Anyone who is close to the inside of a market likely is the first to know about value changes, and trade upon them when profitable. These changes are observed with noise (e.g., Tea’s up!). Profitability, therefore, depends on both the magnitude of these value changes, and the level of noise. In the academic literature, this type of trading has been referred to as “news trading.” News, here, is broadly defined to include press releases, but also, for example, order book updates in highly correlated securities.
News trading creates value, because it makes prices reveal information, right?
Well, maybe not always.
Another important type of informed traders are value traders. These are best thought of as traders who engage in research to learn about the fundamental value of a security. They then trade strategically on this information and, therefore, slowly reduce the gap between value levels and prices. The profit they earn pays for their costly research.
What does this have to do with news trading?
The short answer is:
The next level is the previous level, plus the change in the level.
Therefore, news traders being quick to trade on value changes are eating into the profit of value traders. This could make the latter trade less aggressively on their information, which potentially slows down the revelation of their information into prices.
This interaction of news traders and value traders is at the heart of a new paper I wrote with Lucas Saru. We took the elegant setting of Foucault, Hombert, and Roşu (2016) which is in the tradition of the classic Kyle (1985), and perturbed it to have both types of traders in a single model. The model is solved analytically and the resulting equilibrium dynamics turn out to fit a standard state space model. This enables us to estimate the model parameters and back out equilibrium objects such as the profit of both types of traders.
Deutsche Börse kindly made a proprietary sample available for estimation. The sample runs from 2013 through 2023 and includes all trades in EURO STOXX 50 Index Futures. Importantly, it also contains a principal/agent flag for both the buyer and the seller in the trade. We consider liquidity-taking principal trades to be a proxy for news trades. The idea is that these are likely triggered by signals first seen by exchange members, who are close to the market and trade on it for their own account. Liquidity-taking agency trades are considered a noisy proxy for value trades.
Structural estimation yields the following result:
| News traders | Value traders | |
|---|---|---|
| Gross profit | €670,000 per day | €1,700,000 per day |
| Trading fees | €120,000 per day | €60,000 per day |
| Gross profit net of trading fees | €550,000 per day | €1,640,000 per day |
These numbers show that, net of trading fees, value trading is almost three times as profitable as news trading. The estimation further reveals that the signal that news traders observe, is quite noisy. They learn about a third of the value change after observing the news.
The results further show that the price impact of liquidity demand more than doubles in the second half of the sample. This is due to the news getting noisier and a decline in the overall level of noise trading. Why does noisier news make the market more illiquid? Well, that is the cliff hanger that, hopefully, makes you download the paper and find out.
Please find the paper here.