High Frequency Traders, or HFTs for short, are part of the regulatory storm that has unfolded post the financial crisis. Fuelled by the Flash Crash on the 6 May 2010 in the USA as well as the upcoming revision of MiFID in Europe, opinions about this relatively new class of market participants are being put forward at a staggering pace. But what are the facts we should take into account before we make opinions about how to regulate or consciously not regulate HFTs?
First we should re-establish why HFTs came to prominence in the first place. Second we have to evaluate what their impact on the pre-HFT markets has been. This second question is key before we judge whether they destroy well functioning securities markets or actually render them more efficient. The following ten theses, although not exhaustive, aim to clarify the questions surrounding these two issues. By HFTs we mean professional proprietary traders engaging in high velocity order entry generating a large number of trades on a daily basis. Given the importance of the HFT phenomenon for SIX Swiss Exchange and other market venues in Europe at a crucial time of the MiFID revision, we shall focus on Europe including Switzerland, one of the leading financial markets on the continent.
Why is Europe facing the HFT phenomenon?
1. In Europe, the rise of HFT was fuelled by MiFID
HFTs in Europe rose from virtual non-existence in 2005 to representing an estimated 38% of equities trading volume by 2010. The numbers point to a dramatic rise of HFT activity post 2007, the year in which the European Commission enforced the abandonment of the so-called concentration rule on equities trading. The result of this de-regulation move was to allow equities to be traded not only on their home market(s) of primary listing(s) but on a variety of venues including newly admitted multi-lateral trading facilities.
So how did the de-regulators fuel HFT activity? Very simply, when securities are tradable on several venues at the same time, from a proprietary trading perspective it is nearly impossible to control pricing with a human eye or indeed to process the information in a timely fashion in the human brain. Instead you need an electronic, or computerized, eye in order to manage prices, expected returns and the associated risks. While de-regulation intended to achieve fragmentation of trading, it also created arbitrage opportunities and led to more complex exposure risk management by market makers. As a consequence these are the two core activities of HFTs: arbitrage and market making. No wonder therefore that HFT activity exploded in relative terms as well as in absolute magnitude.
2. The rise of derivatives markets boosts HFT activity
Exchange-traded derivatives (standardized futures and options) and even more so tailor-made warrants and structured products have a long history of centralized trading. However, the financial crisis has also boosted the rise of ETFs and other passive investment products. Each are essentially instruments based on underlying cash market securities. What they have in common is that their on-exchange price formation is to a large degree made possible by market makers which commit to bid-offer prices they “derive” from the underlying securities. As the market makers commit themselves to one market (the derivatives market), any changes in the underlying market (the cash market) leaves them open to a certain “exposure risk” which they intend to minimize. The result is that these market makers will employ all possible and affordable resources to ensure they can minimize this risk while increasing their market making capabilities to more and more securities. This happens to the benefit of a wide variety of market participants, including retail and institutional investors, who invest in these derivatives including ETFs at an increasing rate.
3. Intermediaries such as banks, not just boutiques, also employ HFT
Who are these HFTs? Contrary to some media and political statements, HFT is not solely performed by boutique investment firms. In fact, most of the largest investment banks that are relied upon to provide crucial financing services for our economies are similarly undertaking such activities.
These institutions run large balance sheets (the size of which is open for debate of course) with inherent risks associated to them. It is almost a natural side-effect that HFT is employed as part of their risk management to deal with the proliferation of cash and derivatives markets. This means that the major sell-side intermediaries, alongside specialized firms, have invested substantial amounts in the creation of technology that can deal with the ever rising demands of today’s market making. Not all of these investments are necessarily profitable, because many of them need to be written down over time and the opportunities in deterministic or statistical arbitrage and profitable market making wane as the number of competitors rises.
Have HFTs benefitted or harmed markets with hindsight?
4. Academia is not conclusive about the impact on markets
Sound academic research on HFT and its market impact has only been conducted for about four years. Much of it surrounds the impact on the “healthy” functioning of markets. Healthy markets are normally viewed as those which have low transaction costs and display high liquidity and high transparency. A widely held view is that HFT activity increases competition, hence lowering transaction costs. At the same time it is now common knowledge that fragmentation in Europe has actually decreased transparency, however, this may not be the fault of HFT but rather the effect of de-regulation that was not well thought through.
