What Really Moves Exchange Rate Pass-Through? A New Look at Old Data
Emanuel Kohlscheen, Aaron Mehrotra · July 22, 2026
Central banks and investors have long puzzled over why a currency's swing doesn't always show up in consumer prices. This study from the Bank for International Settlements uses a machine learning tool called random forests to analyze nearly a century of data across almost 100 countries. The key insight: the relationship between exchange rates and inflation is highly non-linear, meaning the impact depends heavily on the economic context. For instance, the pass-through is stronger when inflation is already high, when the economy is open to trade, and when the exchange rate move is large and persistent. The method also reveals that traditional linear models often miss these nuances. For investors, this matters because it suggests that currency-driven inflation surprises are more likely in certain environments—like emerging markets with high inflation and open trade—which could affect bond yields, commodity prices, and consumer stocks. The practical takeaway: don't assume a uniform reaction to currency moves; instead, watch the local conditions that amplify or dampen the effect. This research helps refine inflation forecasts and risk assessments, making it a useful tool for portfolio managers navigating global markets.
— This summary is based on research by the Bank for International Settlements. Read the full paper at bis.org.
Educational content only, generated with AI assistance from public research abstracts. Not investment advice.