Between the 1870s and 1913 almost every developed country and many developing countries were on a gold standard. This meant that in some form or the other their currency was convertible to gold. But the word “gold standard” hides two important facts - there was a large amount of diversity in the gold standard and that some countries could break the rules of the gold standard (for some time) and get away with it.
The Myth of the “Gold Standard”
There was never a single gold standard.
They either had actual gold coins, notes backed by reserves in the central bank, notes backed by other currencies which were backed by gold (mostly for colonies like India and Philippines) or some combination of the last two.
Even if two countries both had the same type of gold standard (say notes backed with gold in the central bank), they too would have different rules for those.
Some like Britain and Russia, allowed a limited amount of money to be issued but not backed by gold, but after that amount of money, every extra pound or ruble had to be issued only if there was a certain amount of gold in the central bank. For Britain this was one troy ounce of gold per £4.25.
But in other countries like the Netherlands or Belgium, gold had to be a certain percentage of the currency (typically 35 to 40 percent of money). Along with gold, countries also held foreign exchange. For colonies like India and Philippines this was imposed on them by the nation that ruled them. But many countries (like Japan) also held foreign exchange in the form of bank deposits because it paid interest, unlike gold which paid none. At the end of the classical gold standard in 1913, this was only around 20 percent of gold reserves.
The idea that there was a single gold standard was a myth because nations had multiple variations on it.
The Rules of the Game
Let's imagine a country set up its currency which was on a gold standard. In this example, the country backs its currency 1:1 with a specific amount of gold, but this model can be generalized to other variations of the gold standard.
If one country ran a trade deficit (for example Britain) against another (France), it would lose gold. This would happen because it spent more money on goods from abroad, than money it received from abroad. French merchants would then have more British currency - pound sterling - than they bought British goods with. French merchants couldn’t use British pounds in France, so they would get those pounds exchanged for gold at the Bank of England. They would then convert this gold to francs at the Bank of France
The BOE would have less gold than before, and the Bank of France would have more gold than before. If the Bank of England backed its currency with gold, an outflow of gold would lead to a reduction in the money supply. And in France, an inflow of gold would lead to an increase in the money supply. This would lead to prices in Britain decreasing, and prices in France increasing.
British exports would become cheaper as prices in Britain were lower than before and French exports would become more expensive. So, if this was iterated multiple times, the trade balance would self-correct as relative prices in these countries changed.
This model is approximately true. The only thing it doesn’t describe well is that there was very little actual gold movements when countries had trade surpluses or deficits. Foreign merchants never actually went to a central bank and exchanged their currency for gold. When central banks anticipated an outflow of gold, they would increase interest rates. This would reduce credit creation (and so the “broad” money supply would decrease), and economic activity in the country would also decrease.This would attract more gold from abroad, stop the outflow of gold but also depress economic activity at home.
Or in the exceptional case where a country didn’t have a central bank (like the USA before 1913), this process would happen automatically with actual gold flows in and out of the country.
This model implies a few rules of the game. The first is that money has to be backed by gold in the central bank. This can be in full (like Britain) or partially (like Belgium). The second is that changes in the money supply have to be caused by changes in gold flows. If gold flows in the country, then the central bank has to issue more money as it has more gold and vice versa.
For a country with no central bank, these rules were automatically followed. The government set the price of the currency and anyone could exchange the currency for gold at the mint or the other way around. But for countries with a central bank, they typically had some leeway in following the rules.
In a world where all countries had equal credibility with markets, all central banks had to follow these rules to the same level. If any of them didn’t follow the rules, their currency would face selling pressure as investors lost confidence that their currency was backed by gold. But this was not true for the gold standard. Some countries due to their higher incomes and long history with the gold standard could break the rules.
Breaking the Rules of the Game
The Bank of England was the center of the financial world when the gold standard was in operation. Britain was the country with the most sophisticated financial markets, and many countries used the sterling as a substitute for gold. So, the Bank of England had a degree of market power in setting its interest rate. Other central banks had to follow the BOE’s lead in setting interest rates, and had to adjust their monetary policy to whatever rate the BOE had set.
Given this market power it had, did the BOE make use of it? The answer most certainly is yes. Had the BOE followed the “rules of the game”, then in months where it faced gold inflows it would increase the money supply, and in months where it faced gold outflows it would have reduced the money supply. But that was not true.
