Robert Shiller on Bitcoin and Bimetallism
When Bitcoin’s price spiked in 2017, so did public mentions of the cryptocurrency, as measured by a Google n-gram plot. When William Jennings Bryan promoted bimetallism during his run for President in 1896, mentions of bimetallism similarly spiked. Noting this parallel, economist Robert Shiller compares the two monetary systems more generally in chapter 12 of his 2019 book Narrative Economics, “The Gold Standard versus Bimetallism.”[1] Shiller was the 2013 Economics Nobel Prize co-recipient. He recently appeared to discuss the book on Russ Roberts’ EconTalk podcast.
Shiller observes (p. 161) that “the enthusiasm for bimetallism in the nineteenth century seems similar to the excitement for Bitcoin we have seen in recent years.” Later he goes a bit further to propose that “narratives about gold and money have a peculiar emotional tone, analogous to the emotions we see in cryptocurrency narratives today.”
“Analogous to”? I would say antithetical to. Shiller overlooks a critical contrast between bimetallists of Bryan’s sort and Bitcoin advocates: bimetallists wanted inflation, whereas Bitcoin maximalists support its deflationary tendency. They praise the hardness of the gold standard as a precursor to Bitcoin. Bryan embraced the softness of bimetallism, and famously cursed the gold standard. Shiller (p. 158) rightly observes that Bryan’s plan—resuming free coinage of silver at 16:1 when the market ratio was 30:1—would have been deliberately inflationary: “Thus the bimetallism proposal would have allowed debtors to cut their debts roughly in half by choosing to repay them in silver rather than gold. In effect, the result would have been a default on about half the value of all debts denominated in US dollars.” But Shiller does not mention that Bitcoin supporters embrace its deliberately deflationary design whereby the stock of Bitcoins grows increasingly slowly and eventually stops growing altogether.
And there is a second important contrast: the 1896 peak discussion of bimetallism came in the context of a political campaign. Bryan sought to impose bimetallism on all American money-users, while the majority of voters rejected the policy. Bitcoin, by contrast, is a payment system whose advocates seek voluntary joiners, one at a time. Theirs is not a political winner-rules-all campaign.
At times Shiller seems implicitly to adopt the attitude that the facts shouldn’t get in the way of a good story. He relates popular narratives about bimetallism and the gold standard but does not fact-check them. Most problematically, he comments (p. 157) that “in history the gold standard has long been associated with prolonged deflation and other economic problems.” Doubtless, there is such an association in many minds. But he might have noted that the association is erroneous in several ways:
- The history of the gold standard shows periods of (mild) deflation alternating with periods of (mild) inflation, making the secular record one of near-zero inflation, not one dominated by deflation. Over the entire “classical gold standard” period, 1880-1914, the average inflation rate in the United States was 0.1 percent per year, as Michael Bordo’s readily accessible encyclopedia entry points out. This mean-reversion of the price level was not a lucky accident, but inherent in the economics of commodity money: any downward price-level movement increases the purchasing power of gold and thereby stimulates more gold production (except when due to a permanent increase in the cost of gold mining).
- Most deflations under the gold standard were not troublesome in themselves, nor symptomatic of problems elsewhere. Bordo, Landon-Lane, and Redish (2010, p. 544) find that “deflation in the late nineteenth century was primarily benign,” the result of rapid growth in the output of goods rather than negative monetary shocks. The negative monetary shocks associated with late-nineteenth-century financial panics in the US were not due to the gold standard (hence Canada didn’t have them), but to banking instability stemming from bank-weakening legal restrictions. Nominal prices fell in gold-standard countries mostly when the world’s real income grew faster than the world stock of gold. Relatively rapid real income growth is of course not a problem. Falling nominal prices with flat nominal wages were how workers enjoyed higher levels of real income.
- Deflation (a decline in overall nominal prices) under the classical gold standard was not normally associated with other economic problems like depression (normally defined as a decline in overall real income). Shiller casually identifies “the 1890s” as a depression based on several years’ high unemployment numbers, but a closer look finds that only 3 years (1893, 1894, 1896) saw a decline in real GDP from the previous year. (He also cites the International Bimetallism Conference in London in 1894 for having “noted that a long slow deflation caused by the gold standard had produced depression in agriculture across much of the world,” but surely relative input and output price movements, rather than monetary forces, must be responsible for a decade-long “depression” in one industry only.) US real GDP in 1900 was 32.5 percent higher than in 1890, for a healthy annualized compound growth rate of 2.86 percent. Deflation characterized half the decade, 1893-88, with the GDP deflator of 1898 almost 3 percent below that of 1893 (for an annualized compound inflation rate of negative 0.57 percent over those five years). Thus, only three years of the 1890s exhibited both deflation and declining real income.[2]
Deflation did coincide with depression during 1920-21 and 1929-34. These sharp deflations were not due to the uninterrupted working of the gold standard, which operates gradually, but followed from the high inflations that the Federal Reserve (and European central banks off the gold standard) had created during the First World War, given their refusal to devalue afterward, as persuasively argued by Mazumdar and Wood (2013).
