“Following the terrorist attacks of
September 11, 2001… the policy followed by the Federal
Reserve resembled Bagehot’s
prescription but for one important
detail: the Fed provided funds at a very
low interest rate.”
In his classic book on central banking, Lombard Street, Walter Bagehot argues a central bank can prevent crises by lending vigorously to banks in trouble. Further, he states these loans should be made at a very high interest rate. Specifically, “there are two rules. First. That these loans should only be made at a very high rate of interest… Second. That at this rate these advances should be made on all good banking securities and as largely as the public ask for them.”1

Following the terrorist attacks of September 11, 2001, disruptions to communications infrastructures resulted in the
temporary breakdown of the interbank market.2 This created a big liquidity problem for many banks. On a normal day, some banks are short of liquidity because their payments coming due exceed their reserves. They may borrow through the interbank market from other banks that have excess liquidity. By lending directly to banks through the discount window, or by lending to the market through open market operations, the Federal Reserve can influence the total amount of liquidity available to banks. Because of the difficulties in communications described above, banks with large liquidity needs were unable to secure funds on the interbank market. The Federal Reserve stepped in and provided a large quantity of funds both to banks directly and also to the market. Hence, the policy followed by the Federal Reserve resembled Bagehot’s prescription but for one important detail: the Fed provided funds at a very low interest rate. This article argues these apparently incompatible differences concerning the cost at which funds should be provided to banks can both be justified and that Bagehot’s basic insight remains relevant today.

To understand Bagehot, it is important to remember that his lending policy was designed for the Bank of England in the 19th century. At that time, the Bank of England operated in a commodity money environment. In such an environment Central Bank money is backed by a commodity—gold in this particular case—and must be redeemed in that commodity at a pre-specified rate. The ability of the Bank of England to provide liquidity was thus limited by its gold reserves. Because there was a risk of running out of reserves, the Bank of England was faced with a difficult problem. On the one hand it wanted to make sure that all banks needing liquidity would have access to sufficient funds, as this was the only way to prevent the panic. On the other hand the Bank needed to protect its reserves from banks that might be overly careful and borrow even though they did not really need to. If too many such banks were able to obtain liquidity, the Bank of England could run out of reserves before all seriously troubled banks were helped.

One way to screen banks that need liquidity most is to set a high rate of interest. Bagehot writes “[a very high interest rate] will operate as a heavy fine on unreasonable timidity, and will prevent the greatest number of applications by persons who don’t require it. The rate should be raised early in the panic, so that the fine may be paid early; that no one may borrow out of idle precaution without paying well for it; that the banking reserve may be protected as far as possible.”3

This policy works because banks that are more likely to need liquidity are willing to pay more for additional funds from the Central Bank than other banks. If the interest rate is sufficiently high, banks facing a low probability of needing funds will prefer to abstain from borrowing and take their chances, leaving more liquidity for banks facing a high probability. This self-selection mechanism is necessary in a commodity money regime because Central Bank reserves are limited.

If the problem is that the Central Bank might run out of reserves, why not make sure that it always holds enough funds to satisfy the needs of all banks? The reason is that such a policy might be too costly in a commodity money environment. Gold held as reserves by a Central Bank has an opportunity cost. It could be used to purchase consumption goods or invested in some productive way. Hence, increasing the Central Bank’s reserves decreases consumption, either directly or indirectly. If severe crises are very rare the cost of additional reserves might exceed the expected benefit of preventing an event that is extremely unlikely to happen.

Today, the United States does not have a commodity money system. Instead, it functions with fiat money. A fiat money system is one in which money has no intrinsic value. The dollar bills we carry in our pockets are just pieces of paper which cannot be exchanged for gold or any other commodity at the Federal Reserve. They are valued in exchanges, not because they represent a claim on some intrinsically valuable commodity, but only because we believe others will accept them later.

In a fiat money system there is virtually no limit to the ability of the Central Bank to increase liquidity. It can just “print” additional pieces of paper. Of course, increasing the money supply too fast for too long would eventually create inflation. However, even massive liquidity injections should have little effect on prices as long as they are quickly reversed. That being the case, a Central Bank does not have to worry about running out of reserves. Consequently, since it can guarantee that all banks will have access to additional reserves, and since the opportunity cost of these reserves is very small, the Central Bank should provide liquidity at a very low cost. This is also desirable because a high interest rate induces a transfer from the banks to the Central Bank which hurts banks and their depositors.

Finally, it is interesting to note that Thornton (1802) does not take exception to the forgoing argument and would probably have concurred. Thornton, who wrote before Bagehot, also recommends lending vigorously in times of panics: “If any one bank fails, a general run on the neighboring ones is apt to take place, which if not checked at the beginning by a pouring into the circulation a large quantity of gold, leads to very extensive mischief.”4 However, he never mentions lending at a high interest rate. In the face of the argument presented in this paper, the reason might be that England was off the gold standard at the time Thornton wrote. Indeed, from 1797 until 1821, the Bank of England was prohibited from paying its notes in gold. However, the Bank was not prohibited from printing notes. Hence, it was functioning in a fiat money environment similar to that of the Federal Reserve today and could provide liquidity by simply issuing new notes. Thus we should not expect Thornton to recommend lending at a high rate.5

Bagehot would probably have congratulated the Federal Reserve for doing a good job after September 11. Indeed, he would have recognized that preventing liquidity crises requires a different policy in a commodity money world than in a fiat money world.


Bagehot, W., 1873, Lombard street, a description of the money market. (Scriber, Armstrong and Co, New York). References from the 1901 edition, Kegan Paul, Trench, Trüber and Co. Ltd. London. [Online at Econlib: Lombard street, a description of the money market.]

McAndrews, J., and S. Potter, 2002, “Liquidity effects of the events of September 11, 2001.” Federal Reserve Bank of New York Economic Policy Review 8, November.

Martin, A., 2002, Reconciling Bagehot with the Fed’s response to September 11. Working paper 02-10, Federal Reserve Bank of Kansas City.

Thornton, H., 1802, An enquiry into the nature and effects of the paper credit of Great Britain. Reprinted in 1939, (Rinehart and Company, Inc., New York).


Page 199. [In the online edition, pars. VII.58-59.—Econlib Ed.]

McAndrews and Potter (2002) note: “The physical disruptions caused by the attacks included outages of telephone switching equipment in Lower Manhattan’s financial district, impaired records processing and communications systems at individual banks, the evacuation of buildings that were the sites for the payments operations of large banks, and the suspended delivery of checks by air couriers.”

Page 199. [In the online edition, par. VII.58.—Econlib Ed.]

Page 180.

I am indebted to Tom Humphrey for pointing this out to me.


*Antoine Martin is an Economist at the Federal Reserve Bank of Kansas City. He can be reached at Antoine.Martin at kc.frb.org.

A more technical exposition of the ideas in this article can be found in Martin (2002) of the bibliography. I thank the editor Russell Roberts and Mike Orlando for useful suggestions. All remaining errors are mine. The views expressed here are those of the author and not necessarily those of the Federal Reserve Bank of Kansas City or the Federal Reserve System.