The Marshall-Lerner condition and the J-curve: does devaluation improve the trade balance?

Suppose the rupee falls. Indian goods cost fewer dollars, so foreigners should buy more of them. American goods cost more rupees, so Indians should buy fewer. Both effects raise net exports, and the textbook remedy for a trade deficit, devaluation, follows. It is a form of expenditure switching: spending is switched from foreign goods to home goods (Hyderabad 2020 asks what devaluation does: it stimulates aggregate demand, so income rises).

But there is a catch, and it is arithmetic. The goods we still import now cost more rupees each. If we hardly cut our purchases, the import bill goes up, not down. Two forces:

• The price effect works against us: every unit of imports costs more.
• The volume effect works for us: we buy fewer imports and foreigners buy more exports.

The trade balance improves only if the volume effect is strong enough to outweigh the price effect. How strong depends on the price elasticities of demand, which is where the condition comes from.

The papers ask the condition in different words. CUET 2023, Hyderabad 2016 and DSE 2011 ask for the sum of the elasticities (greater than one). Hyderabad 2013 asks when devaluation works best: when export and import demand are both elastic. GATE 2021: if the sum exceeds unity, a depreciation eventually contracts the trade deficit. JAM 2022 adds a twist by defining the real exchange rate: if the rupee depreciates and Marshall-Lerner holds, the trade balance improves and foreign demand for Indian goods rises, but imports fall (so the option that imports increase is wrong).

The Marshall-Lerner condition describes where the balance ends up. It says nothing about the road. In the short run, elasticities are low: importers have orders in place, exporters have signed contracts, and it takes time to find new suppliers and customers. Only the price effect is felt, so the balance worsens first and improves later. Plot it against time and it looks like the letter J.

Play with the lab. The default case has elasticities summing to 1.5, so the long run is positive; slow it down or lower the elasticities and watch the dip deepen or the recovery fail.

Hyderabad 2014 and 2017 both ask what the J-curve illustrates, and the answer is the short-term effects of depreciation on the current account. DSE 2007 turns it round: if the exchange rate appreciates, the J-curve suggests the effect on the trade balance is to improve it in the short run (imports get cheaper at once, volumes have not yet adjusted), the mirror image.

A price taker

A small economy that takes the world prices of both its exports and its imports as given can sell or buy any amount at those prices, so its effective elasticities are infinite and the condition is met. Devaluation then improves the balance of trade (JNU 2010 and 2013).

Starting from a deficit

The textbook condition assumes trade starts out balanced. When imports exceed exports, a 10% rise in the price of imports adds more to the bill than a 10% rise in export volume adds to receipts, so the elasticities must be larger. In general the balance improves if ${\epsilon }_x\frac{X}{M}+{\epsilon }_m>1$, which is Robinson’s form. With exports of 60, imports of 100, and elasticities of 0.5 and 0.55, the sum is 1.05, yet 0.5 × 0.6 + 0.55 = 0.85 and the deficit widens.

No real change

What matters is the real exchange rate. If a nominal depreciation raises domestic prices by the same percentage, Indian goods cost foreigners the same as before and nothing improves. JNU 2013’s option that depreciation “will cause domestic inflation, so there will be no impact” assumes this happens always, which is why it is a distractor: it can happen, but not necessarily.

Pass-through

Foreign exporters often absorb part of the change in their margins instead of raising prices in the market they sell in. Then the home-currency price of imports rises by less than the depreciation, and the effective price effect is smaller. This works in favour of the balance, but it also reduces the volume response.

The elasticities approach looks at prices of traded goods. The absorption approach (Sidney Alexander, 1952) looks at the whole economy through an identity you already know from the balance of payments page:

A deficit means the country absorbs more than it produces. A devaluation can cure it only if it raises output (when there is spare capacity) or reduces absorption (for instance by cutting real money balances or real incomes). If the economy is already at full employment, switching demand to home goods raises prices, not output, and the deficit is cured only if spending is cut as well, which is why devaluation is often paired with tighter fiscal and monetary policy. Papers ask who said what: Keynes for income-induced adjustment, Alexander for absorption, Mundell and Fleming for the impossible trinity (UGC NET June 2025 matches them: A with II, B with III, C with IV, D with I).

One question is left out because the elasticity it relies on is defined in a way the stem does not state (JNU 2014); the reasons are listed at the top of the source file. The two DSE questions above Easy are in their own card.