Real exchange rates: competitiveness is not a trading clock

One pair cannot describe every trading relationship

A bilateral exchange rate compares two currencies. An effective exchange-rate index combines several counterparts using weights. That distinction matters when a currency weakens against the dollar but strengthens against most trading partners. A single prominent pair may describe a funding exposure well while describing the country's overall trade-price environment poorly.

An index also depends on its methodology. Partner selection, weights, price measures and rebasing all matter. A level of 120 relative to a base-year value of 100 means an index rose 20% under its definition; it does not by itself mean the currency is 20% overvalued. The base year is a reference point, not a certified equilibrium.

Weights do not simply count neighboring countries. The BIS methodology incorporates manufacturing trade relationships and competition in third markets, so a partner can matter because its exporters compete with domestic firms elsewhere. Weights also change over time. A broad index and a narrow index may therefore answer somewhat different questions even for the same currency. Check the basket and methodology before treating a difference between two published indices as a data error.

Inflation changes relative purchasing costs

A real exchange-rate measure adjusts nominal exchange rates for relative prices. In a convention where a higher index means appreciation, domestic prices rising faster than partner prices can produce real appreciation even with an unchanged nominal rate. This is one reason nominal depreciation need not translate one-for-one into improved cost competitiveness.

The chosen price index affects interpretation. Consumer prices describe a broad consumption basket, not exactly the costs or export prices of every industry. A manufacturer using imported inputs faces different pressures from a domestic service provider. An aggregate real exchange-rate measure is a useful summary, but it cannot replace a sector-specific cost and revenue analysis.

For a bilateral illustration, let E be foreign-currency units per domestic unit. An appreciation-oriented real measure is proportional to E × Pdomestic ÷ Pforeign. If the domestic currency depreciates 10% in that quotation while domestic prices rise 12% and foreign prices rise 2%, the combined factor is 0.90 × 1.12 ÷ 1.02 ≈ 0.9882. The real depreciation is only about 1.18%. Adding percentage changes gives a rough shortcut; multiplication gives the exact result under this simplified definition.

Competitiveness and valuation are different questions

A real appreciation may make some domestic goods more expensive relative to foreign alternatives, but it can also coincide with improved productivity or a shift toward higher-quality exports. Not every historical departure must reverse. Structural changes can alter the level consistent with sustainable trade and financing patterns.

Valuation frameworks therefore require assumptions about productivity, terms of trade, capital flows and external balances. A simple long-run average is a reference statistic rather than a sufficient model. If a currency has spent years above that average, repeatedly predicting immediate reversal does not become more justified merely because the deviation persists. The adjustment horizon is itself uncertain.

Productivity offers one reason a changing real exchange rate need not be a mistake waiting to correct. More productive firms may support higher wages, including in services that are not internationally traded, without losing the same amount of competitiveness in their export industries. Consumer-price-based real appreciation can then coexist with healthy export performance. The question is whether prices, productivity and external financing are consistent—not whether the index has returned to an aesthetically appealing historical average.

Trade and financial exposures can conflict

Currency depreciation can support some exporters while increasing the domestic burden of foreign-currency debt. An economy with large unhedged foreign liabilities may therefore experience financial tightening even when its exporters gain price competitiveness. A trade-weighted measure and a debt-weighted measure answer different questions about the same currency movement.

For a company, map revenue currency, input currency, debt currency and hedges separately. A nominally domestic exporter can have imported inputs and dollar debt that offset part of the revenue benefit. Macro labels such as winner from a weak currency conceal these balance-sheet details. The relevant exposure is the net effect under plausible volumes, prices and financing conditions.

Pass-through also affects who bears a depreciation. An exporter billing in dollars may keep its dollar price unchanged and receive more domestic currency, improving its margin rather than immediately making its goods cheaper to foreign buyers. Imported inputs can offset that gain, and customers may renegotiate prices later. The exchange rate, invoicing currency, contract duration and competitive response jointly determine the outcome. A trade-weighted index alone cannot identify which firms gain or by how much.

Use the measure as context

For historical practice, compare nominal and real indices over a clearly stated period, using the same vintage where possible. Explain whether inflation or nominal exchange rates drove the difference. Record the data frequency; monthly price-based indices cannot identify an intraday entry signal without introducing information that was unavailable at the time.

Then construct an original sensitivity exercise rather than declaring a target price. Ask how much real appreciation remains if domestic inflation exceeds partner inflation by a given amount, or how an exporter fares when imported input costs rise. These exercises make the measure operational while respecting its limits. A valuation argument still needs a catalyst, horizon and risk assessment before it can become any investment thesis.

Repeat the index calculation with a different arbitrary base year. The percentage change between two dates should remain the same even though the displayed index levels change. If your assessment of cheapness changes solely because the base was reset, the assessment is relying on index presentation rather than an economic valuation argument.

Keep rebasing separate from a methodological revision. Changing an index from 2015 = 100 to 2020 = 100 rescales the series and should not alter returns between the same dates if nothing else changes. Updating weights, revising price data or changing coverage can alter those historical returns. Save the release vintage and methodology when reproducing an old study; otherwise, a difference attributed to economic insight may actually come from using a revised series.