Expanding Best Execution Trading to Canada
By: Chris Rice, Head of North America Trading
The main priority of trading, regardless of market, is to implement the decisions of a portfolio manager in a way that maximizes value. Traders are responsible for balancing a variety of factors in the execution of an order with particular focus on timing versus market impact, liquidity sources, general strategy and the portfolio manager’s overall risk tolerance. Superior management of these factors should result in reduced trading costs and ultimately better performance. As investors focus on minimizing transaction costs, best execution in trading is becoming an even more important strategic component of investment management. The following essay focuses on best execution issues in Canada. Unlike SEC requirements for US investment firms, Canadian investment managers are not mandated by the Ontario Securities Commission (OSC) to document their best execution policies. The OSC does encourage Canadian firms to generate “broad” best execution principles and procedures internally, but there is no rule about written documentation for compliance purposes. However, the Canadian investment community expects the OSC will soon standardize best execution policies in much the same way the SEC did in the US. Canadian dealers are already held to the best execution definitions under the Universal Market Integrity Rules (UMIR). While still “broad”, UMIR definitions seek to justify the execution price of any transaction while considering previous trade volumes, security-specific liquidity, spread size 1 and overall direction of the market for the security. The definitions underline the fact that price is not the sole measure of best execution, but rather a variety of other factors must be considered in evaluating the entire framework of a trade. Canadian Market Structure The Canadian marketplace is considered a “dealer market” in which liquidity is not fully displayed to investors. While the US has multiple linked trading venues that compete for equity trading and promote efficiency, Canada is basically a monopolistic system with the TSX dominating both stock trading and listing. Under this one exchange structure, Canadian dealers have developed a so-called “upstairs market” which has typically constituted a considerable portion of overall trading value, mainly in the form of block trades. With such a small concentration of major investors in Canada, dealers have historically filled a necessary void by both providing capital in order to facilitate trading and by acting as agents in the search for natural liquidity. Even as markets have evolved, dealers have been reluctant to relinquish control over this block order flow, as it constitutes a high level of value in terms of information and in turn generates a substantial level of profitability. This dealer control over block flows in Canada has so far hindered the development of electronic crossing networks and direct market access tools shown to increase trading efficiency (in terms of better pricing and lower costs) in markets like the US. However, there has been a gradual shift in how Canadian managers view their benchmarks, with a higher emphasis on best execution concepts including opportunity costs and adding value through trading. Increased foreign investment, both institutional and retail, has also helped to spur demand for greater transparency and more robust trading tools to minimize information leakage and market impact. As such, the Canadian trading environment has been gradually
evolving away from being strictly a dealer market, although inherent differences in technology, capital levels and general culture make it difficult to envisage a complete transformation. Changing Trading Environment in Canada Some aspects of US trading have clearly changed the trading landscape in Canada. Canadian managers and traders are modifying the way they evaluate and execute trading strategies and, as a result, liquidity is migrating away from the historical “upstairs market.” A greater focus is now on portfolio or basket trading, and benchmarks are shifting to strategies with lower opportunity costs such as implementation shortfall, volume-weighted average price (VWAP) and volume participation. Conventional wisdom that the last block print represents the best price, regardless of opportunity cost, has generally given way to a more thorough analysis that includes timing needs, liquidity, risk tolerance and changing market conditions. This shift in attitude has required that traders have an increased array of trading options including algorithmic strategies through direct market access (DMA) tools and electronic crossing networks. It also means that exchanges will need more robust technology to handle increased order flow. With the advent of DMA and algorithmic trading, buy-side traders have been able to wrest away some control from the dealer-dominated Canadian market. Direct market access allows a trader to act in the market without a standard full-service broker. Not only does direct market access give a trader more control over the execution, explicit commissions are typically half what a typical dealer would get. Within the DMA framework, algorithmic trading has become the focus as front-end service providers race to provide traders with the capabilities they demand. Algorithms are computer-designed programs which follow a pre-defined trading strategy electronically. Top algorithms track benchmarks such as VWAP, time slices, specific volume participation levels or implementation shortfall targets with the least amount of tracking error. Notwithstanding the powerful potential of these tools, it’s still the trader’s task to implement algorithms properly in the right situations and to make adjustments to execution strategies as market conditions change. Canadian traders will also soon witness the launch of two new electronic crossing networks (ECN) Liquidnet Canada and TriAct Canada. The goal of both ECNs is to give traders additional options in the search for liquidity and the ability to trade block positions in a more efficient 2 manner. Liquidnet is a buy-side only crossing network where a trader can directly enter a negotiation with another trader who has the offsetting side of a potential