The ECB's research department says US data leaked

The ECB has been researching US economic data and it doesn't like what it has found.

In a report that could shock market participants and regulators, they note advance price moves on embargoed private statistics.

The report says they see 'price drift' in front of some US macroeconomic news, with prices moving in the 'correct' direction about 30 minutes before the releases.

"Seven out of 21 market-moving announcements show evidence of substantial informed trading before the official release time," the researchers say. "The pre-announcement price drift accounts on average for about half of the total price adjustment."

They studied S&P E-Minis and US T-note futures from Jan 2008 to March 2014.

They say further study of 'informed' trades is needed and the report is careful to note that it represents the view of the authors, not the ECB.

The headlines are just crossing. They specifically notice 'private' data so that may rule out some of the biggest releases but not important ones like ISM non-manufacturing or consumer confidence.

"The third category, previously not documented in academic literature, involves an unusual pre-release procedure used in three announcements: Instead of being pre-released in lock-up rooms, these announcements are electronically transmitted to journalists who are asked not to share the information with others. Three announcements in this category are among the seven announcements with strong drift," they write.

Here is the list of data that is most-likely leaked:

  1. ISM non-manufacturing
  2. Pending home salse
  3. ISM manufacturing
  4. Existing home sales
  5. Consumer confidence from the Conference board
  6. Industrial production
  7. The second reading on GDP

Others, including retail sales, the CPI, advance GDP and initial jobless claims also show some evidence.

The largest offender is the ISM non-manufacturing index with 69% of the move in S&P futures happening before the release.

The researchers note that the last time they can find evidence of a government employee getting fired for data leakage was 1986.

The choice of instruments is strange. If I had advance data, the first place I would trade is the foreign exchange market. The volume is easily swallowed up and there is nearly no paper trail.

Here is the paper (pdf).

Newer forecasts don't beat older ones.

Here's a footnote in the research that's as interesting as the main article.

"Since some individual forecasters submit their forecasts days before the releases as described in Section 3 and Bloomberg equal-weights the forecasts, we also test whether more up-to-date forecasts are better predictors of the surprise. The results (available upon request) show that removing stale forecasts does not improve forecasts of the surprise."

I have often been under the impression that the most up-to-date forecasts would be best. I'm still skeptical of the research though because they're look at so many indicators. I noted earlier today that the Chicago PMI (released Friday) was weak and pointed to downside in the ISM manufacturing survey. The data turned out to be soft and I don't think that's a coincidence. There is a strong correlation in the surveys and economists' forecasts aren't all updated.

Takeaways

Existing and pending home sales are not really market movers so those don't bother me. The three that stand out are ISM manufacturing, ISM non-manufacturing and industrial production. The ISM needs to immediately implement much better safeguards. The IP report is from the Fed, and there is absolutely no excuse for them not to have a lockup for the announcement. The Conference Board recently tightened up consumer confidence so they've taken action.