!

The definition of insolvency makes the issue seem more straightforward than it is. For example, sometimes a bank can be solvent using a certain accounting standard and insolvent using a different accounting standard. Suppose that the bank buys discount bonds today for $99 each. The holder of these bonds receives $100 a year from now. If the bank holds the bond to maturity, it will earn around 1%. Now imagine that the price of these discount bonds falls to $95.

In market value, the bonds have fallen by $4 each. If the bank owns a lot of these bonds, the market value of its assets decline a lot.

In terms of maturity value, the value of the bonds might not have changed. If the bank expects to be able to hold the bonds for one year, they will get their $100 regardless of the current price. But it seems to matter why the bond prices fell. If the reason is that people believe the probability that the issuer won’t pay the bond, maturity value isn’t that useful. However, if it the price fell because investors could get higher risk-adjusted returns elsewhere, then the bank will still expect to be paid back.

Herein lies the accounting issue. If bonds are marked to maturity, nothing has changed about the bank’s balance sheet. If bonds are marked to market, then asset values on the bank balance sheet have fallen, perhaps significantly. In fact, if they have fallen by a sufficient amount, the bank might be insolvent.

Does this type of insolvency matter?

Actually, it might not. If no one is paying attention or if no one does the math or if people expect the bonds to pay out and they don’t need their money right away, the bank can continue to operate. Assuming the bonds pay out, everything works out.

However, if people know about the problem, they might get worried. They might look at the balance sheet and say “if they have to sell this stuff at current market prices, I might lose my money.” If a sufficient number of people worry about this, they all show up and demand their money and the bank must sell its assets at these low prices to meet these withdrawal requests. It becomes a self-fulfilling prophecy. If no one shows up, things work out. If everyone worries and shows up, things most certainly do not work out.

But what happens if the bank that we are thinking about is the central bank? A number of people argue that actually, this is fine. All of this is just accounting.

I’m not sure this is correct.

The balance sheet of a central bank has currency and bank reserves as liabilities and tends to have a lot of government bonds as assets. A number of central banks pay interest on reserves. When the central bank tightens monetary policy, the interest rate paid on reserves rises. At the same time, the market value of the bonds on the asset side of the balance sheet decline due to arbitrage.

There are two issues here. One issue is that the central bank can become insolvent in the sense that the value of its assets are less than its liabilities.

A second issue is that when interest paid on reserves gets high enough, the central bank’s income becomes negative because it is receiving less interest than it is paying out. The excess spending over revenue has to be paid for somehow.

One solution would be to sell these assets that now have much lower prices. This eliminates the hope that the problem can be resolved by waiting until maturity.

Another solution would be to just create the money necessary to cover the budget shortfall. The central bank could do this by paying for its operations out of revenue and then paying for interest on reserves by simply crediting the reserve accounts of banks. Of course, this makes the insolvency problem worse. Then again, if everyone pretends there is not a problem, the central bank holds the bonds to maturity, and interest rates normalize then things all work out.

While it is true that people cannot run on the central bank in the same way that they run on commercial banks, this doesn’t mean that there cannot be a run. With commercial banks, depositors who run on the bank demand their money back — either in cash or by moving it to another bank. They want to avoid the counterparty risk associated with the bank. With a central bank, most don’t have an account. And what would they redeem their deposit for if they did? Would a person really run on a central bank to get cash instead of a promise of cash? It seems unlikely since the counterparty on cash would be the same as the counterparty on the central bank deposit.

A run on the central bank would tend to occur by a rapid increase in the selling of central bank liabilities. People holding cash, who expect central bank insolvency to result in higher rates of inflation, will buy assets that they expect to appreciate during inflation or durable, real assets. They might also deposit their money in accounts denominated in foreign currency. Higher inflation expectations will also cause people to sell government bonds, further exacerbating the insolvency problem and increasing the future borrowing costs of the sovereign. The expectation of inflation becomes self-fulfilling.

It isn’t necessarily clear how likely this sort of thing is, but it seems safe to say that it is more likely for central bank money that has a lot of close substitutes.

One way to prevent this sort of thing is to get the government to “buy an equity stake” in the central bank. This would increase the reserves of the central bank without increasing the liabilities. Since central banks turn over profits to the government anyway, this doesn’t seem like that big of a deal in an accounting sense. But this newsletter is about economics, not accounting.

In terms of the economics, injections of equity by the government might come with conditions on the behavior of the central bank. Politicians might use their intervention as an excuse to extract political favors or re-orient the central bank around political goals. This can mean having the central bank do things that it is not good at or simply incapable of accomplishing. It also might lead to higher inflation if the political goals require expanding the central bank’s balance sheet.

This isn’t just a theoretical issue. The Swedish Riksbank suffered a significant loss in 2022 and expects its equity to be negative in 2023 (its assets will be less than its liabilities). The Riksbank currently has a Q&A about this on their website that discusses potential problems that could arise. Anyway, the Riksbank seems concerned about the state of its balance sheet and now you know why.