Views on HFT in Europe, ESMA’s Role in MiFID Shared at E-Finance Lab Event
In the run-up to the end of the Markets in Financial Instruments Directive (MiFID) II Level 2 consultation in early March, I thought Market Integrity Insights’ readers would be interested in the latest perspectives on “Liquidity, Transparency, and Electronic Trading in Europe,” the theme of the annual E-Finance Lab Spring Conference I recently attended at Goethe University Frankfurt.
Rodrigo Buenaventura, head of markets division at the European Securities and Markets Authority (ESMA), gave a presentation on “Liquidity and Transparency — What Does the Regulator Want?,” which focused on the background to MiFID II. Buenaventura explained that ESMA’s role in this key piece of markets legislation is to provide technical standards and technical advice, and guidelines to the European Commission regarding the implementation of the legislation — the so-called “level 2” measures. He emphasized that a top goal of MiFID II was to improve transparency and extend the principles in MiFID I to nonequity instruments. (A full account of MiFID II is provided in our blog post and policy brief.)
Ever-Faster Exchanges — Should We Even Care?
Two perspectives on high-frequency trading (HFT) were offered at the conference — an industry perspective by Remco Lenterman, managing director at IMC Financial Markets and chairman of FIA European Principal Traders Association, and an academic perspective by Albert Menkveld, professor of finance at VU University Amsterdam.
My main takeaway from Menkveld’s presentation was that the marginal decrease in latency on exchanges (i.e., ever-faster exchanges) is now actually detrimental to certain HFTs. The key to understanding why is to realize that there are two types of HFTs: the first group of HFTs he described as high-frequency bandits (HFBs); the second group is comprised of more passive HFT players employing market-making strategies. (This is a variation on the story I have presented in a previous blog post.)
Since traditional investors are already typically too slow to reach a quoted price on a lit market first, and can be “front-run” by predatory HFT strategies, any further decrease in exchange latency will not help traditional investors outwit HFTs. These latency improvements will only further increase the abilities of predatory HFTs (aka HFBs) to get to quotes first, making the toxicity of lit markets for passive HFTs worse, not better, by further improvements in latency. Menkveld concluded that lowering exchange latency could reduce liquidity by increasing adverse selection.
Lenterman’s presentation was a sort of rebuttal that began by arguing the impossibility of defining HFT. His argument was that the definition is so broad that anyone could phrase it in such a way as to support any position on the issue. The problem is that the term HFT describes a frequency, not an activity. He flatly refuted the notion of phantom liquidity and argued that the relatively minute profits made by the HFT industry suggest the entire issue has taken an importance far greater than its economic impact warrants. Lenterman argued that what is being proposed in MiFID II and similar legislation elsewhere is another radical restructuring of a market that has yielded historically low trading costs in favor of an alternative structure that will likely have other unintended consequences.
The panel session after these two presentations was lively, with both the academic and industry contingents presenting the well-known and favoured arguments of each side. The academic perspective focused on the social uselessness of ever-decreasing latencies, and the dangers of completely electronic and homogenous trading, while the industry perspective countered with the observed historically low transaction costs across all markets. In response to my question about the dangers of homogenous liquidity provision in times of market stress, Menkveld agreed that this was a concern while the other panelists said it was an issue ameliorated by the presence of circuit breakers in many markets that would, by definition, not allow runaway scenarios to develop.
As ever, the topic of HFT provides no easy answers, in part due to a lack of suitably detailed data — an issue lamented by Menkveld. The level 2 MiFID II paper describes a process for enforcing market-making obligations on certain HFT participants as well as extensive algorithm certification by investment firms and trading venues. With implementation scheduled for 2017, and no obvious sources of highly detailed data in the interim, it is likely that the question of whether HFT was good or bad will be answered only when it is curtailed by MiFID II. Perhaps we will miss it when it’s gone.
Presentation slides from the E-Finance Lab conference, held annually since 2003 to “develop industrial methods for the impending change process of the financial services industry,” are available here.
If you liked this post, consider subscribing to Market Integrity Insights.
Image credit: iStockphoto.com/Muhla1