Long-Horizon Exchange Rate Expectations

Exchange rates have always been one of the noisiest variables in global finance. 

If you trade interest rates, the rate is either there by date X or it’s not. If you trade bonds, they move based on duration and sometimes credit risk, but the payout is usually known ahead of time.

But exchange rates are perpetual and even managed/fixed rate regimes ultimately have to remain consistent with economic fundamentals.

Traders, economists, and policymakers all look for the inputs that impact the output. Yet short-term movements still appear relatively “random” – i.e., movements that aren’t eminently explainable (let alone predictable) by all input variables. 

The study we cover in this article takes a different view by looking at long-horizon expectations two years ahead and whether that provides a clearer signal.

Does professional judgment, informed by the macroeconomic knowledge that underlies currency movements, actually capture long-term trends?

Key Takeaways – Long-Horizon Exchange Rate Expectations

• Exchange rates look chaotic short term and noise hides the signal.
• Two-year expectations nonetheless show skill, and longer horizons filter daily volatility.
• Professionals predict direction better than chance.
  • Fundamentals eventually drive currency moves.
• Three macro forces matter most: risk covariance, real exchange rate, and current account.
• After those, no hidden “alpha” was observed to remain.
• Markets swing more than fundamentals because they constantly reprice the future.
• Time reveals fundamental trends.

Overview of Long-Horizon Exchange Rate Expectations

Purpose and Research Motivation

The motivation comes from decades of research showing that exchange rates often diverge from economic fundamentals in the short run – only to realign later. 

Short-term currency movements are driven by speculation, while long-term trends eventually reflect underlying economic fundamentals.

Understanding that realignment process could help explain how expectations form and evolve among professionals who trade and advise in the global currency market.

Definition and Context of Long-Horizon Forecasting

Long-horizon forecasting focuses on identifying broad directional trends (e.g., currencies increasing in value because they pay higher interest) rather than the short-term fluctuations. 

For example, if one currency pays 7% per year and it’s funded against a currency that pays only 2% interest, that’s 5% a year in carry.

If the trader’s strategy is based on that, that’s a daily yield of 0.01-0.02%, but obviously the actual daily movement will be more significant due to transactive activity.

Even if they take into account inflation differentials, interest rate parity, and current account balances, daily exchange rate movements are still going to be “overreactive” relative to the slower forces of economic adjustment.

So, studying expectations two years ahead can better allow for testing whether traders truly incorporate these economic signals or simply extrapolate recent performance into the future.

Data from Financial Professional Surveys

The data used in the study come from structured surveys of financial professionals who track major currencies as part of their work. 

These surveys provide a window into the collective expectations of those shaping currency markets, so there’s both informed analysis and market sentiment.

Forecast Horizon and Empirical Approach

Using a two-year horizon allows enough time for macroeconomic fundamentals to influence exchange rates while still keeping forecasts relevant for investment and policy planning. 

Just like with a stock index, if you expect to get 7% per year, after two years you’re at a 14% rise (including just the earnings), which is usually enough to see rises overall that outweigh the noise. But certainly after a day, week, month, or quarter, you’re going to much less sure.

Changes in discounted growth, discounted inflation, discount rates, and risk premiums, as the main four macro variables, will alter the actual return.

Forecasting Performance and Predictive Accuracy

Evidence of Successful Two-Year Forecasts

The results show that financial professionals do, in fact, display skill when forecasting exchange rate movements over a two-year horizon. 

Their average expectations point in the correct direction of future currency appreciation or depreciation more often than chance would predict. 

And this accuracy isn’t limited to specific currencies or time periods but appears broadly consistent across major exchange rate pairs (e.g., USD, EUR, GBP, JPY). 

The finding challenges the belief that exchange rates behave like random walks.

At longer horizons, expectations better reflect genuine information about economic fundamentals.

In-Sample and Out-of-Sample Forecast Comparison

The analysis tests predictive success in both in-sample (within training data) and out-of-sample (outside of training data) settings. 

