Essays on exchange rates and capital flows

This dissertation studies how financial flows move exchange rates and shape stabilization policy. Across three chapters, it (i) measures FX market depth using granular holdings data, (ii) characterizes optimal FX intervention (FXI) with limited reserves, and (iii) examines the opportunity cost of reserve accumulation from an external balance sheet perspective. Chapter 2 uses security-level holdings of mutual funds and a granular instrumental-variable strategy to identify currency demand shocks. These shocks generate significant exchange rate movements in both emerging market (EM) and advanced economy (AE) currencies, consistent with shallow FX markets. For equal-sized flows, EM currencies respond about nine times more than AEs, implying lower depth. Depth is state-dependent: FX markets are highly inelastic when expected exchange rate volatility is high but close to perfectly elastic in tranquil periods. During mutual-fund outflows, EM FX markets are shallower than during inflows. A quantitative small open economy model with segmented FX markets, limited intermediary risk-bearing capacity, and balance sheet constraints rationalizes these patterns. Chapter 3 analyzes optimal time-consistent FXI in a small open economy driven by endowment and currency demand shocks, with endogenous FX market depth and an effective lower bound on reserves. Large flows raise conditional exchange rate volatility and reduce depth, amplifying exchange rate responses. Optimal policy is state-dependent and forward-looking since current interventions affect the risk of reserve depletion. FX sales are more effective than FX purchases, and the optimal response to outflows can be less than or more than one-for-one depending on flow size and the net foreign asset position. Optimal policy yields substantial welfare gains over simple rules, provided reserves are sufficiently above the effective lower bound. Chapter 4 studies stabilization-driven reserve accumulation and its crowding-out of private foreign investment. Reserves used to support FXI can crowd out higher-yielding foreign investment, creating an opportunity cost. The policy problem features two local optima: one with low reserves and positive foreign investment, and one with high reserves and no foreign investment. Which regime is globally optimal depends on the risk–return profile of foreign investment.