In Crisis, Fed Applied Liquidity Lessons from Depression, Says Historian

May 20, 2014

An interview with Michael Bordo, Board of Governors Professor of Economics and director, the Center for Monetary and Financial History, Rutgers University

Amy Hennessy: Welcome. I'm Amy Hennessy with the public affairs department of the Federal Reserve Bank of Atlanta, and joining me today is Rutgers University Professor of Economics Michael Bordo. This evening you are going to speak at our Public Affairs Forum. You're going to be speaking about the historic perspective of the Fed's role as lender of last resort. I thought to set the stage you could provide some context from English history, Bagehot's dictum, around that issue of the role of the central bank as a lender of last resort.

Michael Bordo: The first thing we need to talk about is financial crises or financial panics—because that's where the lender of last resort comes in—and a financial crisis occurs when the public loses confidence in its bank or in the banking system, and everybody tries at the same time to get their money out of the bank.

So the central bank, or the monetary authority, in a sense, just completely opens up its floodgates of liquidity and gives the banking system the wherewithal that they need to satisfy the demand. In fact, if people understand in advance that the central bank is there to do that job, then the fact that the central bank says, "We're ready, we're waiting, we're going to do it" ...that can often, in many cases—and historically—allay the panic.

Now there was this very distinguished journalist, the editor of the Economist magazine, writing and operating in the 1850s and '60s and '70s. And so [Walter] Bagehot wrote a book called Lombard Street in which he said that the responsibility of the central bank is to be a lender of last resort first; it's called the responsibility doctrine. And what the central bank should do—and this is what's called Bagehot's law—is you should lend freely, and you should lend to any solvent bank that's illiquid and also you should lend at a high interest rate.

Hennessy: Exigent circumstances.

Bordo: ...And then the exigent circumstances—that's another term that was in the Federal Reserve Act—came in later when there was a question of, what about banks that were nonmembers. Then the Federal Reserve, Article 13-3 in 1932, said that if they have an extreme circumstance, they should be able to borrow, too. You borrow because you've got to have it. And that's where the word "last resort" comes in; there's nobody else that you can turn to.

Hennessy: Can you provide some context also for the origin of the stigma that began to be associated with accessing the Fed's discount window?

Bordo: So what happened is when the Fed set up shop in 1913, and World War I came along, and the Fed became an agent of the Treasury and helped fund the Treasury's debt, after the war we had high inflation. So what the Fed did was to do what central banks are supposed to do when there's high inflation, and stop it. So they raised interest rates, they raised the discount rate; this led to a serious—very serious—recession in 1920, '21...a huge drop in output, 15 percent. Remember the recent crisis was 5 percent? Three times as bad.

The Fed was heavily criticized for this; and so after 1919 or 1920, they stopped using the discount window with the discount rate as its main policy tool. Also, what they did was, because of the criticism, they started to keep interest rates low.

So what happened was because they kept the interest rate low, they didn't follow Bagehot, Bagehot said "high"; they kept it low. So it meant that banks were coming to the window in the 1920s, borrowing, and then speculating in the stock market, right? The Wall Street boom. The Fed didn't like this. So the banks were really reluctant to borrow. Then, what happens later to sort of complete the stigma story, is during the Great Depression, when the banks really need to borrow, they're afraid that if they go to the bank and borrow that people will find out who they are and will say, well, the reason that they're coming to the discount window is that there's some problem they've got, and the public says, "I want to get my money out of that bank." So the banks were reluctant to borrow, and this is a big problem because it meant that the Fed really couldn't do its job.

Hennessy: And lessons learned from the Great Depression, and the Fed's role through the Great Depression?

Bordo: So there were three problems that they faced; one was the stigma problem that I already told you about. Second was restricted membership. So in the Federal Reserve Act it said that member banks, who are primarily national banks, could come; most of the banks in the United States were small unit state banks, they didn't have access to the window. So this is the problem: the ones that failed, the thousands of banks that failed, were state banks. Very few national banks failed.

And the third problem was called eligibility. The Fed would lend money based on what was called "eligible collateral," that meant commercial paper, or agricultural paper, or government bonds—T-bills. A lot of banks didn't have that, so they couldn't borrow. So for these three reasons, the discount window didn't function properly.

Hennessy: Moving forward, to the most recent financial crisis of 2007–2008, and our role as lender of last resort through that crisis: did we apply lessons learned?

Bordo: We really did. Chairman Bernanke was a scholar of the Great Depression. He came up with a new approach called the credit approach, the credit channel; so what he did was, (a) at the beginning of the recession, the beginning of the crisis in 2007, he viewed it as a liquidity event, and so he did a Bagehot, they just provided liquidity.

When it became apparent that it was more complicated than that and that the financial system was not functioning because there was this great fear of counterparty risk, that the problem of these mortgage-backed securities that many banks and investment banks held was that people didn't know who was holding sound mortgages and who was holding unsound mortgages. And so, market players—all of them, investment banks, commercial paper market, money market funds—all of them pulled back.

They used Article 13-3 (that I mentioned before) to come up with a number of facilities that could provide money directly to nonbank financial intermediaries. Bagehot's rule was, so as far as he was concerned, anybody in the market could receive, could get access to the window. So Bernanke in a sense took that lesson from Bagehot and applied it through using the Article 13-3, which was really set up in the '30s for lending to trust companies and savings and loans and state banks; he used that and he applied that to the commercial paper market, the money market mutual funds, asset-backed commercial paper, investment banks, broker/dealers. They came up with all these different kinds of instruments, and they were successful in allaying the panic. I had some criticisms of some of the legacies that will come out of it, but they were successful.

Hennessy: Can you explain changes that have come about in reference to Article 13-3?

Bordo: Yes. So after the crisis there was some pushback on what the Fed had done in the financial services commission (I forget the exact title)—and then that came up again in the Dodd-Frank bill. So they revised Article 13-3 quite a bit. The Fed could not provide liquidity—in other words, make loans, collateralize loans to particular entities like Bear Stearns—and that they could only lend to broader-based entities (I think that's the term they used), which would mean like to particular markets, or to the market.

They also changed 13-3 to say that before the Fed would institute any activity under Article 13-3, they would have to clear this with the Treasury. I'm not sure that's such a good idea. The reason is because a lender of last resort, by definition, needs to move fast. Bagehot says do not hesitate; you don't wait. If you think there's a panic coming, you just go in there and you provide the liquidity. You lend on the basis of sound collateral.

The third thing that they did, which I think could be problematic, is that after the Fed does one of these operations, it has to publish the list of people who were assisted, as well as details of the assistance that was required. That could be problematic because it means that banks, nonbanks, wanting to go to the window are going to think, "I don't know if two years is enough time for people to forget." And that may discourage them. That may push you back into the stigma problem that came out of the 1920s.

Hennessy: Thank you so much for providing your insights. Thank you for joining us, I encourage you to visit our website at, and to follow us on Twitter and Facebook.