Geology is Bad Monetary Policy

Shut up Goldbugs

Fiat money is a hard topic for a lot of people. Even successful people. Hence the enduring appeal of some sort of commodity-base, currently embraced by crypto people. But there’s a reason we abandoned commodity money, and the best way to understand why fiat currency makes sense is to examine how the classical gold standard actually collapsed, which, contrary to popular opinion, happened well before the First World War as the impossible arithmetic underlying it failed. You cannot tie your money supply to a randomly distributed geological resource and expect it to grow in any reasonable relationship to economic output.

In the popular imagination of that system, it provided automatic discipline in line with Hume’s¹ price-specie flow mechanism. If a country exported more than it imported, gold would flow in, expanding the money supply, raising domestic prices, and eventually making exports less competitive until trade balanced. If you imported too much, gold would flow out, contracting your money supply, lowering prices, and restoring competitiveness. No discretionary policy needed.

In practice, this is not how things worked. That gave rise to what Whale² described as the “banking school” in which central banks followed what Keynes’³ called the “rules of the game.” Rather than a passive acceptance of gold flows, central bankers actively managed interest rates to maintain the system. Gold flowing in? Lower your interest rates to discourage more inflows and maintain equilibrium. Gold flowing out? Raise rates to attract it back. It was a supposedly self-regulating system. And as the economic landscape remained the same and the old boys network running global banking prioritized maintaining the system over anything else it worked. No wonder the court of the Bank of England loved it.⁴

Unfortunately for them, the economic landscape did not remain the same, and a more autonomous monetary system demonstrated just how flawed the underlying logic was.

Between 1897 and 1914, global gold production increased at about 3.5 percent annually⁵⁶. That’s decent, but it was nowhere near fast enough to accommodate the country that had recently⁷ taken the belt as the world’s largest economy: the United States of America. Unlike the staid old world, we didn’t have a central bank run by aristocrats. In fact, since the second bank of the United States was dissolved in a Jacksonian hissy fit, we hadn’t had a central bank at all. Friedman and Schwartz documented that over this 17 year period, American gold holdings jumped from 14 percent of the world’s total to around a quarter. That means that the American gold supply growth was more than seven percent per year, while the rest of the world’s supply grew at less than three percent a year.

Without a central bank, no one was cutting rates enough to repel these inflows. But what about the price-specie flow mechanism? Didn’t it provide a backstop? Friedman and Schwartz show that over that period, American priced did in fact rise faster than those in the rest of the world (Britain, a reasonable enough proxy at the time). But only by between 60 and 100 basis points per year. This was because the American boom was structural. In a story that should not be unfamiliar to observers of the period from 2021-4, the American economy was growing faster than the rest of the industrialized world. And then—as now (or at least before these idiotic tariffs)—the United States was an agricultural and technological powerhouse. And it had grown too large for the gold standard.

The problem with agriculture for a world on the classical gold standard is that it adds a strong element of seasonality, which means that around the last turn of the century, global gold flows temporarily reversed⁸ during the crop moving season. This destabilized the global economy, forcing then Treasury Secretary Leslie Shaw to spend most of his exceptionally long term attempting⁹ to manage it. One tool he created to manage this struggle was “finance bills” which, with their high interest rates, attracted gold to the country during the low money demand season so that during the crop moving season it was already here and did not need to be imported. The Bank of England didn’t love these, as they made gold scarcer for the rest of the world year round.

The picture of the end of the classical gold standard is almost complete. Only three more pieces need to be placed. The first is the technological growth of that period. In Brad DeLong’s excellent 2022 book, Slouching Towards Utopia: An Economic History of the 20th Century, he creates an index of human technology and shows that from 1500 to 1870, it grew on average 0.2 percent per year. For the 140 years after that, it grew at 2.1 percent a year, more than a tenfold increase. In the early years of that period, much of the technological growth was driven by faster communications and electrical power, both of which were reliant on one key technology: copper.

From the 1890s to the start of the First World War, global per-capita copper consumption increased by a factor of ten¹⁰. With rapid global population growth, these numbers are truly astounding. Copper was, in many ways, analogous to the matrix multiplication accelerators that have become key to the current AI boom. And much like those chips, copper production is capital intensive and has long lead times, meaning that its production was sensitive to credit conditions.

While the City of London was very much the center of the financial universe at the time, the American financial system was still potent. It had volved specifically to work around the constraints of limited gold-backed money. Trust companies—lightly regulated¹¹ institutions that faced no reserve requirements until 1906—had grown explosively because they could lend more aggressively than traditional banks. They could do this precisely because they held less gold relative to deposits. This was “innovation” in two senses: it created a larger effective money supply to serve a growing economy, and systemic fragility.

By August 1907, state banks and trust companies held deposits in New York (the center of national finance since the days of Alexander Hamilton¹²) banks nearly equal to those of national banks. The system treated these deposits as equivalent, even though trust company deposits were backed by far less actual gold. Traditional banks, eager for the higher returns that aggressive lending could generate, had tied themselves to trust companies through various channels. You had, in effect, multiple layers of claims on the same limited gold base.

This was not a failure of regulation or a breakdown of ethics. This was the financial system doing what it had to do to provide enough effective money supply for a rapidly growing economy when the actual monetary base—gold—couldn’t keep pace. The problem was that this credit structure was inherently unstable. When confidence faltered and people wanted actual gold (or at least claims backed more directly by gold), the elaborate edifice would collapse.

So in 1906, when the San Francisco earthquake and fire caused damage¹³ of between 1.3 and 1.8 percent of GDP, the gold outflows from British insurers, and the Bank of England responded according to the rules of the game by raising interest rates. As those policies had to be paid anyway, this just caused pain to the British economy, while Shaw’s finance bills prevented the typical summer outflow of gold. As a result, the Bank of England discouraged lending to American banks until Shaw discontinued the instrument. Still, credit tightened and the American economy slowed a little bit. Friedman and Schwartz note that despite slowdown in production and freight loadings, prices did not fall and the rate of commercial failures did not dramatically increase.

For copper producers, however, this was a bit more drastic. You can’t just shut down a mine because prices fall a little bit, which meant that supply remained robust and prices fell sharply. This probably contributed to the fall in the price of shares in United Copper, but after the largest holders of the stock, the Heinze brothers, discovered a large short position, they saw an opportunity for a short squeeze. Even in a declining market, this is not always a bad thing. Porsche’s short squeeze of Volkswagen shares in 2008 is well remembered. But the Heinze brothers failed, and took down the shadow bank they borrowed from.

As the crisis moved through the financial system, an excess demand¹⁴ for currency over bank deposits emerged. This was a problem, because there was a functional limit to how much money exchangeable for gold can be created. Eventually, an ad hoc coalition—a J.P. Morgan led bailout, Regents bank of France desterilizing gold, and a lending scheme by Canadian government—helped stabilize the financial economy and moving demand for currency over bank deposits back in line. This was not the gold standard working. This was the gold standard failing, requiring extraordinary improvisation by private actors and foreign central banks to prevent complete collapse. The system survived only because people were willing to bend and break its rules. To try to prevent that, from happening again, Congress crated the Federal Reserve system.

The strictures of the gold standard had incentivized the financial system to innovate its way to providing enough money for the economy to function, but it had done so by building a towering pyramid of credit on a narrow base of actual gold. The pyramid was necessary given the constraint. The pyramid was also doomed to periodic collapse.

Let me be precise about what killed the gold standard: the complete disconnect between gold supply and economic output. The numbers don’t work. While the velocity of money will, by definition, adapt, if the money supply doesn’t grow in relation to output, bad things happen.

For the most part, the gold supply has grown slower than the economy. For crypto people, who love—but have never experienced—deflation, this seems ideal. If your economy grows at 3 percent per year but your money supply grows at 2 percent, you get 1 percent deflation. That might sound appealing if you’re holding cash, but it’s economically catastrophic. Deflation increases the real burden of every debt contract. It rewards hoarding over investment—why buy capital equipment today when it’ll be cheaper next month? It creates deflationary spirals as consumers postpone purchases, reducing demand, causing more deflation.

