Baby steps in the dark

The popularity of dark pools among institutional investors speaks volumes for their price improvement opportunities, but differences in execution performance are hard to tease out.

Today's buy-side trader is more focused than ever on using data to demonstrate value to the investment process through superior execution performance. How much harder is that when executing in a dark pool compared to a lit venue?

Diversity among dark pools is probably one of the biggest obstacles to proving your dark liquidity strategy is effective.

Dealers at institutional investment firms have enough trouble normalising and comparing execution data delivered end of day by their brokers, following trades conducted on exchanges and lit multilateral trading venues (MTFs) that run fairly standardised matching process on central limit order books. A trade in a large European name might be split across three or four brokers and a similar number of lit trading venues. You might, for example, expect to get fills for an order for a CAC-40 stock from Euronext Paris, BATS Chi-X Europe, Turquoise and Equiduct. A lack of uniformity in post-trade reporting across venues and brokers can mask a host of small differences in performance that make it hard to be sure which channel was the most effective.

In the dark, the picture is more blurred yet. Dark MTFs must follow rules as laid down in MiFID and interpreted by the European Securities and Markets Authority (ESMA) but there is still much variety in process here, not least because of the different business models and client profiles of operators such as the London Stock Exchange, Goldman Sachs and Liquidnet, to name three at random. Beyond these dark MTFs, broker crossing networks (BCNs) - which account for roughly the same amount of European dark trading volumes as MTFs - provide both more opportunity, but more complexity too.

Like dark pools, these problems have been with us for quite a while now. What's been done so far to help the buy-side trading desk compare apples with apples more easily?

Quite a bit actually, but it tends to be the firms with the greatest resources or influence that stand the best chance of achieving the most meaningful analysis. In 2011, a number of global brokers signed up to the global 'Open TCA' initiative spearheaded by vendor and transaction cost analysis provider TradingScreen to standardise the execution information they provide to clients. Trading venues and brokers have worked closely with regulators in Europe to standardise the way they categorise different types of trades. Brokers and vendors have worked to provide clients with greater transparency of the venue-specific information carried by FIX messages, partly in response to the feedback provided to their execution consulting teams from clients that want to understand more about how their orders are routed.

In the dark trading space specifically, analytics specialists such as London-based Liquidmetrix have built rapidly growing businesses on the provision of broker-independent reports to the buy-side on the quality of their execution performance across non-lit venues. Meanwhile in the US, Barclays has worked with a number of buy-side clients to develop a standardised methodology that it believes would provide a universal basis for comparison of dark executions if adopted by all operators of dark pools.

But most buy-siders are still in the dark in the dark?

Don't make me repeat myself. Only the largest asset management firms are set up to handle large volumes of data, which prolongs dependence on third parties for effective analysis and interpretation of post-trade execution data. Inevitably, these third parties can face conflict of interest when it comes to providing data which will be used to judge their service quality.

This doesn't only affect the trading desk. A recent paper published by investment research group Investit and buy-side technology provider SimCorp found that 60% of firms are concerned about their ability to support data management growth. The report calls for institutional investors to adhere to data governance principles to improve the quality and control of data on an enterprise-wide basis. Such an approach may well help dealing desks to develop a more robust framework for analysis of post-trade data from lit and dark venues, but it's hard to escape the conclusion that the advantage will lie with the larger asset managers.

To vote in this month's poll, click here.