Are Low Rates A Subsidy to Banks?

Paul Krugman responds to (and disagrees with) Axel Leijonhufvud:

Are Low Rates A Subsidy to Banks?, by Paul Krugman: Mark Thoma sends us to Axel Leijonhfvud, who declares that

The Fed is supplying the banks with reserves at a near-zero rate. Not much results in bank lending to business, but banks can buy Treasuries that pay 3% to 4%.

This hefty subsidy to the banking system is ultimately borne by taxpayers. Neither the subsidy, nor the tax liability has been voted for by Congress.

This is a common view. But it misses the key point, which is maturity: short-term rates are near zero, while those 3-4 percent Treasuries are long-term.

Here’s a stylized picture:

Short rates will (and should) remain low until the economy recovers substantially; thereafter, they’ll rise as we get closer to full employment. Long-term rates are, to a first approximation, the average future expected short-term rate — because investors choose whether to park their funds short-term or buy long bonds based on which they think will yield more over the next 10 years.

So what can we say about a bank that gets short-term deposits or loans and puts the money into long-term Treasuries? Yes, it’s earning more interest now than it’s paying. But it’s also tying up funds in long-term assets; if and when short rates rise, it will either find itself paying more interest than it receives, or have to sell those long-term bonds at a capital loss. There’s no subsidy here.

Now, there is a question about reported earnings: do rosy numbers on bank earnings take into account the likely future losses on those long-term bonds? I suspect not, or at least not sufficiently — which means that reports of the revival of the financial sector are exaggerated, as are bonuses. But that misreporting is the issue — not the alleged subsidy to the banks.