The denizens of the classical system understood this problem, which is why they embraced credit creation: banks lending out multiples of their gold reserves, creating money through fractional reserve banking. But this just converted one problem into another. Instead of chronic deflation, you got credit booms that inevitably ended in crashes when people wanted their actual gold back. The choice was between slowly suffocating the economy with deflation or creating an unstable credit structure prone to regular crises.

Moreover, gold discoveries bore no relationship to economic needs. The California Gold Rush and later discoveries in South Africa and Alaska happened to create modest inflation in the late 19th century—but that was pure luck, not systematic policy. When gold discoveries slowed in the 1870s and 1880s, the world experienced deflation despite ongoing economic growth. When they accelerated again after 1897, prices rose despite no particular change in underlying productivity.

This is monetary policy by geological accident. You wouldn’t design a system this way if you were trying to create stable, prosperous economic growth. The fact that it ever worked at all required either slow economic growth (so money supply could roughly keep pace) or increasingly creative financial engineering (which eventually became unstable).

By 1907, the American economy was growing too fast for the gold supply to keep up, even with unprecedented accumulation. The only way to maintain gold convertibility was through elaborate credit structures that substituted bank money for base money. When those structures inevitably cracked, the gold standard offered no solution. You couldn’t print more gold.

Which brings us to cryptocurrency. Its enthusiasts have misunderstood the fundamental lesson: they think algorithmic scarcity solves the problems of fiat currency, when in reality they’ve just recreated the deflationary ideology of the gold standard—minus the excuse of living through its consequences.

Confidence in the system, on the other hand is durable. As J.P. Morgan said in his testimony to the Pujo committee, “character” is the most important thing in finance.

1. Hume, David. 1752. Of the Balance of Trade.

2. Whale, P Barrett. 1937. “The working of the pre-war gold standard.” Economica 4 (13):

18–32.

3. Keynes, John Maynard. 1919. The Economic Consequences of the Peace.

4. And basically destroyed the UK’s economy to try to get back to it as quickly as possible after the First World War. See Ahamed, Liaquat. Lords of finance: 1929, the Great Depression, and the bankers who broke the world. Random House, 2011.

5. Klein Goldewijk, Kees, and Jonathan Fink-Jensen. 2015. “Gold Production.” IISH Data

Collection.

6. Friedman, Milton, and Anna Jacobson Schwartz. 1971. A Monetary History of the United

States, 1867-1960. Vol. 3. Princeton University Press.

7. Crafts, Nicholas. 1998. “Forging ahead and falling behind: the rise and relative decline of

the first industrial nation.” Journal of Economic Perspectives 12 (2): 193–210.

8. Rich, Georg. 1989. “Canadian banks, gold, and the crisis of 1907.” Explorations in Economic

History 26 (2): 135–160.

9. Sprague, Oliver Mitchell Wentworth. 1910. History of crises under the national banking

system. 5624. US Government Printing Office.

10. Rodgers, Mary T, and James E Payne. 2018. “Monetary policy and the copper price bust:

a reassessment of the causes of the 1907 panic.” In Research in Economic History,

34:99–133. Emerald Publishing Limited.

11. Hansen, Bradley A. 2014. “A failure of regulation? Reinterpreting the panic of 1907.” Busi-

ness History Review 88 (3): 545–569.

12. Beckhart, BH. 1922. “Outline of banking history: from the first bank of the United States

through the Panic of 1907.” The Annals of the American Academy of Political and

Social Science 99 (1): 1–16.

13. Odell, Kerry A, and Marc D Weidenmier. 2004. “Real shock, monetary aftershock: The 1906

San Francisco earthquake and the panic of 1907.” The Journal of Economic History 64

(4): 1002–1027.

14. Andrew, A Piatt. 1908. “Hoarding in the Panic of 1907.” The Quarterly Journal of Economics 22 (2):

290–299.