Every day in the FX market, capital shifts between two competing instincts: the hunt for yield and growth, and the flight to safety. Currencies like the Australian dollar, New Zealand dollar, and Canadian dollar tend to strengthen when investors feel confident — risk is being embraced. The Japanese yen and Swiss franc tend to strengthen most sharply when fear dominates — risk is being avoided. The US dollar occupies a nuanced middle ground: it is a safe haven against emerging market and commodity currencies, but in extreme risk-off episodes, it can weaken against JPY and CHF as Japanese investors repatriate holdings and carry trades are unwound. This alternation between confidence and caution is what traders call the risk-on / risk-off dynamic, and it is one of the most persistent structural forces in global FX.
FXMacroData now publishes a composite daily Risk Sentiment indicator that distils this dynamic into a single score ranging from -1.0 (extreme risk-off) to +1.0 (extreme risk-on). Each day is tagged with a regime label: risk_on, neutral, or risk_off.
API ENDPOINT
The indicator is available at:
GET /api/v1/risk_sentiment?start_date=2024-01-01&end_date=2025-01-01&api_key=YOUR_KEY
Why Risk Sentiment Matters for FX
Currency markets do not move in isolation. While macro fundamentals — growth differentials, inflation, central bank policy — set the long-run trend, risk appetite is often the dominant driver of day-to-day and week-to-week moves in commodity and growth-linked currencies.
This happens for structural reasons. Countries like Australia and Canada are large commodity exporters. When the global economy is growing strongly, commodity demand rises and their currencies strengthen. Simultaneously, investors tend to reach for higher-yielding assets in strong economic environments — again favouring currencies like AUD and NZD that historically carry positive interest rate differentials. The reverse happens when fear grips markets: capital retreats into the deepest, most liquid markets in the world. US Treasury bonds, Japanese government bonds, and Swiss franc cash are the traditional beneficiaries.
Understanding whether the current market environment is risk-on or risk-off helps traders:
- Identify which FX pairs are likely to outperform in the current regime.
- Size positions appropriately when market volatility shifts regime rapidly.
- Filter directional macro signals — a bullish fundamental case for AUD/USD carries different weight in a risk-on vs risk-off environment.
- Anticipate momentum and mean-reversion dynamics across correlated asset classes.
How the Composite Score is Constructed
The FXMacroData Risk Sentiment score is a weighted composite of four independently normalised signals, each sourced from official or widely recognised public data providers.
Component 1: OFR FSI — Financial Stress Index (weight: 40%)
The OFR Financial Stress Index (OFR FSI), published daily by the U.S. Treasury's Office of Financial Research, measures the level of stress in the U.S. and global financial system. It aggregates conditions across credit markets, equity volatility, funding markets, and safe-haven demand into a single index value — where positive readings indicate elevated financial stress and lower or negative readings indicate calmer, more orderly market conditions.
In the composite, the OFR FSI is inverted: a high reading — signalling widespread financial stress — contributes negatively to the score, pushing it toward risk-off. A low or negative OFR FSI reading (calm markets) contributes positively to the risk-on signal. The OFR FSI carries the largest weight of any single component at 40%, reflecting its breadth: unlike single-market indicators, it synthesises stress signals from across the entire financial system into one daily number.
Component 2: Gold Price — Safe-Haven Demand (weight: 20%)
Gold (XAU/USD) is the world's oldest safe-haven asset. When uncertainty rises, institutional and retail investors alike increase gold allocations as a hedge against volatility, currency debasement, and tail risk. Central banks also increase gold reserves during periods of geopolitical instability.
The gold price used in this composite is sourced from The Royal Mint (https://www.royalmint.com/gold-price/), the globally recognised benchmark price in USD/troy oz published daily on London business days. This is the same data behind the /api/v1/commodities/gold endpoint.
Gold is also inverted in the composite: a rising gold price, reflecting heightened safe-haven demand, pushes the score toward risk-off. A falling gold price — suggesting investors are confident enough to reduce defensive holdings — contributes positively to the risk-on reading.
Component 3: AUD/USD — Risk-Sensitive Currency Pair (weight: 20%)
AUD/USD is one of the FX market's most reliable risk barometers. The Australian dollar is heavily linked to global commodity prices (particularly iron ore and copper), Chinese economic activity, and the global growth cycle. It is also a carry currency: when risk appetite is healthy, investors tend to hold AUD for its yield advantage over the safe-haven alternatives.
A rising AUD/USD implies investors are moving toward risk, and this contributes positively to the composite. A falling AUD/USD signals risk aversion and pushes the score negative.
Component 4: USD/JPY — Safe-Haven Flow Proxy (weight: 20%)
USD/JPY captures a different dimension of risk appetite: the flight-to-JPY dynamic. The Japanese yen benefits from a structural safe-haven premium rooted in Japan's status as the world's largest net creditor nation. During risk-off episodes, Japanese investors repatriate overseas holdings, and global investors unwind JPY-funded carry trades — both dynamics driving JPY higher (USD/JPY lower).
