Financial History to the Rescue: The Harder Money Wins Out

This article is part of a series on bitcoin (and bitcoiners’) arguments about money and particularly financial history. See also:

(1) ‘On Bitcoiners’ Many Troubles’, Joakim Book, NotesOnLiberty (2019-08-13)
(2): ‘Rothbard’s First Impressions on Free Banking in Scotland Were Correct’, Joakim Book,
AIER (2019-08-18)

(4): ‘Bitcoin’s Fixed Money Supply Is a Weakness’, Joakim Book, AIER (2019-08-28)

The great monetary economist and early Nobel Laureate John Hicks used to say that monetary theory “belongs to monetary history, in a way that economic theory does not always belong to economic history.”

Today I’m going to illustrate exactly that with respect to the Bitcoiner’s (mistaken) progressivism in another episode of Financial History to the Rescue.

In the game of monetary competition, the Bitcoin maximalists posit, the “harder” money always wins out. I’ve been uneasy with the statement as it (1) isn’t clear to me what “harder” money (or money’s “hardness”) really means, and (2) probably isn’t historically true. So we end up with something that’s false, or vague – or both! Clearly unsatisfactory. As I pointed out in my overview post to this series, financial and monetary history is almost always more nuanced than what such simple generalizations allow.

Luckily enough, Saifedean Ammous at the Soho Forum debate last week, did inadvertently provide me with a useable definition – and I intend to use it to debunk the idea that money’s history is one of increased hardness. Repeatedly Saif claimed that monetary history, before the advent of central banking, showed us that the harder money always won out: whenever two monetary networks clashed (shells and silver; wampum and gold) the “harder” money won. The obvious implication is that Bitcoin, being the “hardest” money, will similarly win out. Right off the bat, there’s some serious problems here.

First, it’s not altogether clear that such “This time is not different” arguments apply. Yes, economic history teaches us not to discount what seems to be long-standing or universally applicable phenomena – but also to take notice of the institutional setting in which they happen. Outcomes specific to, say, the Classical Gold Standard, rarely generalize into our hyper-modern financial markets with inflation targeting central banks.

Second, over the twentieth century we literally went from the hardest money (gold) to the “softest” money (central bank-created fiat paper money). Sure, you can argue that this was unfair or imposed upon us from above by wars and welfare states, but discounting it as irrelevant strikes me as overly cherry-picking. If the hardest money “lost” before, what makes you think that your new fancy money will win out this time around?

Then Saif returned to the topic of hardness and defined it as a money whose supply is “the hardest to increase.” The hardness of Cowrie shells or Wampum or gold or Whale’s teeth or Rai stones or the other early money that Jevons listed and discussed in 1875, all rely on a difficult, costly and inconvenient process of extraction and/or production. Getting Rai stones from far-away islands, stringing beads together into extended strips of Wampum, or digging up gold from inaccessible patches of the earth were all cumbersome and expensive processes. In Saif’s mind, this contributed to their hardness. Their money stock were simply difficult to expand – in jargon: their money supplies were inelastic.

The early 1600s Dutch Republic struggled with another problem. As the main financial centre of the time, countless hard money (coins) from all over the world were used in Amsterdam. Estimates say over a thousand legally recognized kinds of coins – and presumably even more unrecognized coins. A prime setting for monetary competition: they were all pretty hard (Saif’s definition: difficult and costly to expand) commodity moneys, of various quality, origin, and recognition in trade.

Another feature of 17th century Amsterdam was the international environment of Bills of Exchange (circulating private credit notes). Briefly summarized, merchants across the world traded debts on Amsterdam bankers or traders, and rather than holding and transporting bullion across the world, they transported the debt of the most trustworthy and reliable Dutch financiers. As all such bills required a settlement medium in Amsterdam, trade on thin margins was very sensitive to fluctuations in prices between the commodity moneys in which their bills were denominated – and very sensitive to debasements and re-defined values by various European proto-governments.

