Gold Standard
No automatic pilot

The pre-1914 gold standard is often cited as an example of an international monetary regime which ensured economic stability and growth without continuing adjustments of fiscal and monetary policy by the participating countries. Can stability be achieved with so little effort by governments? In Discussion Paper No. 125, Research Fellow Willem Buiter studies an idealized gold standard using a simple two-country model. Each country is assumed to guarantee the convertibility of its currency into gold at a fixed price as long as its gold reserves exceed some threshold. When a country's gold reserves fall below that threshold, convertibility is ended and a period of free floating begins. The quantity of gold is assumed to be fixed in the model. Buiter considers how monetary and fiscal policies affect the survival of the gold standard and the likelihood of either currency being the victim of the speculative selling attack that ends it.

The behaviour of reserves and the survival of the gold standard in Buiter's model depends on the difference between home and foreign Domestic Credit Expansion (DCE), and the difference between changes in the home and foreign demand for money. Buiter assumes that these variables can be represented by continuous- time versions of a random walk (in which the value next period will be equal to the current value plus an independent random disturbance). Buiter also incorporates trend behaviour in these variables by allowing 'drift' in the random walk.

If money demand growth and DCE behave as random walks, the gold standard is not viable: it is certain to collapse within a finite period. There are, however, some policies that help postpone the inevitable collapse. Not surprisingly, a rise in the world stock of physical gold, or of 'paper gold' created by international organizations such as the IMF, increases the expected duration of the gold standard. Buiter also finds that a given stock of reserves can be redistributed between the two countries so as to increase the life expectancy of the gold standard. Such redistributions may be difficult to achieve voluntarily, however, since they make the country that has given up reserves more likely to be the victim of the selling attack that ends the gold standard.

Policies which change the behaviour of DCEs alter the expected survival period of the gold standard, but unless policy completely offsets the exogenous shocks to the economy, it only postpones the inevitable exhaustion of one country's reserves. Buiter demonstrates that responsive or 'conditional' policy rules for DCE can sustain the gold standard indefinitely. One such policy rule adjusts DCE immediately to accommodate changes in the demand for money which originate in shocks to the economy. By continuously matching variations in the stock of domestic credit with real-income-related variations in money demand, the level of reserves is stabilized perfectly and the viability of the gold standard seems assured.
Buiter notes that this policy is problematic, however, for by stabilizing the stock of reserves it risks destabilizing the national debt. Suppose that in addition to following this policy for DCE, the authorities balance the budget continuously: variations in the stock of domestic credit are therefore matched by variations in the national debt. In the simplest case, the national debt will follow a random walk and as a result will, over time, exceed any limit, however high. An international reserve crisis has been exchanged for a government solvency crisis.

Buiter argues that the gold standard is essentially an extreme version of a commodity price stabilization scheme, which aims not merely to stabilize but to fix the price of a commodity-gold. He concludes that its operation is far from automatic and its survival is not assured. It cannot survive if governments follow 'unconditional' policy rules. To avoid both the collapse of the gold standard and a public sector solvency crisis, monetary and fiscal policy must respond flexibly to disturbances to the economy.


A Gold Standard Isn't Viable Unless Supported by Sufficiently Flexible Monetary and Fiscal Policy
Willem Buiter

Discussion Paper No. 125, August 1986 (IM)