In-sample performance confirms that survey expectations align with actual currency changes within the studied dataset. 

More impressively, when models trained on one time period are used to forecast a later one, predictive accuracy holds up. 

Accordingly, professional forecasts reflect stable relationships that persist over time. 

Out-of-sample validity is important because it mirrors real-world forecasting. Decisions must rely on information available at the time.

Contrast with Short-Term Forecast Limitations

Short-term forecasts often fail because exchange rates react to unpredictable shocks like policy announcements, risk sentiment swings, influences from other markets, or sudden capital flows. 

Markets are always trying to price the future, which leads to price movements that are larger than the fundamentals would otherwise dictate.

Markets discount the future into the present value. Because small changes in assumptions, like discount rates, growth or inflation expectations, or risk premium (when it comes to many asset classes), can greatly alter those present valuations, prices typically swing far more than the underlying fundamentals themselves.

Over two years, however, these shocks tend to balance out, which allows deeper macroeconomic forces to emerge. 

The professionals in the surveys in these surveys seem to intuitively capture these slower-moving fundamentals rather than chase daily volatility. 

Their success at longer horizons shows that time itself is generally the best way filter signal from noise in currency markets.

Statistical Tests of Predictive Strength

Formal statistical tests by the authors confirm that survey-based expectations significantly predict future exchange rate changes, even after accounting for randomness and benchmark models. 

The magnitude of predictive power is modest but meaningful.

Accordingly, informed expectations carry real content. 

Determinants and Interpretation of Expectations

Key Macro-Finance Variables Explaining Expectations

When researchers examined what drives professional currency forecasts, three macro-finance variables stood out.  Together, they explained most of the variation in exchange rate expectations across time and currencies. 

These were:

• the risk-neutral covariance* between the exchange rate and the equity market
• the real exchange rate**, and
• the current account balance relative to GDP**

Each represents a different way in which economic and financial conditions shape beliefs about future currency values.

* Measuring how a currency moves with the stock markets once traders’ attitudes toward risk are factored in. Some currencies are “risk-on” (e.g., commodity currencies) while some are more “risk-off” (e.g., CHF).

** The nominal exchange rate adjusted for inflation differentials between countries.

*** A country’s total income from exports, investments, and transfers versus what it spends abroad. This shows whether it’s a net saver (surplus) or borrower (deficit) in the global economy.

Risk-Neutral Covariance Between Exchange Rate and Equity Markets

The risk-neutral covariance measures how currencies move relative to stock markets in an environment free of risk preferences. 

When currencies tend to strengthen in bad equity markets, traders/investors view them as safe havens. Over recent history (i.e., decades), this has commonly meant JPY, CHF, or gold, though it depends.

“Safe haven status” increases expected long-term appreciation. 

When currencies weaken alongside falling equities, expectations adjust downward. 

This link shows how risk perception and portfolio diversification drive professional judgment about future exchange rates.

Effects of the Real Exchange Rate and Current Account to GDP

The real exchange rate measures how expensive a country’s goods and services are compared with those abroad. 

When it becomes too high, professionals often expect eventual relative depreciation in a currency to restore competitiveness. 

The current account to GDP ratio shows whether a nation saves more than it spends/invests. 

Persistent surpluses or deficits signal structural pressures that shape how traders anticipate future currency changes.

Surpluses are positive for a currency. Deficits are negative for a currency.

Testing for Residual Informational Value (“No Secret Sauce”)

Once these three variables are considered, they found that the remaining variation in survey expectations adds little predictive power. 

In others words, there was no hidden layer of private insight or trader instinct that improves forecasts beyond fundamentals.

Broader Implications for Exchange Rate Modeling and Expectation Formation

Overall, this evidence suggests that professionals’ expectations are pretty well grounded in economic reality. 

It reinforces the idea that exchange rates can be noisy over short-term timeframes, but they eventually align with measurable macroeconomic forces.