Fritz Machlup, “The Need for Monetary Reserves” (1966)

Machlup’s article is a conceptual and empirical critique of the language of “reserve needs” in the Bretton Woods system. Its starting point is not a technical formula but a prior question: whether monetary authorities, individually or collectively, can be said to need reserves in any objective sense.

This article will address itself to the question whether it is possible to find any objective criteria for the need of monetary reserves, either for individual countries or for the world at large.

The essay first narrows the meaning of international liquidity. Machlup distinguishes measurable official reserves from broader and more elusive sources of liquidity, such as borrowing capacity or assets that might be sold in an emergency. This narrowing is important because many debates about reserve adequacy smuggle in unmeasured possibilities while presenting conclusions as statistical facts. He also insists that “need,” “desire,” and “demand” are not interchangeable terms.

It may be a little tedious if I point here to the popular confusion among three words: need, desire, and demand. I have done it before, but find it important enough to repeat (1).

For Machlup, a need is defined by the consequences that follow if the needed object is absent. Applied to reserves, this means asking what reserves prevent: devaluation, deflation, exchange controls, trade restrictions, or loss of confidence. This framing allows him to criticize the familiar reserve ratios then used in policy discussion. Ratios of reserves to imports, exports, money supply, central-bank liabilities, or past balance-of-payments deficits may describe historical patterns, but they do not prove objective requirements. They often confuse convention with causation.

A major part of the article examines why gross reserve figures dominate international-liquidity statistics. Machlup argues that this is not accidental: under the gold-exchange standard, a country’s liabilities to foreign monetary authorities were also another country’s reserve assets. Netting them out would erase the very structure that made reserve-currency holdings function as reserves.

That most statistical tables of international liquidity report gross reserves, rather than net reserves, is fully consistent with the principles of the gold-exchange standard.

The empirical appendix reinforces the argument by showing wide variation among fourteen industrial countries from 1949 to 1965. Some countries maintained very large reserves relative to trade or monetary aggregates, while others operated with much lower ratios. These differences cannot be explained by a single reserve formula. Nor do past reserve losses justify observed holdings: some countries held far more reserves than any plausible insurance calculation would require, while reserve-currency countries faced special vulnerabilities.

Machlup’s deeper point is behavioral and institutional. Reserve holdings are not simply chosen according to rational demand; they are often the result of what monetary authorities are willing to accept under fixed exchange rates. A surplus country may accumulate reserves because resisting inflows would require politically difficult measures, such as appreciation, import liberalization, or domestic expansion. A deficit country may want more reserves not because a formula says so, but because reserve losses threaten policy embarrassment and pressure toward controls.

Since the propensities to spend, lend, and invest differ among the holders of cash balances and among the holders of reserves, the same total may mean very different rates of spending, lending, and investing, depending on its distribution.

The article therefore rejects the search for a timeless global stock of “adequate” reserves. But it does not deny every practical case for reserve creation. Machlup shifts the issue from the level of reserves to additions to reserves. In a growing world economy with fixed exchange rates and politically constrained adjustment, incremental reserves may reduce the frequency with which countries respond to deficits by imposing restrictions or pursuing contractionary policies.

It is in precisely this sense that one may speak of the need for additions to total reserves of the monetary authorities in the countries that form the international economy (in the non-communist world).

Machlup’s conclusion is anti-mechanical but not anti-reform. He denies that reserve adequacy can be read from simple ratios, yet allows that additional international reserves may serve a liberal policy purpose if they help sustain open trade and payments. The article’s lasting contribution is to replace mystical reserve rules with an account of central-bank behavior, political tolerance for adjustment, and the institutional logic of the gold-exchange standard.