Price Discovery Across Venues: Information Share Basics
When the same asset trades in many places, which venue moves first? An introduction to measuring where price discovery actually happens.
Bitcoin trades on dozens of spot and derivatives venues simultaneously. Arbitrage keeps their prices close, so in a sense they share a single efficient price. But when news arrives, some venues incorporate it faster than others. Measuring which venues lead is the study of price discovery.
The standard approaches
The two most widely used measures come from the market microstructure literature. Hasbrouck’s information share attributes the variance of innovations in the common efficient price to each venue. The Gonzalo–Granger component share instead measures each venue’s weight in constructing the common price. Both start from the observation that venue prices are cointegrated — they cannot drift apart indefinitely.
- A venue with a high share leads: its price changes tend to be permanent and others follow.
- A venue with a low share follows: its price changes tend to be corrections toward the leader.
- Spot and perpetual futures markets frequently differ in their contribution, and the balance can shift over time.
Why data quality decides the answer
Lead-lag measurement at high frequency is extremely sensitive to timestamps. If one venue’s data carries more delivery delay than another’s, it will appear to lag even if it moved first. Consistent use of exchange timestamps, explicit treatment of clock offset, and sampling intervals appropriate to the precision of the data are prerequisites for any credible result.
This publication is provided for informational purposes only and does not constitute investment, legal, or tax advice, or an offer or solicitation to buy or sell any asset. Live figures are computed from third-party public market data and may be delayed, incomplete, or inaccurate.
