Misreading Walter Bagehot: What Lombard Street Really Means for Central Banking


Review of Lombard Street: A Description of the Money Market, by Walter Bagehot

London: Henry S. King and Co., 1873

Among central bankers and fellow traveling academics and financiers, a common trope has arisen to celebrate this once obscure nineteenth century editor of The Economist magazine. That trope goes something like this. Walter Bagehot is Victorian prophet of central banking, the author of the “bible of central banking,” Lombard Street: A Description of the Money Market. In Lombard Street, Bagehot became the first to articulate what a central bank should do to prevent a panic from becoming a crisis. It should simply follow Bagehot’s dictum: “lend freely, at a penalty rate, against good collateral.” This dictum—sometimes referred to as a Rule, sometimes as a Law—is the gospel for central bankers, as applicable to today’s markets as they were at the time he wrote it 140 years ago. Indeed, to continue the religious imagery, Bagehot’s own name has become something of a shibboleth in central banking circles. Pronounce it “Bag-ih-hot” and you will expose yourself as a central banking outsider.

Except for the fact that “Bagehot” isn’t pronounced as written—it’s “BADGE-it,” in American English, “BADGE-ot” in the Queen’s English—there’s little about the above paragraph that is true. The rest is part of a mythology that has arisen around Bagehot. Lombard Street has become a classic in the sense of the word often attributed to Mark Twain: a book everyone cites but no one reads.

That doesn’t mean that Bagehot has nothing to offer scholars and practitioners of (central) banking today. To the contrary, Bagehot was a wordsmith nonpareil. The book is immensely readable and contains pithy and wise observations on nearly every page that provoke thought about our current set of institutional financial arrangements currently in place. No financial commentator has ever matched him, in my view, in the 140 years since his death. He’s better than Keynes, Galbraith, or Friedman, and rewards the patient reader for those quotable insights alone.

Bagehot is much more than a wordsmith. To the extent the book can be a guide to modern central bankers, it is not in the banks’ capacity as lenders of last resort. It is in the need to speak clearly, plainly, and transparently about what the central bank’s actual functions are. Doing the right thing and then denying that it was done was the cardinal sin in Bagehot’s version of the “money market.” Lombard Street is therefore not a broadside against a central bank that fails to use its authority; it is a diatribe against a central bank that failed to acknowledge, publicly and transparently, that it had done so.

This essay will first discuss what Lombard Street is not, second what Lombard Street is. I invert the usual order because the mythologies that have arisen around Bagehot are so pervasive and so flawed. It will then conclude with a reflection on whether the U.S. Federal Reserve System, specifically, is measuring up to Bagehot’s principles, and whether anyone should care.

What Lombard Street Is Not

There are five problems with the Bagehot dictum/rule/law trope as usually stated. First, the general articulation of a “central bank”—to use a common if misleading prolepsis—as a lender of last resort hardly originated with Bagehot. Although Lombard Street doesn’t acknowledge the debt, Bagehot’s conception that there should be a readily acknowledged banking backstop in the event of crisis comes not from his pen, but from Henry Thornton’s. Thornton, a lawyer, banker, and economist, fleshed out the idea in his 1802 treatise, An Enquiry into the Nature and Effects of the Paper Credit of Great Britain. Thornton just didn’t write as well, and perhaps for this reason missed out on the credit. In any case, it is surely central banking’s most prominent example of Stigler’s law of eponymy.

So Bagehot wasn’t the original thinker some have supposed. Here’s the second problem. Bagehot is also not the author of this dictum/rule/law, usually attributed to him, whether first or ever. A researcher who flips through Lombard Street for that citation will search in vain (this is why those who do cite the “rule” cite other researchers who purport to cite Bagehot).

Instead, Bagehot cited the three elements of the dictum—that a central bank should, in a crisis, lend freely, against good collateral, at a penalty rate—over the course of the book. The problem with reducing this long discussion to a single apothegm is that Bagehot placed such disproportionate emphasis on the first prong: a central bank should, in a crisis, lend freely. The other two prongs are given short shrift: they appear only briefly and in passing as two provisos to his rather sweeping explanation of “free lending.” For example, here is one of Lombard Street’s most famous passages, worth quoting at length to get a sense of Bagehot’s argument:

A panic, in a word, is a species of neuralgia, and according to the rules of science you must not starve it. The holders of the cash reserve must be ready not only to keep it for their own liabilities, but to advance it most freely for the liabilities of others. They must lend to merchants, to minor bankers, to ‘this man and that man,’ whenever the security is good. In wild periods of alarm, one failure makes many, and the best way to prevent the derivative failures is to arrest the primary failure which causes them.