More importantly, the coincidence of depression with deflation was not typical of the classical gold standard. Shiller (p. 158) counsels that “Because the gold standard was talked about very much during those depressions [the 1890s and the Great Depression], we ought to consider how the gold standard narratives relate to the potential for severe depression.” Surely we should also fact-check those narratives. Atkeson and Kehoe (2004) considered the facts of the matter in a panel of 17 nations over 100 years. They found that “nearly 90% of the episodes with deflation did not have depression.” (A measured “episode” was a five-year period in a nation.) They summarized the overall pattern as follows: “A broad historical look finds many more periods of deflation with reasonable growth than with depression, and many more periods of depression with inflation than with deflation. Overall, the data show virtually no link between deflation and depression.” Shiller appears to be innocent of these and similar findings.
Shiller blurs some other details that would undermine his story of a parallelism between the bimetallic and Bitcoin eras. He likens (p. 162) “the Silverites of the 1890s” to “the supporters of Donald J. Trump in the 2016 US presidential election, both in their sympathies and in the contempt that many intellectuals held for them.” But the Silverites were left-wing or progressive populists, more like some supporters of Bernie Sanders, not right-wing or reactionary populists like some supporters of Trump.
Bimetallism sought to re-establish the unlimited coinage of full-bodied silver dollars and fractional denominations. But with the development of the monometallic gold “standard model,” full-bodied silver coins had become obsolete as a way of solving “the big problem of small change,” that is, keeping small currency in circulation at par value. Shiller characterizes the standard model, which “first came about during the eighteenth century in the United Kingdom,” as consisting of “a single gold coin representing legal tender, subsidiary coinage of base metal, and paper money with value based on the government’s unqualified willingness to exchange it for legal tender.” That characterization is nearly correct, except that the value of British paper money in the 18th and 19th centuries—issued by the then-private Bank of England and other private banks—did not rest on any “unqualified willingness” of the UK government to exchange that paper money for legal tender (gold coins). Redeeming notes was the obligation only of the issuing banks. (To be fair, although Parliament did not exchange Bank of England notes for gold coins, it did pass a law to make the notes themselves legal tender during the suspension period of 1797-1819.) The standard model came about from gold monometallism plus the development of trustworthy commercial banknotes that displaced subsidiary coins. In the United States as well, both before and after the Civil War, most currency notes were privately issued and the government did not exchange them for legal tender.
Finally, when Shiller (p. 167) quotes the notorious concluding line from William Jennings Bryan’s speech at the 1896 Democratic Party convention (“we will answer their demand for a gold standard by saying to them: ‘You shall not press down upon the brow of labor this crown of thorns; you shall not crucify mankind upon a cross of gold’”), he fails to consider that its crucifixion trope was anti-Semitic. He comments (p. 168): “Likely the working classes connected their economic suffering with the imagery of Jesus’s suffering a brutal execution at the hands of the powerful Romans.” But as writer Chad Weisman has put it, Bryan was implying “that the religious group responsible for the death of Christ was also the source of the United States’ economic woes.” Bryan and other bimetallists like “Coin” Harvey saw international Jewish bankers, not the Romans, conspiring to impose the gold standard. Bryan had denounced the Rothschild family in earlier speeches. A caricatured Rothschild repeatedly appeared as an object of scorn in the political cartoons of the bimetallist periodical Sound Money. One such cartoon made the subtext of Bryan’s “Cross of Gold” speech explicit by showing Republican Party leader Mark Hanna pressing a crown of thorns down onto a worker, who is nailed to a cross labeled “GOLD,” while Hanna stands on the shoulders of a bearded man labeled “Rothschild, owner of most of the gold in the world.”
In conclusion, while we can applaud Shiller’s attention to the influence of popular economic narratives, we should not take the veracity of those narratives for granted. Some narratives are myths. To study them seriously requires a careful evaluation of their factual basis.
[1] See James Grant’s review of the book here.
[2] All percentage calculations are mine, from the real GDP and GDP deflator data at MeasuringWorth.com.
