At the time of writing, nominal exchange rates have been particularly volatile. Large exchange rate movements can weigh on export competitiveness, engender price instability, and disrupt financial systems. To mitigate these swings, the Fed has extended \$90 billion in new dollar swap lines to nine countries, the Norges Bank has conducted its first foreign exchange intervention (FXI) in 20 years, and the Swiss National Bank has sold $8 billion worth of franc in a single week. This dissertation studies the effectiveness FXI and investigates secular changes in the responsiveness of trade to exchange rate movements.
Chapter 1 investigates trends in expenditure-switching for 17 countries using a factor-augmented vector auto-regression with time-varying loadings. I find that exports and imports have become less responsive to unexpected exchange rate movements, but export and import price responses have been stable. This implies that country export supply curves have become more inelastic while global export demand has become more elastic. Similarly, country import demand curves have become more inelastic, but global import supply has become more elastic. This is consistent with the rise in global value chains: multiple stages of production make it difficult for exporters to swiftly increase production in response to demand shocks, and intermediate inputs make imports less responsive to price changes. It is also consistent with the view that trade integration has increased global competition.
Chapters 2 and 3 are adopted from a joint paper with Nelson Mark. A key transmission channel for achieving central bank policy objectives works via managing financial market expectations. Chapter 2 provides daily estimates of market expectations of future foreign exchange (FX) spot prices. Exchange rate expectations are risk-adjusted futures prices implied by a term-structure model of FX futures.
Chapter 3 employs the event study method to test if FXI causes changes in exchange rate expectations. Endogeneity of interventions are addressed via synthetic control. From a panel of 9 countries, we find that FXI leads the market to revise expectations by an average of 50 basis points in the direction predicted by the intervention. Under certain conditions, larger changes and success rates of 90\% are achievable. Changes in risk premia are also detected, but are less systematic, suggesting that FXI may contribute to FX market uncertainty.