trade. Once the negotiation is complete the execution is posted on the exchange without dealer intervention between the two participating parties. As with any alternative trading source, liquidity will be key. TriAct 3 is another crossing option through which liquidity providers (both buy-side and dealers can participate) can enter a matching process in an attempt to locate the opposite side of the trade. The system aims to offer “dark liquidity,” in which a trader can enter an order into the pool with its being visible only to the counterparties showing an off-setting position. Canadian dealers could use this tool to internalize their trade flows by first checking TriAct for natural liquidity before moving orders directly to the market. Matches in TriAct are executed in 30 second intervals and print at prices inside the current Canadian National Best Bid/Offer (NBBO). In addition to dealer flow, an ongoing stream of market-flow orders from ITG users glance into the TriAct liquidity pool in an attempt to match before moving on to other specified destinations. Another notable development in Canada is the rise of the Canadian Trading and Quote System (CNQ). This alternative stock exchange represents the first real domestic competition that the incumbent TSX has faced. There is speculation that if the CNQ can provide better execution and
liquidity options, two of the TSX’s major revenue sources, trading (directly) and listings (indirectly), could be infringed. The CNQ has created a trading facility called “Pure Trading,” which its developers describe as faster and cheaper than the TSX. With Pure Trading, CNQ aims to provide enhanced overall system performance and connectivity as well as a fee structure that is more attractive for algorithmic trading (i.e., smaller slices, more frequent orders). 4 The goal is to generate more algorithmic trade flows through the Pure Trading platform based on the system offering faster execution speeds (relative to the TSX) at a lower overall cost. It will be interesting to see the extent to which Pure Trading enhances the amount of displayed liquidity in Canada and the amount of algorithmic trade flow the exchange captures. Our Approach to Capturing Value in Canada In Canada, the approach to trading a specific order depends on the pooled investment vehicle’s strategy as well as specific characteristics of the basket. Typically, it’s understood that passive strategies have little information driving how and when trades are submitted. That being said, opportunity cost is less of a concern, so finding available liquidity (while minimizing transaction costs) becomes the primary goal. Passive order lists can be further broken down to determine high impact names (significant percentage of average daily volume) vs. low impact names (insignificant percentage of average daily volume). Ideally, low impact names should be executed through a DMA application at a reduced commission rate. The preferred method for handling high impact names would be to locate natural liquidity in the market, typically through a Canadian dealer, and then pair off so as to minimize price impact. This is where the importance of relationships with the Canadian dealer community comes into play as they control the majority of non-displayed liquidity. Orders for actively managed strategies can be thought of as having a high level of information content with the expectation that a certain level of alpha will be generated. Therefore, opportunity cost is much more of a focus in the execution strategy compared with trades for passive strategies. Generally, arrival price is used as a benchmark by traders for active orders as it’s assumed that the portfolio manager would ideally own or sell the position at current pricing levels. All alpha forecasts based on arrival price must take into account expected trading costs, easily generated through pre-trade analytic programs. The trader should then collaborate closely with the portfolio manager over the life of the trade so there is no disconnect between how the order is executed and the expected alpha the trade should generate. In such situations, explicit costs like commissions become less of a concern, especially for high-impact names. Moreover, brokers may be asked to trade principally (provide capital commitment) in order to minimize the timeframe for completion of the trade and to ensure minimal erosion in the value of the trade idea (alpha). Other factors such as cash constraints, sector movements, and economic news and commodity prices obviously all play a role in how a specific list is worked by a trader. With the ongoing advancement of technology in Canada and the continued development of DMA tools and ECNs, the ability of traders to minimize information leakage and maximize trading efficiency will only be enhanced. These developments will help emphasize the value a trader can add to the investment process and result in less dependence on the brokerage community. It goes without saying that traders will continue to seek out all available tools in an effort to achieve best execution and maximize portfolio value.
1 2 3 4
http://www.rs.ca/en/mktpol/umir01.asp?printVersion=no&loc1=mktPol&loc2=umir&loc3=01 http://www.liquidnet.com/company/ http://www.triactcanada.com/main.html http://www.cnq.ca/Page.asp?PageID=122&ContentID=657&SiteNodeID=190&BL_ExpandID=1399
October 3, 2006 __________________________________________________________________________
This material is for your private information only. The views and comments expressed in this commentary are the views of Chris Rise only through the period ended October 3, 2006 and are subject to change based on market and other conditions. The information provided does not constitute legal or investment advice and it should not be relied on as such. It should not be considered a solicitation to buy or an offer to sell a security. It does not take into account any investor's particular investment objectives, strategies, tax status or investment horizon. All material has been obtained from sources believed to be reliable, but its accuracy is not guaranteed. There is no representation or warranty as to the current accuracy of, nor liability for, decisions based on such information. Past performance is no guarantee of future results.