The Specie

The promise behind a gold currency was wonderfully concrete: bring the money, receive the specified weight of metal. That promise gave the unit a reference outside the daily decisions of its issuer, and currencies redeemable into the same metal acquired a common measure. In an invented arrangement, one unit of A promises a tenth of an ounce and one unit of B a fiftieth. Their gold contents imply five B for one A, before the costs and risks of conversion. A merchant can write a contract with that relationship in view. Yet fixing the weight of gold is different from fixing the quantity of bread. The metal’s purchasing power can change, and redemption still depends on institutions honoring their undertakings. The anchor made one promise unusually definite; it did not make every meaning of monetary stability the same.¹

The constraint becomes more revealing when the issuer holds metal against only a fraction of its promises. Imagine a monetary authority with forty in gold, valued at its fixed parity, against one hundred of redeemable liabilities. Its rule requires forty-percent gold cover, and it starts exactly at that minimum. A redemption of ten reduces gold to thirty and liabilities to ninety. Cover has fallen to one third. If the authority cannot replenish reserves and must restore the rule, liabilities must fall to seventy-five: another fifteen of contraction after the initial ten. A small outflow can therefore require a larger monetary adjustment. The example deliberately leaves out reserve borrowing, substitute reserve assets, changes in the rule and buffers above the minimum. Give the authority fifty in gold to begin with and the same redemption leaves forty against ninety; that particular rule no longer demands the additional contraction.

The international adjustment also depends on what happens at the other end of the gold shipment. A country facing an outflow may restrain credit and spending to defend conversion. A country receiving gold can permit easier monetary conditions, helping demand move in the opposite direction, or can accumulate reserves while limiting their domestic effect. There is no automatic symmetry between the two decisions: the first may approach a minimum reserve requirement while the second faces no corresponding maximum. In a difficult adjustment, lower domestic spending may reduce imports and put downward pressure on prices, but output and employment can fall while wages and contracts are slow to change. Capital movements and confidence matter alongside trade. The metal does not carry an instruction that makes all these quantities move smoothly, or obliges the surplus country to share the work of adjustment.²

The interwar experience exposed the conflict between defending the external promise and supporting the domestic economy. Bernanke recounts how Britain’s departure from gold in September 1931 was followed by pressure on the dollar and an American interest-rate increase to defend gold, while the banking system was already in distress. A measure directed at retaining reserves could make conditions harder for borrowers and banks at home. But the conflict must be located in its institutions and moment. Reserve cushions, international assistance and policy choices affected how tightly it bound. Bernanke and James expressly distinguish the initial American contraction in 1929, which they say was not forced by a binding external constraint. Gold was neither a passive ornament nor an explanation that made every decision inevitable. A commitment can transmit a crisis without having dictated every act that began it.³

Leaving the old parity could remove a constraint that had made monetary expansion difficult to sustain. The comparison developed by Bernanke and James associates earlier departures with more favorable subsequent monetary and output developments, supporting their explanation of gold’s part in transmitting the Depression. That is an argument from historical variation, not an experiment that makes all the countries identical. Departure coincided with other changes, and its timing itself requires explanation. The comparison identifies a constraint that governments could remove: domestic conditions no longer had to answer to the defense of an increasingly costly conversion rate. They still had banks to repair, debts to work through and confidence to rebuild. Cutting the external tie created room to act; it did not by itself perform every action that recovery required.⁴

The later American announcement of August 15, 1971 made the hierarchy of promises explicit again. Nixon directed the Treasury to suspend the dollar’s convertibility into gold or other reserve assets, describing the suspension as temporary and allowing exceptions on stated policy grounds. In the same address he announced a temporary wage-and-price freeze. Those were different undertakings: one concerned exchanging a monetary claim for a reserve asset, the other attempted to restrain the prices and wages at which domestic transactions occurred. Ending the first promise could not logically establish the success of the second. Nor does the speech by itself explain every subsequent currency arrangement. What it makes plain is that the metallic undertaking could be subordinated to other declared objectives. A rule meant to restrain political discretion ultimately depended on political authorities continuing to maintain it.⁵

The attraction of specie was a commitment whose content could be weighed. Its difficulty was that the price of keeping that commitment could fall on incomes, credit and employment when the adjustment worked badly. An anchor is useful because it limits movement; whether that limit serves the economy depends on the movement being prevented and the means of bearing the strain. The choice is therefore not between institutions and their absence. A metallic standard needs institutions to define redemption, maintain reserves, govern banks and decide what happens in an emergency. A currency without metallic redemption needs institutions capable of making a different promise credible. Gold gave the unit a definite external reference. The anchor was cut when maintaining that reference lost priority to objectives its defenders were no longer willing, or able, to subordinate to it.