The focus of academic analysis has therefore been on the impact of HFT on liquidity. But there is the rub. Liquidity is not only defined as the speed of execution (latency) and the bid-offer spread of securities prices, but also by the depth (the available sizes at different and notably the most narrow, bid-offer spreads) and the resiliency of markets (how long does a market need to recover after a large order is executed). The latter is particularly important for large institutional investors. This is where academia is so far inconclusive. Research seems to indicate that HFTs indeed lower latency and spreads and probably also increase resiliency, but they might have a detrimental effect on market depth.
5. HFTs are viewed as speculators but good markets need both investors and speculators
The average retail or institutional investor has an opinion about the absolute value of a security and rarely goes home after market closure with a flat book. HFTs on the other hand only trade on relative values and prefer going to sleep with flat positions to start from scratch the next morning. Does this mean that the two types of actors should not be angling in the same pool? The answer is no. Genuine “investors” and the lesser appreciated species of “speculators” have a long history of co-existence. This was true for open outcry markets as it is valid for electronic markets such as on-exchange ETF markets. At SIX Swiss Exchange, the ETF issuing and trading volume has risen every single year since its establishment one decade ago, regardless of market volatility, in a great symbiosis of investors and speculators or multiple market makers. Better market making has attracted more fundamental investors ensuring that as recently as in 2010 our ETF segment has been the fastest growing one of all major exchanges in Europe.
6. Anticipatory trading by a machine is not different from anticipatory trading by a human
If the presence of speculators and investors is mutually beneficial in a “healthy” market, what if HFT speculators manage to systematically anticipate and trade in front of fundamental investors? The principles of regulators and self-regulating authorities such as the SIX Group is to guarantee market integrity. Front running, which is associated with a fiduciary relationship, by human traders is a well-known criminal offence. Anticipatory trading by HFTs has been well defined by IOSCO as the application of “locked trades” but is potentially less well identified to the present day. While HFTs today are a main body of participants putting the efficient capital market hypothesis into action, namely that prices should at all times reflect publicly available information, this does not give them carte blanche to pick up on an order entry of a third party to step in and trade in front of them. Here regulators need to be in a position to monitor and enforce potential application of “locked trades”. Our experience at SIX Swiss Exchange is that this is not a core activity of HFTs, but in order to keep pace with the technological advancement of our member firms we have invested dozens of millions of Swiss Francs to update our market surveillance systems. Other venues need to do the same as do governmental regulatory bodies. And this is a key challenge indeed.
7. HFTs have made new markets come into existence
It is easy to overlook that HFTs not only pose potential problems for existing market venues but also have actively created new venues. In Europe, similar to ECNs in the US markets, new MTFs were often co-founded by large HFT firms as has been the case for example with Chi-X Europe. HFT firms such as Citadel and Getco hereby generally act as seed investors and seed liquidity providers simultaneously. By providing consistent bid-offer spreads for securities listed on regulated markets, applying for instance techniques such as lit pegging, they make otherwise illiquid market venues come to life. Academically speaking they overcome “adverse selection” which might prevail in new venues because fundamental investors shun away from markets with supposedly large informational asymmetries. Practically speaking such HFT seeders are acting as market makers on new venues linked to securities listed in their home markets similarly as they do in exchange-listed derivatives linked to underlying cash markets. This quality of HFTs makes them a highly valuable additional set of market players in an environment such as Europe where de-regulation from 2007 was specifically introduced to create competition between market venues.
8. HFTs can damage existing markets by driving certain participants out
But “adverse selection” not only exists for new venues, which generally benefit from the introduction of HFT activity. It can also appear as a phenomenon in established regulated markets, or exchanges, when the arrival of HFTs can drive out the pre-existing fundamental investors. Such investors may feel that their usually much larger buy or sell orders generate signals to HFTs which they might employ to generate directional or statistical arbitrage trades to the detriment of the former. The danger is that such investors turn to the OTC market in order to cross trades outside the domain of public lit order books. This effect has indeed taken place to some extent, the size of which is at the core of an ongoing debate.
