Lombard Street is the original treatise on money markets written by Walter Bagehot in 1873, a piece that is especially relevant after the Great Financial Crisis. During Bagehot’s time, the now-familiar ideas of fractional reserve banking and cheap liquidity were new to the world. Before strong banking infrastructure was created, banks would always keep an equal amount of money in reserves as they lent out. If everyone wanted to withdraw their deposits at the same time there would be no issue. As the banking infrastructure grew, however, banks began to lend out more money than they held. They realized that they could increase profits by lending out more than they held in deposits, confident that they could meet any withdrawal demands as such demands were unlikely to occur all at once. This created the new and frightening possibility of an alarm. An alarm would lead to a panic if everyone wanted to withdraw their money at once and were prevented from doing so by the banks.
Bagehot is the first authority on how central banks ought to address such a panic, and his theory has received much attention as of late. While the theory has grown dramatically since it was first written, implementation—and prevention—can still be difficult. The U.S. was able to prevent a run on the money markets in 2008 from collapsing the economy, but was not able to prevent the run in the first place. The consensus seems to be, based on a recent Federal Reserve working paper, that we followed the monetarist model as put forward by Milton Friedman. However, this model is remarkably similar to Bagehot’s model, and I will compare the two in the final part of this series.
The upcoming blog posts will be a four-part series on Lombard Street, specifically tying the theories of Bagehot into contemporary issues of panics and central banking.
- Part One: The nature of panics and liquidity crises
- Part Two: What Bagehot suggests we do in a liquidity crisis, and how it applies today.
- Part Three: How money markets and central banks came to exist.
- Part Four: The Friedman ‘Monetarist’ models the Federal Reserve chose in 2008 and how it differs from Bagehot and other economic models.
[Or join the Facebook group, SchoolsAndThought, to read the first three parts early! That’s incentive!]
Part One: The Nature of Panics
Peel’s act was passed in England in 1844. It restricted the creation of currency to the central bank. Prior to 1844, all banks could issue their own legal tender. This system created the Bank of England as England’s central bank. It was the only bank that could issue currency. As a result all subsidiary banks were forced to work with this bank. This is the same system in use today with the Federal Reserve in the U.S. and the European Central Bank in the E.U.
Panics are intuitive in nature, but it is important to understand the working parts in a panic to understand its underlying economic causes. Even though panics all share the similar result of every person trying to withdraw their money at once, the reasons for the panic can vary drastically. These reasons must be considered when countering the panic.
In Lombard Street, Bagehot distinguishes between domestic and external (foreign) causes of a panic. Domestic causes for panic arise from credit disturbances within a country. These are the issues that spiral as those who panic withdraw their money. A panic is at its core a simple phenomenon. Banks start worrying and rumors spread about the credit worthiness of a bank. Foreign causes are when the country’s reserves are slowly depleted due to ‘unfavorable’ exchange rates. Fixing this issue is generally easy if given sufficient time. A country needs only to raise its interest rates to attract capital. Since all nations are now on fiat currency and not on fixed or metallic currencies this is not as necessary for current monetary crises. In the past if all the gold reserves dissipated the entire system, central bank included, would be (potentially) insolvent.
The primary entities that withdraw their money from banks are banking subsidiaries and merchants, or in contemporary languages, corporations and firms. Banking subsidiaries generally specialize in offering better rates to firms by devoting more time to analyzing their risk. For example, while a massive firm might offer a flat rate to firms that fall within a statistical confine, a smaller subsidiary might research deeper into their books to potentially offer them a better deal. The subsidiary can then borrow from a large dealer and lend it out at a higher interest rate to another firm. These firms are not primary dealers. They use their extra time and market intelligence to make money, not their massive capital and infrastructure. In a time of panic they go to principal dealers to get cash forwards so they can honor their lending promises to other firms. Larger dealers must lend them the money or people will suspect they don’t have the money, leading to a crisis of confidence and a potential run on the bank.
Primary dealers must, in fact, lend to everyone. This creates a brutal spiral that defines a panic. When a panic occurs all banks want to amass a reserve. But since everyone else is afraid the banks will run out of money they all attempt to withdraw their money. If the bank refuses it will collapse instantly since they will lose all reputation. However, if they lend it out they will quickly run out of money. This is the point where a central bank must intervene. It is important to address the primary failure. There is a difference between a failure due to fundamental failure and pure counter-party risk.
Drawing the line between the two is far nicer in theory than in reality. And even if you can in reality it is still not a simple separation. Due to the nature of banking all primary dealers can be thought of as prisoners on the deck of a ship attached by the ankle with chains. The bank that starts the downward spiral is the first prisoner to fall off the ship, bringing all of his fellow prisoners down with him. If a central bank can quickly and efficiently cut off both of the first prisoner’s legs the system will stabilize. The assumption here is that one bank caused the crash and the rest are sound. In 2008 it is more likely that all banks were rotten, but some were more rotten than others, so they simply fell off the ship first.