In the composite, a falling USD/JPY (JPY strengthening) is risk-off, and a rising USD/JPY (JPY weakening) reflects risk appetite returning. This component complements AUD/USD by providing a cross-check from the safe-haven demand side, rather than the risk asset demand side.
Signal Normalisation: Rolling Z-Scores
Each of the four raw series moves in very different units — financial stress index values, USD/troy oz, and exchange rate levels. Direct comparison is meaningless without normalisation. The composite uses a 252-day rolling z-score to normalise each component: a measurement of how many standard deviations the current reading is from its rolling one-year average.
This has several important properties:
- Scale-invariant: an OFR FSI reading of 0.5 is only meaningful in the context of its recent range. The z-score contextualises the current level relative to recent history.
- Regime-adaptive: the rolling window means the normalisation adjusts as market conditions structurally shift. An OFR FSI of 1.5 in a calm period signals elevated stress; in a crisis period it may be moderate.
- Comparable across components: once expressed as z-scores, all four series can be weighted and combined on equal mathematical footing.
The composite weighted z-score is then passed through a tanh(0.6 × z) compression, which maps the raw score into the bounded [-1.0, +1.0] range while preserving the ordering and reducing the influence of extreme outliers. The choice of 0.6 as the scale factor means a one-standard-deviation move maps to approximately ±0.54 — leaving room for genuine extreme readings to approach ±1.0 without the score being saturated by normal volatility.
Regime Labels
Each daily observation is classified into one of three discrete regimes based on the final composite score:
- risk_on (score ≥ 0.25): Broad risk appetite. AUD, NZD, CAD typically outperform. JPY and CHF tend to underperform. Carry trades and commodity-linked positions are typically favoured.
- neutral (-0.25 < score < 0.25): Mixed or transitional conditions. No clear directional bias from the risk sentiment indicator alone. Other macro signals should dominate positioning decisions.
- risk_off (score ≤ -0.25): Broad risk aversion. JPY and CHF typically outperform most strongly. AUD, NZD, CAD, and EM currencies tend to underperform. USD also benefits as a safe haven against commodity and EM currencies, though it may weaken against JPY and CHF in extreme episodes as carry trades unwind. Defensive positioning and safe-haven demand dominate.
API Response Format
Each record returned by the endpoint includes the composite score, the regime label, and the normalised contribution of each component:
{
"start_date": "2025-01-01",
"end_date": "2025-03-31",
"data": [
{
"date": "2025-03-31",
"val": -0.42,
"regime": "risk_off",
"components": {
"ofr_fsi": -0.38,
"gold": -0.31,
"aud_usd": 0.12,
"usd_jpy": -0.29
}
}
]
}
The components values are the normalised z-scores as they contribute to the composite (i.e., after inversion where applicable). Negative values in components.ofr_fsi and components.gold indicate elevated financial stress and elevated gold prices relative to their rolling average — both risk-off signals.
Practical Applications
Filtering Directional Macro Signals
Suppose the FXMacroData announcements API shows that Australia has just released a stronger-than-expected employment report. In isolation, this is AUD-bullish — it reduces the likelihood of RBA rate cuts and signals a healthier domestic economy. However, if the risk sentiment indicator is firmly in risk-off territory on that date, the AUD-bullish fundamental signal may be overwhelmed by broad risk aversion. Tracking both together provides a more complete picture of the likely market response.
Regime-Specific Carry Trade Positioning
Carry trades — borrowing in low-yielding currencies (JPY, CHF) to hold high-yielding ones (AUD, NZD) — are structurally profitable in calm, risk-on environments but violently unwind during risk-off episodes. The risk sentiment indicator provides a systematic way to assess whether current conditions are conducive to carry strategies, reducing exposure during periods of elevated risk aversion.
Correlating with COT Positioning
The CFTC Commitments of Traders (COT) data available via the FXMacroData COT endpoint shows the net speculative positioning of managed money in major FX futures. Periods of extreme risk-off often correlate with large short positions in AUD, NZD, and CAD — and large long positions in JPY. Comparing the risk sentiment score to COT extremes can identify mean-reversion opportunities when positioning becomes one-sided.
Data Sources and Licensing
All four underlying data series are sourced from official or widely recognised public providers:
- OFR Financial Stress Index: Published daily by the U.S. Treasury Office of Financial Research (OFR) at financialresearch.gov. Free to use and publicly accessible.
- Gold (spot price): The Royal Mint (https://www.royalmint.com/gold-price/). Gold spot price in USD/troy oz — free for public use.
- AUD/USD and USD/JPY exchange rates: Frankfurter API — an open-source currency data service backed by the European Central Bank reference rates.
Getting Started
The Risk Sentiment endpoint is available to all FXMacroData API subscribers. Query it directly with your API key:
GET https://fxmacrodata.com/api/v1/risk_sentiment
?start_date=2024-01-01
&end_date=2025-01-01
&api_key=YOUR_API_KEY
Use start_date and end_date (YYYY-MM-DD) to specify your desired date range. The default range is the trailing 365 days. The endpoint returns data sorted chronologically.
To explore the full API reference and available parameters, visit the API documentation.
