There’s an intuitive logic which states that the more liquidity providers (LPs) that a client puts in their aggregator, the better prices they should get. After all, increased competition should cause LPs to tighten their prices in order to win the trade.
However, as Roel Oomen, managing director, electronic FX spot trading at Deutsche Bank, explains, this logic only holds up in a static environment, which the FX market most certainly is not. The reality is that LPs alter their spreads depending on how they perceive the liquidity environment to be at any given point in time.
“If you think about it from a liquidity provider’s perspective, the client executes on best price so the only way for me to win a deal in this aggregator is if my price is more competitive than, let’s say, my 19 competitors in the aggregator. Now, it might be the case that we show a smarter and more accurate price than any of the 19 people in the aggregator, but what’s more likely is in fact our price is inaccurate, our price is too aggressive and so when we win the deal it’s a signal to us that we’ve been mispricing the market. And this is referred to as the “winner’s curse” in game theory literature,” he says.
Oomen continues: “Initially you’re happy because you own the deal, and you’re out there to get flow, but you also know that you’re probably showing the wrong price and you suffer adverse selection subsequently. Because of this mechanism, liquidity providers adjust their spread or maybe their last look criteria and reject more trades, and so if you look at the net-net impact of adding liquidity providers to your aggregator over time as all these adjustments take place there’s no guarantee that you end up with lower transaction costs in your aggregator.”
Of course, there’s no guarantee that transaction costs will get worse either. But Oomen points out that there is a practical issue here, which is that there are fixed costs associated with having an LP in the aggregator - the client might need cross connects, they have to spend more time on relationship management, etc. And from the LP’s perspective, the incentive to continue providing good liquidity to a client’s aggregator is diminished if they know that they are one of 50 LPs in there instead of one of four valued LPs.
Oomen adds: “There’s a second, perhaps more important mechanism at work here. When you start increasing the number of liquidity providers in your aggregator, the risk that you end up with a mixture of liquidity providers that follow different risk management styles increases, ie, it is more likely that you end up with a mix of internalisers and externalisers in your liquidity pool. In many scenarios if you mix liquidity providers with different risk management styles you can easily end up in a ‘prisoner’s dilemma’, meaning that actually everyone is worse off and transactions costs are maximised rather than minimised.”
He concludes: “Rather than just taking every liquidity provider you can find on the Street, choose them very carefully to make sure that their risk management style and the liquidity they offer is consistent with your execution objectives.”
The full video can be viewed here: