The U.C. Berkeley Faculty...
Twitterstorm delong: November 29, 2011

Department of "Huh?!": Unclear on What Central Banks Are Department

I see that Benn Steil, a student of geoeconomics at the Maurice R. Greenberg Center for Geoeconomic Studies at the Council on Foreign Relations, is trying to scare his listeners into thinking that something very bad might happen because (i) the ECB is undercapitalized and (ii) the ECB is busy buying up the bonds of europeriphery sovereigns:

The Future of the Eurozone: The ECB only has 81 billion euros in capital. That could easily be wiped out with, say, just a 25 percent haircut in PIG debt -- Portugal, Ireland and Greece. That's it. Gone.

Now, a central bank can operate for a brief period without any capital. But eventually any central bank will have to tighten monetary policy at some point in the future. And in order to do that, they need assets to sell. And unless the market believes that the ECB is going to be credibly recapitalized, there will be a god-almighty run against the euro and it will, quite frankly, collapse.

That's not the same situation as the Feds in the United States. In fact, if you look at the Fed's reported capital, it's only $58 billion…. But nobody doubts that the U.S. government is ultimately going to stand behind the Fed. But investors do doubt that the German taxpayer is ultimately going to stand behind the ECB. And that's a big, big problem…


It is not.

A central bank--like the ECB--is a very, very different animal than a normal bank.

A normal bank needs capital to prevent runs precisely because its depositors can force it to shrink its balance sheet whenever they wish. Depositors can come to it and demand their money. It then needs to sell some of its assets in order to raise the cash to pay its depositors. And if assets are less than liabilities--if there is no capital--then depositors can see the endgame, and so they all rush to pull their money out because the last ones in the queue don't get anything. There is a run on the bank. The bank goes bankrupt. That is why a normal, commercial (or shadow) bank must be well-capitalized.

But, as economists have known for centuries, a central bank is different.

If you go to a central bank and say "I want to pull my money out", the central bank says "OK". It then prints up a number of banknotes. It then hands the banknotes to you. It doesn't need to sell assets to give you your money. All it needs to do is to transforms one of its liabilities--a reserve deposit--into another one of its liabilities--a bank note.

There is nothing anybody can do to force a central bank to shrink its balance sheet.

A central bank shrinks its balance sheet only when it wishes.

At the moment, if the ECB wanted (and if it could transact at accounting values), the ECB could shrink its liabilities--the monetary base of the eurozone--to zero and still have 81 billion euros of assets left over. (Of course, the ECB would never want to do that.)

Suppose that the ECB were to take a big haircut of 200 billion euros on its PIIGS asset holdings. What would that mean? Well, it would mean that the ECB could only shrink the eurozone monetary base down to 119 billion euros before it would have to stop. But there is no conceivable situation in which the ECB would ever want to shrink the eurozone monetary base that much. So taking losses on its PIIGS bonds is not, as Steil claims, a very worrisome thing. Taking losses on its PIIGS bonds is a nothingburger.

So why does Steil think it is a "big problem"?

I have no clue.

He says there will be a "god-almighty run against the euro and it will, quite frankly, collapse" but I have no idea what he could possibly mean...

I think it was in April 2009 that Christina Romer told me that the most astonishing thing about the situation was the number of people confidently pronouncing on issues they could not or had not thought through, and that she wished they would just sit down and be quiet, and then we would all be much better off...