"information of many macroeconomic announcements is known by some market participants in advance"
- European Central Bank
Remember when it was merely another conspiracy theory that "some" traders "trade" (with zero risk) with the benefit of leaked material non-public information? As of moments ago it is merely the latest conspiracy fact, confirmed by none other than the ECB, which earlier today published a research paper which finds substantial "informed" trading before the official release in seven of 21 market-moving U.S. macroeconomic announcements.
The paper has studied trading patterns from Jan. 2008 to March 2014 and finds that "prices begin to move in the ‘correct’ direction about 30 minutes before the release time. The pre-announcement price drift accounts on average for about half of the total price adjustment."
Translation: chronic insider trading, which until recently was an allegation that the serious outlets would relentless mock as another conspiracy by tinfoil hat fringe websites.
The paper also finds "strong evidence of pre-announcement drift" in ahead of releases such as the Conference Board's consumer confidence index; NAR existing home and pending home sales; preliminary GDP; Federal Reserve’s industrial production (yes, even Fed data has been leaked repeatedly); ISM manufacturing and non- manufacturing indexes, and countless other examples.
While the ECB does not blame the entire "drift" on leaks of insider information, saying instead that "the evidence suggests that the pre-announcement drift likely comes from a combination of information leakage and superior forecasting" it is clear what the real culprit is for outperformance ahead of major data release, and - spoiler alert - it isn't "superior forecasting."
The paper also says that according to back-of-the-envelope calculations, since 2008 pre-release trading in S&P E-mini futures market profits amount to about $20 million per year. Curiously, the ECB did not find evidence of "pre-announcement drift" in non-farm employment data. We doubt that especially since just this website has repeatedly caught BLS data being leaked in advance, an example of which we reported several years ago in "Bureau Of Labor Statistics Caught Red Handed Leaking Confidential Employment Data."
From the paper's non-technical summary:
Macroeconomic indicators play an important role in business cycle forecasting and are closely watched by financial markets. Some of these indicators appear to influence financial market prices even ahead of their social release time. This paper examines the prevalence of pre-announcement price drift in U.S. stock and bond markets and looks for possible explanations.
We study the impact of announcements on second-by-second E-mini S&P 500 stock index and 10-year Treasury note futures from January 2008 to March 2014. The study is based on 21 market-moving announcements among a sample of 30 U.S. macroeconomic announcements. Eleven out of these 21 announcements exhibit some pre-announcement price drift in the \correct" direction, i.e., in the direction of the price change consistent with the announcement surprise. For seven of these announcements the drift is substantial. Prices start to move about 30 minutes before the ocial release time, and this pre-announcement price move accounts on average for about a half of the total price adjustment.
These facts are uncovered by an outlier-robust procedure (MM weighted least squares), but are similarly striking in cumulative average return graphs and order ow imbalances.
The paper shows that these results are robust to controlling for, among others, outliers, data snooping, nearby announcements and the choice of the event window length.
Extending the sample period back to 2003 with minute-by-minute data reveals both a higher announcement impact and a stronger pre-announcement drift since 2008, especially in the S&P E-mini futures market. Based on a back-of-the-envelope calculation, we estimate that since 2008 in the S&P E-mini futures market alone the prots associated with trading prior to the social announcement release time have amounted to about 20 million USD per year.
The late start of pre-release price drift, which becomes signicant only about 30 minutes before the social release time, reveals an interesting property of prevalent trading strategies. Assuming that informed traders possess their informational advantage already more than 30 minutes ahead of the release, the question arises why they wait with trading on their knowledge until shortly before the release time. A possible explanation is that trading close to the release time minimizes the exposure to other risks that are unrelated to macroeconomic announcements.
The difficulty of identifying the causes of pre-announcement drift stems from the relatively small number of announcements that actually move financial markets. Nevertheless, we find that an implementation of strict release procedures makes pre-release drift less likely. This applies in particular to data released under the Principal Federal Economic Indicator (PFEI) guidelines, which impose strict security procedures. There is no evidence that modifying the calculation of market expectations, e.g., a focus on the most recent survey responses, helps in predicting the commonly used announcement surprise.
Public information, such as internet activity data, predicts the surprise in a few cases where the public information closely corresponds to the forecasting target. Analogously, improvements in data processing render privately collecting large amounts of comparable information feasible, which can be used for generating proprietary forecasts ahead of time. This early information -leaked or self-calculated- does not need to be precise in order to a have a large price impact. Under Bayesian learning, even if the information available before the ocial release is noisy, it can have a large price impact because of its timing. For a Bayesian learner, early availability makes up for less precision and a potentially smaller surprise. Thus, the incentives for privately collecting information and for leakage are high.
The main policy implications of this paper are twofold. First, the total impact of macroeconomic news is larger than measured in most event studies, which ignore the pre-release price drift. Therefore, the total impact of macroeconomic news on financial markets is larger, and financial markets are linked more tightly to the real economy than usually found. Second, information of many macroeconomic announcements is known by some market participants in advance. To ensure fairness in financial markets, strict release procedures need to be implemented for all market-moving announcements including announcements originating in the private sector.
Translation: the entire market is rigged.
For those who have been accused for years of being insane in claiming precisely what the ECB just "confirmed", you can read the full paper here.