It has also given rise to the establishment of a specific and very successful MTF for fundamental buy-side investors, namely Liquidnet, and other attempts to create venues outside the reach of HFTs. At SIX Swiss Exchange we have always believed that there is a natural raison d’être for certain large in scale trading to happen off-exchange and we have therefore announced a landmark cooperation agreement with Liquidnet in order to connect our own buy-side with this special network. Through this link we ensure that all our members can interact efficiently in both a highly liquid lit order book as well as the most efficient large in scale-only venue globally.
For exchanges generally the potential crowding out effect of HFTs in lit order books remains an issue similar to the one of cohabitation between lions and the buffalos: too many lions killing too many buffalos will make the buffalo herd extinct, thereby undermining the existence of the lions themselves. But as with nature, markets also have a tendency, although no guarantee, to regulate themselves. It is therefore questionable whether regulators should set a percentage limit on HFT activity in a market or whether markets will find the right equilibriums themselves. So far neither academia nor public debate have determined how to derive an ideal share of HFT in mature securities markets. As it stands, the US experience indicates that a higher level of HFT activity overall increases velocity of markets beyond the level seen today in Europe.
9. Ensuring the integrity of markets requires large technology investments and co-ordination
So if market structure does not prevent a further rise of HFT activity with the described benefits and drawbacks, demands for keeping the integrity of these markets will rise. More complicated surveillance, rising quote-to-order ratios and so forth leave market operators and regulators with two principal challenges: better technology and better co-ordination.
There is a clear trend in the public debate that this is being more and more understood by all relevant parties. At SIX Swiss Exchange we have seen this trend arise several years ago. The result is that we have single security circuit breakers for all security classes both for price jumps and price avalanches as well as individual capacity limits for member connections. We have also been pressing our national regulator to establish Memoranda of Understanding with regulators in the EU in order to make sure coordination happens for simple issues such as trading halts.
Not all market venues however have such circuit breakers, as the US and EU observers painfully found out during the Flash Crash. Neither have EU regulators embraced the Swiss cooperation attempts seriously enough to the present day. This must change. When it does, hopefully sooner rather than later, the question will arise who has to ultimately bear the costs, in particular on the technology side. The standard answer so far under the unlevel playing field presented by the current MiFID regime has been to load the costs onto regulated markets. This is mainly because the operational availability of regulated markets has the highest benchmarks. In contrast MTFs and other venues have got away lightly under MiFID rules and in practical enforcement. In the future, a more balanced cost allocation ultimately including a higher cost sharing by intermediaries such as banks and other securities dealers must be considered.
10. HFT regulation should be risk-return based
In the final analysis, HFTs should be seen as both makers and potential breakers of markets. The current market structure in Europe, itself a product of intentional fragmentation, simply does not allow for a radical clamp-down on HFT anymore without endangering the benefits of all the new competition that has followed since MiFID. If re-regulation is needed, and this is still an “if”, it should be based on a well-founded analysis of the benefits and risks of such a move.
Simple answers such as the introduction of a Tobin tax, for example in the form of a punitive stamp duty per transaction, must clearly be rejected. We, as operators of highly liquid markets in the full range of financial securities, do and will help to establish fundamental data and findings for such an analysis together with regulatory and political bodies. We will also continue to invest in leading trading and market surveillance technology and push for international coordination as we have for many years.
But we must also ask for a more equitable cost sharing between all market participants in order to stop a race to the bottom in market integrity, triggered by the well-intended introduction of competition and growing fragmentation. Securities exchanges still represent the cornerstone venues in modern market economies that allow for the efficient formation of capital and best resource allocation. Our aim as exchange operators is through time to strengthen the economic role we play, while prevailing in fair competition, to the benefit of the whole public. The aim of the public itself should be the same.
About Dr. Christian Katz
Dr. Christian Katz is CEO of SIX Group's Cash Markets Division. The Division encompasses SIX Swiss Exchange and SIX Exfeed, a leading supplier of raw financial data. Before joining SIX Group in early 2009, Christian Katz headed the representative office of Goldman Sachs International in Switzerland and spent eight years with JP Morgan Chase in London. Christian Katz holds a Masters in Business and Finance and a PhD in Finance from the University of St. Gallen, Switzerland.