The way in which the panic of 1825 was stopped by advancing money has been described in so broad and graphic a way that the passage has become classical. ‘We lent it,’ said Mr. Harman, on behalf of the Bank of England, ‘by every possible means and in modes we had never adopted before; we took in stock on security, we purchased Exchequer bills, we made advances on Exchequer bills, we not only discounted outright, but we made advances on the deposit of bills of exchange to an immense amount, in short, by every possible means consistent with the safety of the Bank, and we were not on some occasions over-nice. Seeing the dreadful state in which the public were, we rendered every assistance in our power.’ After a day or two of this treatment, the entire panic subsided, and the ‘City’ was quite calm.

In other words, Bagehot is congratulating the Bank of England for its past behavior and uses that behavior as the example of the Bank doing what it ought to do.

Lending freely, “to this man and that man,” then, is the entire point. That kind of liquid generosity is aimed at flooding the market with the reassurance that there is no scarcity, that there is a financial safety net. Only in this way can the financial “neuralgia” be arrested and calm restored.

This brings up the third problem with Bagehot’s rule/law/dictum: the idea that Bagehot has much to say about “good collateral” that is easily translatable to our own time. There are two problems here. First, as the quote about Mr. Harman’s masterful central banking resolution indicates, Bagehot approved of central banking flexibility when drawing the inherently subjective lines about collateral during a crisis. Bagehot liked central bankers who lend, in a crisis, “by every possible means and in modes” they perhaps have “never adopted before.” So it is that we should take a dim view of those who use Bagehot to justify, for example, the Fed’s decision to allow Lehman Brothers to fail on the basis of non-conforming collateral. Maybe that was the correct course of action, maybe it wasn’t. But Bagehot doesn’t provide the answer.

The second problem with a collateral-focused interpretation of Bagehot is that “collateral” in his time meant something very different from modern usage. As Carolyn Sissoko argues, essentially all the collateral presented for discount at the Bank of England was “good” in Bagehot’s day, because all collateral was subject to a commercial bills doctrine that required self-financing bills. These bills of exchange are those that represent transactions that are already in the past. Say, for example, I sell coffee in Philadelphia that I import from Colombia. There’s a boat in Bogota loaded with my coffee beans, and all my seller needs is the money to ship. What Bagehot had in mind for all collateral was this kind of bill of exchange, between my bank and my Colombian counter-party’s bank, that pointed to that waiting transaction. There could be no “collateral” for me to finance an expedition on to Rio de Janeiro for coffee whose existence was speculative.

This historical difference is why Bagehot devotes essentially no time in Lombard Street to the question of collateral: the doctrines undergirding international finance he had in mind was essentially taken for granted. Again, from Bagehot: “[n]o advances indeed need be made by which the Bank will ultimately lose. The amount of bad business in commercial countries is an infinitesimally small fraction of the whole business.”

We don’t practice that kind of central banking anymore. First of all, central banks discount all kinds of bills, based on all kinds of collateral. My favorite example from the 2008 financial crisis is that the Fed accepted the primary loan secured by the Crossroads Mall, in Oklahoma City, based on speculative projections of the mall’s future revenues. When the mall stopped paying, the Fed foreclosed and became the mall’s owner. In the 1990s, I did my part to keep the mall solvent by buying a lot of Orange Juliuses, not realizing at the time that I was in the shadow of the Federal Reserve System. The point is that while there are some statutes and regulations governing what qualifies as collateral, those standards are a far cry from the self-financing paper that Bagehot saw as the beginning and end of central bank practices.

That takes us to the fourth problem with the law/rule/dictum, the idea of a “penalty rate.” Bagehot never used the term “penalty rate” at all, and what exactly he meant by the “very high rate” that he does propose isn’t exactly clear. At one point, he argues that interest rates should be very high as “the best remedy for the worst malady of the money market,” because the alternative was no money at all. That alternative was anathema to Bagehot: “Any notion that money . . . may not be had at any price, only raises alarm to panic and enhances panic to madness.”

Many critics of the Bernanke Fed point to the fact that the Fed failed to follow Bagehot’s rule because it lent money so freely, at no obvious penalty rate. But Bagehot’s argument was not necessarily relevant to that decision. He wanted to argue against those banks that would not lend at all. High rates suggested that the markets were functioning according to basic principles of supply and demand—high rates for scarce money, not zero rates for zero money.

