One of the great clichés of the dotcom boom was, “information wants to be free.” This conveyed a number of messages simultaneously, among them: that the demise of any for-profit model of information distribution is a matter of historical necessity, a dialectical truth, and; that as a sort of moral axiom anyone who tries to wall up data is in the wrong: is engaged in an activity akin to locking up prisoners of conscience.
You may judge the moral point as you like, but the dialectical-necessity argument seems to have vanished into the wind.
Gennaro Bernile and two associates revisit some of the implications of this cliché in a new paper, Can Information be locked-Up? Informed Trading Ahead of Macro-News Announcements.
Bernile, Assistant Professor of Finance at Singapore Management University, and his colleagues, affiliated with the same institution, begin their paper with the simple fact that many U.S. agencies routinely pre-release economic data to news agencies under promise of embargo. The deal is simple: “The official announcement date is June 1st. You, Clark, may have this pre-release copy on May 29th, which will give you a couple of days to work it into a fuller story that can be released when the embargo is lifted. But you have to promise not to tell a soul before that time, and serious consequences may befall if that promise is broken.”
Leaks to Traders
Are such promises in fact kept or does the information (wanting to be free and all) routinely leak out to traders who can profit from it? [Of course, at AllAboutAlpha embargo promises are always scrupulously honored.]
Bernile et al. conclude that sometimes information leaks out, and that sometimes it doesn’t. There exists “robust evidence of informed trading during lockup periods ahead of the Federal Open Market Committee … monetary policy announcements” but the same authors found “no evidence of informed trading ahead of nonfarm payroll., CPI, and GDP data releases by other government agencies.”
So the answer to the question, “can information be locked up?” would seem to be “yes,” although sometimes what can be accomplished is not in fact accomplished.
Most of the paper is devoted to the FOMC lock-ups, and their failure. One of the types of data at stake, then, is the Federal funds target rate, conventionally released during the trading date, most often at 2:15.
What kind of investor can best capitalize on, say, having a chatterbox reporter friend who spills the news that FOMC seeks to manage? The kind who trades in instruments with high systematic, but low idiosyncratic, risk exposure. After all, FOMC data yields few opportunities to get a jump on other market participants in one’s understanding of the specific characteristics of a security or its issuer. But it might inform some quicker than others on the possibility that interests rates are due for a drop, which may be very good news for the leverage-dependent.
The Bottom Line
Bernile and his colleagues paid especial attention to the E-mini S&P 500 futures in part because it meets this criterion of systematic risk exposure. They also examined E-mini Nasdaq futures, the SPDR S&P 500 ETF, and the Powershares QQQ ETF that tracks the Nasdaq 100 index, although the authors refer to these as “supplemental tests.”
Their methods required that they distinguish between “surprise” and “non-surprise” announcements from the FOMC. Presumably, the slippage of data through the metaphorical walls of a news blockage will matter more when the data in question defies conventional wisdom in the market, than it will when the data is merely a confirmation of the relevant conventions.
The bottom line number is that for E-mini S&P 500 futures, the abnormal price run-up during embargo periods prior to FOMC surprise announcements was 20.5 basis points higher than it was for non-surprise announcements. Price run-ups and more indicative here than price drops. There is an asymmetry due to the constraints on short sales, and perhaps due to the use of limit orders by informed traders/tippees ahead of bad news.
Another key number: abnormal order imbalances. These are 8.4% to 9.5% higher for FOMC surprise announcements than for nonsurprise announcements.
Finally, examination of the securities used as supplemental tests show “similar patterns.”