Measuring the Information Content of Stock Trades
Read the paperopens doi.org in a new tab
What they found
Hasbrouck developed a way to separate the permanent price impact of a trade (the part that reflects information) from the temporary impact (inventory effects and bid-ask bounce that reverse). Using a vector autoregression of quote changes and signed trades on NYSE stocks, he found that trades carry substantial information, that the price response to a trade unfolds over several subsequent trades rather than instantly, and that the information content of trades is larger for smaller firms and larger trades.
What you can use
- Price impact has two parts: the permanent part is the market learning from your trade; the temporary part is what you get back if you are patient.
- The full impact of a trade takes several subsequent trades to show up, so the immediate price move understates it.
- Trades in small stocks are more informative and therefore move prices more, which is why hiding size matters there.
Caveats
Requires signing trades (buyer- versus seller-initiated), which is imperfect. Based on 1989 NYSE data with a specialist system. Technical econometrics.
Tags: microstructure, price-impact, information, econometrics
Summaries are our own reading of the paper, not the authors' words. Educational only, not advice. Discuss it in Book Club.