Elsewhere, Bagehot suggests that the “high rates” are meant to “operate as a heavy fine on unreasonable timidity.” Some say this is to penalize banks who have to come to the discount window for failure to maintain sufficient liquidity to stem the crisis on their own. Bignon, Flandreau, and Ugolini have a fascinating alternative explanation for this. They suggest that the “high rates” should be understood not as a fine on bankers’ failure to manage their liquidity—the stigma theory of Bagehotian high rates—but to penalize them for their reluctance to trust each other. As Bignon et al. put it, the high rate is “an encouragement to make use of the information they have on one another rather than seek the safety of the Bank, for this may result in a complete collapse of interbank lending and destruction of information.” Central banks need information during panics; the high rates, when themselves not destabilizing, can encourage the production of more information.

If these critiques of the received wisdom about Lombard Street nibble at the edges, there’s a fifth and final argument that should render Bagehot worship a dead letter. It’s not at all obvious what Bagehot can tell us about modern central banking. Most of Lombard Street concerns itself with the Bank of England in times of financial panic. Outside of that context, Bagehot has very little to say of resonance to modern central bankers. Inflation, deflation, negotiating the Phillips curve (does it still exist?), liquidity traps (did they ever exist?), and the rest: all of these concepts came after Bagehot. He can hardly be faulted for focusing on the issues of his day, but he can also not be credited for midwifing the issues of ours.

Putting Bagehot in the policy chair 140 years after his death is problematic, too, because of the radically different nature of central banking in the late 19th century. Lombard Street is about the Bank of England as the keeper of the nation’s gold reserve, and what maintaining that reserve—what Bagehot called and revered as the “caste-iron system” because of the ways it removed discretion from central bankers’ policy portfolios—should look like. He published his book at the very beginning of the “classic gold standard,” a period of stable prices and international cooperation in trade and finance that many still cite in their enthusiasm for abolishing central banks (and fractional reserve banking, and deposit insurance, and much else of governmental participation in finance). And, of course, central banks themselves were completely different institutions: they were private, or quasi-private, in a way that they have not been for decades. Charles Goodhart calls this transition central banking’s core “evolution.” Bagehot would have been appalled by the change.

Those libertarian enthusiasms for free money and private central banking aside, there is today no global gold standard. Central banks are largely public institutions. It is fiat currency all the way down. This reality was anathema to Bagehot, though not unfamiliar to him. “In America,” Bagehot writes with a detectable sneer, “it is quite enough for a banker to hold greenbacks,’ though the value of these changes as the Government chooses to enlarge or contract the issue.” In England, there was no such flimsy system. It was within that London-centric system that Bagehot wrote.

Today, though, fiat currencies managed by central bank discretion is the whole ball game. Translating Bagehot to the present, as so many seem to want to do, must confront this untranslatable reality.

What Lombard Street Did Argue: Plain Speaking

Bagehot is thus not the 19th century founder of central banking that those who praise him in the 21st century would have him be. But he did have a message for his day with important implications for our own. That issue is central bank transparency.

Bagehot makes two arguments about transparency, one stylistic point and another substantive one.

Stylistically, he has this to say, in the very first lines of the book:

I venture to call this Essay ‘Lombard Street,’ and not the ‘Money Market,’ or any such phrase, because I wish to deal, and to show that I mean to deal, with concrete realities. A notion prevails that the Money Market is something so impalpable that it can only be spoken of in very abstract words, and that therefore books on it must always be exceedingly difficult. But I maintain that the Money Market is as concrete and real as anything else; that it can be described in as plain words; that it is the writer’s fault if what he says is not clear.

The emphasis is mine. Part of the mystique of central banking is that, stylistically, there is such a hedge built around the technical details of central banking that it has become a playground for specialists only. Jargon abounds, sometimes in ways that facilitate communication among specialists but always with the consequence that many non-specialists—including all of the general public—will be blocked from participation.

Bagehot’s stylistic argument, then, is simple and damning: Complex ideas can be expressed simply and elegantly. It’s the fault of the writer if the final result is full of ambiguity, misdirection, unnecessarily technical language, or simply very boring sentences. Here is a lesson for central bankers that they are too often too willing to ignore. So much writing on this vital subject violates Bagehot’s real dictum. We are all the worse for it.

Bagehot also has something substantive to say to central bankers as applicable today as it was in his time. Lombard Street was not his effort to argue what the Bank of England should do during liquidity crises, as almost all people assume; it was an argument about what the Bank of England should openly acknowledge that it had already done.

Recall that the example he gave in the first extended quote above, the example of a Bank doing exactly what he wants it to do, is from the Bank of England in 1825, roughly fifty years before Lombard Street was published. His beef with the Bank was its directors’ refusal to acknowledge their role in doing what he and they agreed was the correct policy. He puts it this way:

The practice of the Bank [as a lender of last resort] has, as we all know, been much and greatly improved. They do not now manage like the other Banks in Lombard Street. They keep an altogether different kind and quantity of reserve; but though the practice is mended the theory is not. There has never been a distinct resolution passed by the Directors of the Bank of England, and communicated by them to the public, stating even in the most general manner, how much reserve they mean to keep or how much they do not mean, or by what principle in this important matter they will be guided.

It is this failure of the Bank of England—the failure to acknowledge that they kept the central reserve, that they were the lender of last resort—that set Bagehot off. The rest of the book is an argument about why this acknowledgment is so important.

That’s Bagehot’s real rule. There is little gained and much lost when we, in a constitutional republic, delegate substantial authority to central bankers who then obfuscate about what they do with that authority.

How are central banks doing in this Bagehotian endeavor? I’d give the Fed under Bernanke and Yellen a B-. They are a world away from Montagu Norman, the Governor of the Bank of England during the 1920s and 1930s, who infamously announced a rule of his own on central banking: “never explain, never apologize.” Alan Greenspan adopted a similar view: “Since becoming a central banker, I have learned to mumble with great incoherence. If I seem unduly clear to you, you must have misunderstood what I said.”

Given that tradition of mumbled incoherence, it should come as no surprise just how many institutions of Fed transparency are new. Until 1994, the Fed never even announced its decisions. In the same year, the Fed started releasing its minutes and (on a five-year lag) transcripts. And it wasn’t until 2011 that the most important economic regulator in the United States starting holding press conferences.

So yes, we’ve come a long way. But there’s a long way to go. Consider an example. The Fed’s decision to seal the so-called “Doomsday Book”—a collection of legal opinions that outlines the Fed’s emergency lending authorities—is a troubling one worthy of Bagehotian fury. The public knew that such a book existed at least since former Treasury Secretary Tim Geithner’s memoir was published. But now that we know it exists, we also know that we can’t examine the thing. Even the massive and embarrassing disclosures about the Fed’s internal thinking that came to us through the AIG litigation weren’t enough to dislodge it (one lawyer called the Fed’s original high terms for the AIG bailout “loan sharky,” to give but one example). The very authority that the Fed believes it has to respond in a crisis is wrapped in court-sanctioned secrecy. In other words, the central bank may well be doing the right thing, but it refuses to acknowledge the basis for reaching that conclusion.

To those central bankers who continue to insist on the Doomsday Book’s secrecy, might I recommend a classic if misunderstood book?

*  *  *

Elsewhere in Lombard Street, Bagehot laments that the Bank of England has become such a point of “fierce controversy” and “so much animosity” that “a single sentence respecting it is far more interesting to many than a whole book on any other part of the subject.” How ironic that this nuanced and complex book has been reduced to a single sentence that Bagehot himself didn’t write. Here, then, is a proposal to give Bagehot his due, but not more. If scholars and central bankers continue to misread Lombard Street to offer a comprehensive theory of central banking that is at best only partially correct and partially relevant, while ignoring the book’s main lessons, I propose Conti-Brown’s Rule/Dictum/Law be invoked against them. Conti-Brown’s Rule/Dictum/Law has two prongs. First, it requires anyone, central banking critics and defenders alike, who cites Bagehot to point to the exact, full passage of Lombard Street that supports the speaker’s argument. Secondary citations are strictly prohibited. Second, the speaker must identify why Bagehot, given his very different focus and very different historical context, has anything useful to say in support of the speaker’s contentions.

I hope Conti-Brown’s Rule/Dictum/Law will lead people to pause before bringing the great Bagehot’s name into conversations where he doesn’t belong. He was no great seer into the intricacies of 21st century central banking. But his tartly-worded invectives against central bankers who like to keep secrets remains as relevant today as ever. Central bankers and their critics should give Lombard Street a read, in most cases for the first time.

Posted on 14 December 2015

PETER CONTI-BROWN is an assistant professor at the Wharton School of the University of Pennsylvania. He is the author of The Power and Independence of the Federal Reserve (Princeton University Press, 2016). Some of these ideas were first presented on the blog at the Yale Journal on Regulation.