Who Acquires Information in Dealer Markets?

A market maker is a broker-dealer who at all times undertakes to buy or sell assets at specified prices. Market makers thus play a key role in financial markets. The workhorse models in financial theory assume that speculators are better informed than market makers, and that market makers make up what they lose to speculators by trading occasionally with noise traders. But, if information can be acquired at a cost by speculators, surely it can also be acquired (at the same cost) by market makers. Therefore, the dichotomy between informed speculators and uninformed market does not seem like a foregone conclusion. Furthermore, empirical evidence suggests that market makers may be informed above and beyond what speculators know, and that price discovery – the process by which prices move toward fundamental asset values – is sometimes driven by market makers rather than speculators. Surprisingly, though, this issue has never been addressed by the theoretical literature. Our paper aims to fill this gap, and provides important insights for regulators and policy makers. In particular, we show in the paper that changing the cost of information may have non-monotonic and, sometimes, surprising effects. Any regulation designed to affect incentives to acquire information must take these effects into account.

We first show that indeed the ‘standard information structure’, in which speculators are informed and market makers not, arises when information cost is in an intermediary range. Speculators acquire information because they can always make a profit when they are better informed, but market makers are more constrained – for instance, if they learn that the asset is undervalued, they can only profit from this knowledge if they can buy the asset from a noise trader, and this opportunity does not always present itself.

However, two things occur as the cost of acquiring information cost decreases. First, cheaper information makes it more attractive for market makers to become informed. But second, and less obvious, the more informed trading there is, the more attractive it is for market makers to become informed. This observation is unusual, since information is normally a substitute good, yet in our model, it is a complementary good. In particular, information acquisition by speculators leads uninformed market makers to increase their spreads (the difference between the buying and selling price) to protect themselves against adverse selection (losses incurred from trading with better informed speculators). But this allows informed market makers to also set a larger spread, thereby making more profit. As a consequence of these two effects, if the information cost is low enough, market makers will start acquiring information. In fact, when the information cost is sufficiently low, the market makers will completely crowd out the speculators, and only noise traders will be left trading.

We then explore liquidity and price discovery. An (inverse) measure of liquidity in the market is the spread between buy and sell prices. We show that this is non-monotonic in the information cost, and is maximized for some intermediary cost level. The reason is that adverse selection is at its highest when both speculators and market makers acquire information with some probability, since in this case a market maker who remains uninformed faces adverse selection both from speculators and from fellow market makers, who may be informed. Turning now to price discovery, we note that this is driven by two factors: informed trading (whereby the direction of the trade contains information) and informed market making (whereby the quoted price contains information). Price discovery is (almost) perfect for low information cost, and poor for high information cost. But, in the region where spreads rise due to double adverse selection, lower information cost may lead to worse price discovery, since high spreads imply that the realized prices can be quite far from the actual value of the asset.

Jesper Rudiger is an Assistant Professor in the Department of Economics at the University of Copenhagen.

Adrien Vigier is an Associate Professor in the Department of Economics at the University of Oxford.