Economic Surprise Dashboard
Enter the most recent actual print and consensus forecast for each indicator. The tool calculates normalized surprise scores and a composite CESI-style index.
Direct Answer
An economic surprise dashboard converts each economic release's actual print versus its consensus forecast into a normalized surprise score, then combines those scores into a composite CESI-style index. A positive composite reading means data has been beating expectations, which tends to pressure yields higher and support the dollar, while a negative reading signals data missing forecasts. Enter the latest actual and consensus figures for each indicator below to calculate individual surprise scores and the composite index.
| Indicator | Actual | Consensus | Surprise (z) | Signal | Strength |
|---|
Macro Interpretation
Asset Class Implications
This tool uses illustrative historical standard deviations for normalization. Real-time CESI scores (Bloomberg, Citigroup) use proprietary rolling windows and larger series sets. Results are educational only and do not constitute investment advice.
Assumptions and Limitations
- Fixed, illustrative standard deviations: Each indicator's surprise score is normalized by a hard-coded standard deviation (e.g. ±65k for Non-Farm Payrolls, ±0.13% for CPI) drawn from approximate historical release-to-release variability, not a rolling, indicator-specific volatility window recalculated from live data.
- User-entered actual and consensus figures: The tool performs no live data fetch, it computes entirely from the actual and consensus values you type in for each indicator, so results are only as current and accurate as the numbers you enter.
- Simple equal-weighted composite: The composite surprise index averages the six indicators' z-scores with equal weight and no cross-indicator correlation adjustment; proprietary indexes such as Citigroup's Economic Surprise Index use different indicator sets, weighting schemes, and rolling time windows.
- Directional interpretation is illustrative, not predictive: The macro and asset-class narrative text describes typical historical relationships between surprise direction and asset classes. It is not a forecast, backtested signal, or trading recommendation for any specific position.
Frequently Asked Questions
What does a surprise score of one standard deviation mean?
It means the release came in one typical miss away from consensus, where typical is measured by how much that particular series has historically deviated from forecasts. The point of the conversion is comparability: a payrolls figure and an inflation figure are measured in different units and have different normal error sizes, so raw gaps cannot be added together. Expressed in standard deviations they can be, which is what allows a composite score to exist at all.
What happens if a release's typical volatility has changed since the normalizing figures were set?
The scores drift systematically. A series that has become more volatile than the fixed divisor assumes will generate scores that look large every month, and one that has become steadier will look permanently quiet. The direction of each surprise stays correct; the magnitude stops being comparable across series. Reading the composite against its own recent range, rather than against a fixed threshold, limits how much that drift affects the interpretation.
Where do consensus figures come from?
They are compiled by data providers and news organizations from surveys of economists, and they are published ahead of each release. Different providers survey different panels and can publish slightly different medians, so the same release can be a small beat against one consensus and a small miss against another. The figure a market actually traded against is the one most widely circulated, which is why the source matters more for small gaps than for large ones.
Why does a surprise composite center on zero rather than tracking the level of the economy?
Because it measures forecast error, not activity. Forecasters revise their expectations toward whatever the data has been doing, so a run of beats gets incorporated and the next forecast is harder to beat. That feedback pulls the composite back toward zero regardless of whether the economy is strong or weak. A persistently strong economy that everyone correctly expects produces a composite near zero, which is the behavior the construction is designed to produce.
Does a positive composite reading mean the economy is strong?
No. It means data has been arriving better than forecasters expected, which is a statement about expectations as much as about conditions. A weak economy can produce a strongly positive composite if forecasts had been even weaker, and a healthy economy can produce a negative one if optimism had run ahead of it. The composite is useful for anticipating how markets respond to the next release, not for judging the state of activity.
How long does an elevated composite reading usually persist?
Not indefinitely, because the mechanism that produces it is self-correcting. Forecasters observe the pattern of beats and adjust, which raises the bar for the next release and pulls the composite back toward zero over subsequent releases. That mean reversion is a structural property of the construction rather than an empirical tendency, which is why extreme readings in either direction have historically been followed by moves back toward the middle.
Do jobless claims use the same sign convention as payrolls?
They need to be inverted before they mean the same thing. A payrolls figure above consensus is a positive growth surprise, while a claims figure above consensus means more people filed for unemployment benefits, which is a negative growth surprise. Entering a claims beat as a positive score would push the composite in the wrong direction. Any series where a higher number is worse for growth, including inventories in some contexts, needs the same treatment.
How should a revision to a prior month be handled?
Separately from the current surprise. The score compares the newly released figure against the consensus for that period, and a revision changes a period that was already scored. Folding a revision into the current month's surprise double counts it and attributes old information to a new release. Markets do react to large revisions, so they are worth noting alongside the score, but they belong outside the surprise calculation itself.
What do the asset class implications assume?
They assume the reaction pattern that has been typical when growth data surprises: firmer data pushing yields higher and supporting the currency, softer data doing the reverse. That mapping is regime dependent. In periods when inflation is the dominant concern, a strong growth surprise can be read as hawkish and pressure equities, while in a growth scare the same surprise can be welcomed. The implications are a starting interpretation rather than a fixed rule.