Currency Depreciation and the Terms of Trade

This short theoretical note is a diagrammatic and elasticity-based intervention in international monetary theory. Its scope is narrow but policy-charged: Haberler asks whether currency depreciation necessarily worsens a country’s terms of trade, and how that question differs from the balance-of-payments effect of devaluation. He begins by rejecting a common inference:

It is very often uncritically taken for granted that devaluation of a country's currency will necessarily result in a deterioration of its terms of trade, that is, in a rise of the average price of its imports in relation to the average price of its exports.

Haberler credits Frank D. Graham and Joan Robinson with correcting this assumption, while also admitting his own earlier carelessness. His main thesis is asymmetrical: in the “regular” case, where depreciation improves the balance of payments, the terms of trade may either improve or deteriorate; in the “perverse” case, where depreciation worsens the balance of payments, the terms of trade must deteriorate.

The first conceptual move is to separate the unavoidable cost of external adjustment from a possible terms-of-trade loss. Under full employment, eliminating a deficit means exporting more or importing less, so domestic absorption must fall whether relative trade prices change or not.

This basic burden is unavoidable even if the terms of trade remain unchanged.

This distinction lets Haberler defend devaluation against protectionist criticism without pretending it is costless. Inflation after devaluation, in his account, is not an independent refutation of exchange-rate adjustment: it reflects wage, fiscal, and monetary policies that prevent society from accepting the real expenditure cut required by balance-of-payments correction.

The formal core comes in the second section, where Haberler uses demand and supply curves for imports and exports, with prices expressed in dollars. A franc depreciation lowers the dollar demand curve for imports; import prices fall in dollars and rise in francs. Export prices likewise fall in dollars and rise in francs. This is the key anti-illusion: one cannot compare dearer imports in francs with cheaper exports in dollars and infer a terms-of-trade loss.

We have, then, the result that export and import prices move in the same direction.

The terms of trade depend on which dollar price falls more. Haberler’s elasticity condition is that the depreciating country’s terms improve when the product of import and export demand elasticities exceeds the product of the corresponding supply elasticities; they deteriorate when the reverse holds. But the condition for a “normal” balance-of-payments response is different, with the familiar sufficient condition that the sum of the import and export demand elasticities exceed unity. Thus a normal balance-of-payments improvement does not imply adverse terms of trade.

Summing up, we may say that, from the prevailing (optimistic) presumption that the balance of payments will be improved by devaluation, it cannot be concluded that there is a similar (pessimistic) presumption that the terms of trade will deteriorate.

The exception proves the asymmetry. If devaluation perversely worsens the balance of payments, export value must fall more than import value; since export quantity rises while import quantity falls, export prices must fall more than import prices. In that case the terms of trade necessarily deteriorate. But the converse does not hold: successful devaluation leaves the terms-of-trade outcome open.

The closing discussion resists Robinson’s suggested presumption that supply elasticities generally exceed demand elasticities because countries are specialized exporters and diversified importers. Haberler, following Hinshaw, regards that as country-specific rather than general, perhaps plausible for some raw-material exporters but not for industrial economies.

I doubt very much whether any broad generalization can be made in this matter.

The brief empirical coda on the 1949 sterling devaluation reinforces the theoretical point: Britain’s deterioration was not decisive evidence, since U.S. terms also worsened despite dollar appreciation, while Australia’s improved despite depreciation. The note’s relevance lies in this disciplined separation of issues: devaluation entails a real adjustment burden, but a terms-of-trade loss is not a theorem and should not be used as a blanket argument for protectionism.