Shhh, don’t tell: the struggle to keep skew under wraps

The ‘six degrees of separation’ theory posits that people are never more than six social connections away from each other – even if they are complete strangers and move in very different circles. The theory has been popularised via the game Six Degrees of Kevin Bacon, in which players are challenged to connect any given actor to the eponymous movie star in six steps or less.

Similar daisy-chain relationships exist in foreign exchange markets – but it can be a costly game to play.

When a dealer is axed to trade a particular side of a currency, it may skew its pricing – that is, move either the bid or the offer closer to the market’s mid – to make it more appealing. This information is potentially valuable to other market participants, allowing them to infer the dealer’s position and trading intent, and some algorithms are designed specifically to exploit it. This has led to the development of ‘skew safe’ pools where banks can offload risk on trading platforms with trusted counterparties.

But the problem persists. In recent years, large liquidity providers (LPs) have set up direct streaming relationships with smaller dealers that can pass along the original pricing to their end-clients. If the original LP offers a skewed priced to a regional bank or a prime-of-prime broker client, with the client then recycling it to its own customer, the skew information can still be broadcast to the wider market.

Another issue is when a counterparty to the original skewing LP uses this pricing as a market data source for its own execution algorithms. The original LP loses control of the information.

Trying to keep skewed prices quiet is like trying to lose the Six Degrees of Kevin Bacon game. You don’t want to be connected back to the prolific actor; you don’t want to be drawn into a network where connections proliferate and you are just a step or two away from hungry, skew-sniffing sharks. So, banks have developed sophisticated tools to try and avoid leaky market nodes – clients that are transmitting information about the bank’s positions to the rest of the market.

Some have even assigned ‘skew leakage’ scores to certain clients to determine how often they can skew a price to them. If the client has a bad score and fails to make improvements, then the LP can decide to stop offering better prices.

However, detecting information leakage is incredibly difficult and the signals used to distinguish where, when and how skewed prices are being leaked are fairly weak. Banks also don’t want to be scared away from showing better prices to their clients because that would ultimately translate into lost business.

It’s possible to imagine a solution to all of this – an industry-wide effort that would knit together LPs, their clients, and trading platforms to ensure skewed prices are seen only by the intended recipient. Possible, but extraordinarily difficult. Until the day arrives when banks feel footloose and free to skew, they will continue playing the Six Degrees game, and hoping that their price and position information can be contained.