In 1609, the City of Amsterdam created the Wisselbank (initially a 100% reserve exchange bank) specifically tasked with standardizing the coinage and to insulate the bill market from currency fluctuations (through providing a ‘neutral’ unit of account for bills settlement). The Bank accepted deposit of whatever coin at the legally recognized rate (unrecognized at metal content) and delivered ”high-quality Dutch trade coins” upon withdrawal. To fund itself, it added a withdrawal fee of 1.5%, but no internal transfer fee, which made holding currency at the Bank very expensive in the short-term, but very cheap in the long-term. Merchants also avoided much of the withdrawal fee by simply trading balances with one another rather than depositing and withdrawing trade coins. In return for this cost-saving, sellers of bank balances would share a portion of the funds saved with the buyer in what’s known as the “Agio”: the price of Bank money in terms of current money outside the Bank’s accounts. This price would fluctuate like any other price on the market and would indicate the stance of liquidity demands.

In a classic example of Alchian’s monetary competition by transaction costs, Dutch merchants and financiers “outsourced” the screening and assaying of unfamiliar coins. They preferred settling their transactions through the (cheaper) medium that was deposits in the Bank.

And it gets worse for the bitcoiner’s story. In 1683, the Bank coupled its deposits with specific receipts for withdrawal; to gain access to coins, one was required both to hold balances and to purchase a receipt issued by the Bank (they also changed the pricing). Roughly speaking, the Bank became a fractional reserved bank (with capped withdrawals) overnight – and contrary to what the hardness argument would imply, the agio on Bank money rose to above par!

Two monetary historians, Stephen Quinn and William Roberds, summarize one of their many writings on the Wisselbank as follows:

“imaginary money on the Bank’s ledgers succeeded because it was more reliable than the real stuff. […] The most liquid asset in the economy was no longer coin, but a sort of ‘virtual banknote’ residing in Bank of Amsterdam accounts.”

Further,

“the evolution of the agio shows that the market valued irredeemable balances as if they were closely tied to backing trade coins” (my emphasis)

The story of the Amsterdam Wisselbank’s monetary experiments and innovations show us that monetary adaption relies on many more dimensions than “hardness.” Sometimes “hard” money is defeated by “soft” money, since the softer money brought other benefits to its users – in this case a cheap and reliable settling medium.

The lesson for bitcoin-vs-fiat-vs-FinTech is pretty clear: hard money doesn’t always “win”; and sometimes “soft” money can better serve the needs of consumers in a free market.

Rosenbloom on the Colonial American Economy

Joshua Rosenbloom is an economic historian worth following if you are interested in American economic history during the colonial era. He has recently published what appears to be an overview article of the topic (probably for a book or an invited symposium) which perfectly summarizes the current state of the research. I believe that this should be widely read by interested parties.  Here are key excerpts for some of the topics he discusses. I provide some comments to enrich his contribution, but these should be understood as complements rather than substitutes to this excellent overview of the American economy during the colonial era.

On Economic Growth 

Mancall and Weiss (…) concluded that likely rates of per capita GDP growth could not have been higher than 0.1 percent per year and were likely closer to zero. In subsequent work, Mancall, Rosenbloom and Weiss (2004) and Rosenbloom and Weiss (2014) have constructed similar estimates for the colonies and states of the Lower South and the Mid-Atlantic regions, respectively. Applying the method of controlled conjectures at a regional level allowed them to incorporate additional, region-specific, evidence about agricultural  productivity and exports, and reinforced the finding that there was little if any growth in GDP per capita during the eighteenth century. Lindert and Williamson (2016b) have also attempted to backcast their estimates of colonial incomes. Their estimates rely in part on the regional estimates of Mancall, Rosenbloom and Weiss, but the independent evidence they present is consistent with the view that economic growth was quite slow during the eighteenth century.

This is still a contentious point (see notably this article by McCusker), but I believe that they are correct. In my own work, using both wages and incomes, I have found similar results for Canada and Leticia Arroyo Abad and Jan Luiten Van Zanden have found something roughly similar for the Latin American economies (Mexico and Peru).

It is also consistent with even simplistic accounts of the neoclassical growth model. The New World was an economy of abundant land input whose outputs (agricultural produce) were mostly meant for local consumption. If one wanted to increase his income, all he had to do was use more inputs at really low costs. There is very little in this situation to invest in increasing total factor productivity and incomes would only increase at the dis-aggregated level (following the same region over time) as we are capturing the extent of inputs included over time (e.g. the long-settled farmer has a high income because he has had the time to build his farm, but the short-settled farmer brings the average down because he is just starting that process).

On Monetary History and Monetary Puzzles

In lieu of specie, the colonists relied heavily on barter for local exchange. In the Chesapeake transactions were often denominated in weights of tobacco. However, tobacco was not used as a medium of exchange. Rather merchants might advance credit to planters for the purchase of imported items, to be repaid at harvest with the specified quantity of tobacco. Elsewhere book credit accounts helped to facilitate transactions and reduce the need for currency. The colonists regularly complained about the shortage of specie, but as Perkins (1988, p. 165) observed, the long run history of prices does not suggest any tendency of prices to fall, as would be expected if the money supply was too small. (…) With only a few exceptions the colonies issuance of these notes did not give rise to inflationary pressures. There is by now a large literature that has analyzed the relationship between note issuance and prices, and finds little evidence of any correlation between the series (Weiss 1970, 1974; Wicker 1985; Smith 1985; Grubb 2016. As Grubb (2016) has argued, this suggests that while the circulation of bills of credit may have facilitated exchange by substituting for book credit or other forms of barter, they did not assume the role of currency.

In this, Rosenbloom summarizes a puzzle which has been the subject of debates since the 1970s (starting with West in 1978 in this Economic Inquiry article). In many instances (like South Carolina and Pennsylvania), the large issues of paper money had no measurable effect on prices.  This is a puzzle given the quantity theory of the price level. The proposition to solve the puzzle is that as the paper money printed by colonies tended to be backed by future assets, they were securities that could circulate as a medium of exchange. If properly backed and redeemed, people would form expectations that these injections were temporary injections and there would be no effect on the price level all else being equal. Inflation would only occur if redemption promises were not held or were believed to be humbug. This proposition has been heavily contested given the limited information we hold for the stock of other media of exchange and trade balances. I have my own take on this debate on which I weigh using a similar Canadian monetary experiment (see here), but this is a serious debate. Basically, it is a historical battleground between the proponents of the fiscal theory of the price level (see notably the classical Sargent and Wallace article) and the proponents of the quantity theory of the price level.  Anyone interested in the wider macroeconomic debate should really focus on these colonial experiments because they really are the perfect testing grounds (which Rosenbloom summarizes efficiently).

On Mercantilism, the Navigation Acts and American Living Standards

The requirement that major colonial exports pass through England on their way to continental markets and that manufactures be imported from England was the equivalent of imposing a tax on this trade. The resulting price wedge reduced the volume of trade and shifted some of the producer and consumer surplus to the providers of shipping and merchant services. A number of cliometric studies have attempted to estimate the magnitude of these effects to determine whether they played a role in encouraging the movement for independence (Harper 1939; Thomas 1968; Ransom 1968; McClelland 1969). The major difference in these studies arises from different approaches to formulating a counterfactual estimate of how large trade would have been in the absence of the Navigation Acts. In general, the estimates suggest that the cost to the colonists was relatively modest, in the range of 1-3 percent of annual income. Moreover, this figure needs to be set against the benefits of membership in the empire, which included the protection the British Navy afforded colonial merchants and military protection from hostile natives and other European powers.

The Navigation Acts were often cited as a burden that the colonists despised, but many economic historians have gone over their impact and they appear to have been minimal. It does not mean that they were insignificant to political events (rent-seeking coalitions tend to include small parties with intense preferences). However, it does imply that the action lies elsewhere if someone wants to explain the root causes of the revolution or that one must consider distributional effects (see notably this article here).

These are the sections that I found the most interesting (as they relate to some of my research agendas), but the entire article provides an effective summary for anyone interested in initiating research on the topic of American economic history during the colonial era. I really recommend reading it even if all that you seek is an overview for general culture.