A series of papers showed that the BOE did not do what the rules suggested. In months where the BOE faced inflows, it reduced the money supply about half the time. (Bloomfield (1959)). A stronger test of the “rules of the game” checked if the Bank of England changed the money supply based on expected changes in the trade deficit. But the BOE didn’t change that either, and the correlation was weakly negative. (Dutton (1984))
But in the long run the BOE’s record was impeccable. For periods longer than a year, the BOE’s handling of the money supply matched changes in its gold reserves. (Pippenger (1984)) This suggests that the gold standard was credible because investors had confidence that no matter what, the BOE would follow the rules and maintain gold credibility.
But what about other central banks? Did they break the rules of the game and get away with it?
The answer is that they did, to some extent. The Bank of France also broke the rules of the game. A good way to measure this is if Bank of France interest rates increased when it was facing a drain on gold (gold was leaving the country). The answer is that it didn’t. When the Bank of England raised its interest rate and money flowed from the French money market to Britain, this made money more expensive in France (and increased the interest rate in French money markets).
But the Bank of France did not raise its interest rate to prevent the flow of money to Britain. It expanded the amount of domestic lending it did to reduce the interest rate in France. In theory this would mean that there would be questions on French convertibility.
If gold was leaving the country and the Bank of France was increasing the money supply, did they really back all their currency with gold? That should have left a doubt in traders' minds. But it did not.
During times when gold was flowing into the country, the Bank of France did not use all of it to back money. It kept more than it had to for times when gold was flowing out of France, so it could use it then to back the new credit it was creating.
But the strongest evidence comes from a paper published in 2019 where researchers estimated the impact of BOE interest rate increases on other central bank’s interest rates. If all the central banks had followed the rules of the game, they too would have increased their interest rates in line with the BOE. For every 1% increase in the BOE Bank Rate, foreign central banks in the “core” (developed world) increased their interest rate by only 0.24%. They faced gold outflows of 1.8% of their capital, but covered it up by increasing lending when that happened.
In other words, almost no country followed the rules of the gold standard all the time that supposedly made it so stable. They didn’t increase interest rates when gold flowed out of the country, and they didn’t always maintain perfect gold convertibility. A good test of whether this was caused by the actions of central banks or not is to compare the interest change in the developed world which did have central banks (Germany, France, Austria-Hungary) with the only developed country that did not have a central bank (the USA till 1913). The US had a much sharper change in interest rates (0.5% for every 1% change in the BOE rate) versus the rest of the developed world (0.24%).
How did it work then?
The answer is that monetary authorities in the long run always went back to the rules of the game. They could not escape it for too long. This was because of a few reasons.
The first was that central banks and governments made maintaining the convertibility of their currency an utmost priority in the long run. In the “core” countries of France, Britain, Germany, the US, and Austria-Hungary, there was not a single devaluation in the period from 1870 to 1913. Even during crises like the failure of Barings Bank in 1890, the Bank of England was able to secure emergency gold financing from the Bank of France. Or during the Panic of 1907, the American government would rather let the money supply fall than break the link with gold.
This led to stabilizing speculation. Everytime central banks in the developed world went against the rules of the game, speculators bet that they would still maintain the gold parity. There is extremely strong evidence for this. Contracts for hedging currency risk between gold backed currencies were unheard of during the gold standard period. Along with this, perfect capital mobility led to speculators betting that governments would always honor their gold commitment. (An example is an advertisement in the 1890s which promised "Short-term money moved at an hour's notice from any point of the globe to another.”)
The last factor was political support. Maintaining a gold standard was politically costly. It required interest rate increases during economic downturns. But before World War One, the right to vote was limited to men with property (and completely banned for those who were part of racial minorities). Along with that, there were no parties representing the interests of organized labour, and so governments did not have a strong incentive to boost employment. So investors had the confidence that the government was not going to prioritize employment over the value of the currency.
In the end there are a few takeaways from this
To “cheat” the gold standard, a country had to first build up credibility. Otherwise, markets would lose confidence in it
A central bank could break the rules of the game to get some space for monetary policy in the short run
If the underlying factors which allowed for investor credibility changed, then they could no longer cheat it