Breaking Parity: How Dealer Bank Frictions Shape Currency Premia and Exchange Rates

In the working paper “Breaking Parity: Equilibrium Exchange Rates and Currency Premia”, economists Mai Chi Dao (IMF), Pierre-Olivier Gourinchas (University of California, Berkeley and IMF), and Oleg Itskhoki (Princeton University) propose a paradigm-shifting explanation for exchange rate dynamics that diverges sharply from standard macroeconomic models. Published by the International Monetary Fund (IMF), the paper addresses a key puzzle in international finance: why do exchange rates exhibit weak and inconsistent links with macroeconomic variables such as interest rates, inflation, or output? The authors present a partial equilibrium model that emphasizes the role of frictions in currency intermediation, arguing that the true drivers of exchange rates are the constraints and behaviors of financial intermediaries, particularly global dealer banks, rather than macroeconomic fundamentals alone.

Dealer Banks and Currency Premia: A Micro-Level Mechanism

The crux of the paper lies in how it reframes currency premia, traditionally seen as market inefficiencies, as rational outcomes arising from frictional intermediation. In this framework, intermediary banks face value-at-risk constraints on their balance sheets, requiring compensation to absorb net positions in foreign exchange markets. This compensation takes the form of covered and uncovered interest parity (CIP and UIP) premia. Unlike standard models that assume risk-free arbitrage and perfect capital mobility, this model recognizes that intermediaries are capital-constrained and thus only engage in swaps or forward trades when returns justify the added risk or liquidity cost.

This leads to a clear division across currencies. Funding currencies like the Japanese yen (JPY) and Swiss franc (CHF), which have a surplus of local savings, tend to exhibit low interest rates and negative UIP and CIP premia. On the other hand, investment and commodity-linked currencies like the Australian dollar (AUD), New Zealand dollar (NZD), and Mexican peso (MXN) are characterized by high interest rates, scarce domestic savings, and positive currency premia. Investors seeking higher returns must hedge their exposures by selling these currencies forward, which increases demand for dollar hedging and raises the forward premium.

Futures Positions as a Window into Currency Demand

One of the paper’s key empirical contributions is the use of dealer banks’ net futures positions, drawn from the CFTC’s Traders in Financial Futures (TFF) reports, as a proxy for currency-specific demand shocks. These net positions serve as a quantifiable and timely indicator of shifts in hedging and speculative pressures in the FX market. The authors demonstrate that large changes in these positions lead to immediate depreciations in the spot exchange rate, followed by gradual and predictable appreciations. This pattern creates the UIP premium: investors entering after the initial shock earn a small, positive return over time.

What makes this relationship particularly powerful is its consistency. Across major currencies in the G7 and select emerging markets, changes in dealer futures positions explain over 60% of the monthly variation in spot exchange rates. The response is not only statistically significant but also economically meaningful. A one standard deviation increase in dealer net positions leads to a depreciation of around 0.9–1.3% in the local currency, depending on the market. These exchange rate adjustments are persistent, typically exhibiting half-lives of six to eight months.

Time Series vs. Cross Section: The Double Life of Currency Premia

A striking aspect of the analysis is its distinction between cross-sectional and time-series behavior of currency premia. In the cross-section, UIP premia align well with interest rate differentials: high-interest currencies tend to offer positive expected returns, while low-interest ones deliver negative or negligible returns. This is consistent with the classic carry trade logic. However, in the time series, the relationship breaks down. The authors show that fluctuations in UIP premia are largely unrelated to changes in interest rate differentials and are instead tightly tied to dealer bank activity. In contrast, CIP premia exhibit minimal volatility over time and respond mostly to aggregate financial conditions such as global risk aversion (measured by the VIX) or shifts in dollar funding demand.

This divergence is key to solving the forward premium puzzle, the empirical finding that high-interest currencies do not depreciate as much as UIP would predict. In the authors’ model, frictional intermediation explains the puzzle: spot and forward exchange rates adjust to ensure equilibrium, not interest rates, as intermediary banks accommodate shocks in net currency demand.

Implications for Policy and Global Finance

The insights from this research carry profound implications for currency policy, FX market monitoring, and macroeconomic modeling. By emphasizing intermediary balance sheet constraints over macro fundamentals, the authors suggest that central banks and financial authorities must pay closer attention to the behavior of dealer banks and currency-specific demand flows. Traditional monetary policy tools, such as interest rate adjustments, may be less effective at influencing exchange rates than assumed, especially during periods of global financial stress when intermediary constraints tighten.

Moreover, the methodology highlights the predictive power of monitoring FX futures data for assessing potential volatility in currency markets. The framework also bridges a gap between microstructure-based and macroeconomic approaches to exchange rates, offering a rare synthesis that is both theoretically elegant and empirically grounded. In a world of globalized finance, where cross-border capital flows are increasingly intermediated by a small set of powerful financial institutions, understanding the mechanics of currency intermediation is no longer optional, it is essential.