New$10 Sign Up Bonus & $30 Free Trial Reward
Start challenge

Economic surprise

Also called data surprise, news surprise

The difference between a released economic figure and the consensus forecast for it. Research on US releases finds that exchange rates and bond yields jump in response to these surprises.

Reading
2 min
Sources
4ResearchCentral bank
Updated
On this page· 5 sections
  1. How it works
  2. Example
  3. Watch out
  4. Sources
  5. Related terms

A strong jobs number is old news if everyone expected it. Research on market reactions looks at the part nobody expected: the released value minus the market’s expectation, called the surprise or “news”[1][2].

In 5-minute exchange rate data from 1992 to 1998, Andersen, Bollerslev, Diebold and Vega found that these surprises make the dollar jump, and that bad news had a greater impact than good news[1].

How it works

For one release, the surprise is the actual figure minus the median forecast from a survey of market participants[2][3]. Because units differ from release to release, the surprise is usually standardized: divided by the standard deviation of that release’s surprises over the sample[1][4]. A value of +1 then means a typical-sized surprise above the forecast, whatever the units[1].

A sign convention makes surprises comparable. A New York Fed study counts a surprise as positive when it shows the US economy stronger than expected, such as higher payrolls or retail sales, or lower unemployment; for inflation, higher than expected counts as positive[3].

A surprise index adds many releases into one number. Two methods are documented:

  • The Citi Economic Surprise Index, as the 2013 New York Fed study describes it, aggregates US data surprises against the Bloomberg survey median over the past three months. Older releases get declining weights, and each indicator’s weight comes from how much a one-standard-deviation surprise in it has moved exchange rates. A positive reading means releases have on balance come in above consensus[3].
  • Scotti’s index at the Federal Reserve Board takes a weighted average of standardized surprises in real activity data: for the United States, GDP, industrial production, payrolls, retail sales, the ISM manufacturing index and personal income. The weights come from each indicator’s role in a business conditions index, fade as a release ages and drop to zero for months with no data. Each surprise is signed so that a positive value means good news for the economy, which flips the sign of the unemployment rate where it is used[4].

Example

Say three releases come out in one week, each with its own record of past surprises.

  1. Payrolls: 180,000 against a consensus of 150,000. Past surprises have a standard deviation of 60,000. Standardized: 30,000 ÷ 60,000 = +0.5.
  2. Unemployment rate: 4.3% against 4.1%. Standard deviation 0.1 point. That is 0.2 ÷ 0.1 = 2.0, and higher unemployment counts as negative: −2.0.
  3. Retail sales: +0.6% against +0.4%. Standard deviation 0.4 point: +0.5.
  4. Equal weights: (0.5 − 2.0 + 0.5) ÷ 3 = −0.33.
  5. Weights of 0.4, 0.4 and 0.2: 0.2 − 0.8 + 0.1 = −0.5.

The same three numbers give a different index depending on the weights.

Watch out

An index measures data against expectations, not the state of the economy. Scotti reads a positive value as market participants having been more pessimistic than the data turned out[4].

The reaction is not fixed. In data from 2000 to 2011, government bond yields rose on good news, but less so when risk was elevated[3].

The consensus used has changed over time. Studies have drawn it from Money Market Services, Action Economics and Bloomberg[2][3][4].

Sources

  1. ResearchT. Andersen, T. Bollerslev, F. Diebold, C. Vega / NBERMicro Effects of Macro Announcements: Real-Time Price Discovery in Foreign Exchange (NBER WP 8959; American Economic Review 93(1), 2003)Checked November 9, 2026
  2. Central bankBoard of Governors of the Federal Reserve SystemThe Excess Sensitivity of Long-Term Interest Rates: Evidence and Implications for Macroeconomic Models (FEDS 2003-50, R. Gürkaynak, B. Sack, E. Swanson)Checked November 9, 2026
  3. Central bankFederal Reserve Bank of New YorkTime Variation in Asset Price Responses to Macro Announcements (Staff Report No. 626, 2013, L. S. Goldberg and C. Grisse)Checked November 9, 2026
  4. Central bankBoard of Governors of the Federal Reserve SystemSurprise and Uncertainty Indexes: Real-Time Aggregation of Real-Activity Macro Surprises (IFDP 1093, 2013, C. Scotti)Checked November 9, 2026
All terms in Tronexopedia

Your challenge starts here

Start your challengeFree TrialNew
Registration bonus

Get $10 free in your Wallet when you sign up

Register and use your $10 Wallet credit to reduce the price of your first challenge.

The $10 bonus is credited to your personal Wallet and can only be used to reduce the price of a challenge purchase. Credits cannot be withdrawn or used for payouts.

Start now & get $10