Why Markets Can Move the Other Way When Economic Data Beat Expectations
Read economic surprises by separating actual, expected, previous, and revised figures. Use a hypothetical comparison to examine forecast distributions, units, and economic meaning together.
An economic data surprise is the difference between the released figure and expectations formed before publication. Beating expectations means the number is higher; it does not mean good news for every asset. The headline and market reaction can seem inconsistent when the timing and distribution of forecasts, revisions to previous figures, and underlying components change together.
Forecasts are different data from official observations
A statistical agency's release results from specified survey and estimation methods, while market expectations are forecasts from separate surveys or models. Economic calendars can differ in participating forecasters, survey cutoffs, and use of means or medians. When comparing actual and expected figures, therefore record the source and the time the forecast was saved. A screen updated after the release can make it difficult to reconstruct what the market originally expected. A wide spread of high and low forecasts does not imply the same degree of surprise as a situation where everyone expected similar results.
Interpret the size and direction of the gap for the specific indicator
Higher growth or employment figures can be positive for economic activity, but higher inflation or unemployment means something different. Federal Reserve research also distinguishes the signs of upside and downside surprises according to their relationship with the direction of the economy. Investors care not just about the raw difference but about how much policy expectations and earnings outlooks change. Surprises across different indicators can be compared by scaling them by the size of past forecast errors, but this requires a sample period and stationarity assumptions. Listing absolute differences without considering units distorts comparisons.
Revisions change the historical starting point
Statistics revised as additional survey information arrives, such as monthly employment, change previous months' levels as well as reporting the latest result. Even if the latest release beats expectations, substantial downward revisions to the past can weaken the cumulative picture. However, do not call the simple sum of this month's surprise and past revisions an official surprise indicator: the comparisons use different points in time and reference bases. Recording the actual figure, prerelease expectation, previous release, revised previous figure, and components separately lets you investigate why the reaction differed.
Worked example
Hypothetical example: If payroll growth is expected to be 150,000 and comes in at 180,000, this month's result is 30,000 above expectations. Suppose the previous month's increase is simultaneously revised down by 60,000, from 200,000 to 140,000. The expected two-month total based on the previous release was 350,000, while the new two-month total is 320,000, leaving the cumulative picture 30,000 lower. This comparison illustrates the revision effect; it is not a statistical agency's official surprise formula.
What to check
- Record the forecast provider and when the forecast was saved before the release.
- Align the periods, seasonal adjustment, and units of actual and expected figures.
- Do not call an upside surprise good news for every asset.
- Separate this month's gap from the impact of past revisions.
Sources and primary materials
This is educational material about research methods. For an actual company decision, verify the filings and primary sources for the relevant period.