Shining a Light on Trading Costs

There is a danger that some market participants might view regulations such as the Markets in Financial Instruments Directive II (MiFID II) as nothing more than a compliance burden. How might new regulatory requirements result in positive changes to investment processes and practices around reporting and transparency?

David Weisberger: Any new rule or regulation that forces asset managers to predict, measure and control execution costs as part of their investment process can improve returns for their investors. While it is too soon to speculate on the percentage of managers that will merely pay lip service to the regulations instead of embracing them, it is likely that most will make substantive changes.

Under MiFID II, European best execution regulations have been strengthened for transactions in asset classes including derivatives, fixed income and foreign exchange, as well as equities. This means firms must have policies in place to explain transactional elements such as venue selection, timing, price and trading costs to clients. Asset managers that not only comply with these new rules but actually incorporate predicted trading costs directly into the portfolio construction and asset selection process could see major performance benefits by choosing more suitable position sizes.  

The asset management process consists of three interrelated disciplines: asset selection for the purpose of generating alpha; portfolio construction for the purpose of risk control versus the manager’s benchmark; and trading for the purpose of implementing the desired portfolio while capturing as much alpha as possible.

It is axiomatic that the more integrated the trading function is within the asset selection and portfolio construction process, the higher the returns will be for the asset owners. This is due to the simple investing principle that all strategies suffer from diminishing marginal returns as the amount of money invested in that strategy increases. Every investment strategy has an optimal size that maximizes total return, and every investment strategy eventually reaches a certain size, beyond which it loses money on an absolute basis (see figure 1).

This is because, as position sizes increase, the cost of entering and exiting the chosen positions also increases. For example, if a portfolio manager really believes in a stock and thinks it will outperform its peers by 10%, it is likely the manager will purchase a sizeable position. If, on the other hand, a position over €25 million is predicted to cost 5%, then it would not make sense to try to establish such a position without tightly controlling the trading process. Otherwise, the cost of entering and exiting the position would erase the predicted outperformance. The only way to ensure the construction of a portfolio is at the optimal point on this curve is to integrate predicted trading costs directly into the portfolio construction and asset selection process.

It is important to understand that this concept is also true when trading for the purpose of diversification. Consider an asset manager that is underweight in a particular sector and has five stocks with similar return profiles to choose from to reduce this risk. Imagine further that three of these stocks have wider spreads and lower liquidity than the other two. If that manager has no opinion on the prospects of the five companies, the correct selection would likely be to purchase more of the stocks with tighter spreads and higher liquidity. This type of decision-making is very hard when the trading process is not part of the investment process. 

By integrating predicted trading costs into the portfolio construction and selection process, managers can insure against such underperformance, positioning their portfolios at the optimal point on the curve illustrated in figure 1.

When evaluating current and prospective asset managers, should asset owners consider trading process and measurement as part of their due diligence? If so, why is this important?

David Weisberger: Absolutely. One of the most important questions facing asset owners when they evaluate investment managers is whether or not the manager’s process will yield consistent returns. Most managers will provide a track record of their asset performance as well as their fees, and some will demonstrate how they back-test new strategies and changes to their strategies. While this is more likely with quantitative managers, fundamental managers may do something similar by describing their stock-picking methodology. In order to demonstrate consistency, however, it is also important to evaluate trading costs as measured by price slippage from asset selection to implementation. This is because trading costs, relative to the volatility and liquidity of assets, should be extremely consistent over time. Managers that outperform in this area tend to do so consistently, and that makes their return profiles more predictable. As a result, managers that demonstrate a systematic approach to trading as well as documenting their implementation costs are more likely to have their future returns correlate to their past returns. 

Should new financial regulations emphasize the importance of third-party data validation? If so, what are the main benefits of having data validated by a third party?

David Weisberger: Third-party validation should be a key element of the best execution evaluation process—for example, public companies are required to engage auditors to approve their financial results for very good reasons. There are many benefits to the use of third-party validation, but the most important are the elimination of conflicts of interest and the guarantee of more accurate results. Trading systems that use flawed market data or have bugs in their processes will not identify those issues if the evaluation process uses the same flawed data. As such, the use of independent data sources to validate market prices should lead to improved accuracy.  

According to a recent survey conducted by Markit and WatersTechnology, less than half of organizations currently use pre-trade analytics for equities, less than one-third use this type of analysis for fixed-income transactions and about 40% use it for foreign exchange trading. Why are these figures so low? What benefits can investment managers derive from using pre-trade analytics?

David Weisberger: These figures are extremely troubling. Integrating pre-trade analysis into the asset selection and portfolio construction process is the only way to ensure that such decisions are optimal. In addition, pre-trade benchmarks that accurately predict the degree of difficulty of individual trades are the only way to fairly evaluate brokers and strategies used by asset managers. Otherwise, there is a risk that inferior brokers or strategies that were allocated much easier trades will have superior metrics and will, therefore, be allocated more order flow than they deserve.  

Portfolio managers should of course strive to ensure accuracy when it comes to pre-trade analytics, but even the existence of such tools should boost performance. A manager who has access to pre-trade analytics is able to distinguish between assets on a relative basis even if those cost estimates are off by 50%. This will allow for better decision-making. Take the example of a portfolio manager that must choose between two stocks in the same industry with virtually the same return profiles, but one is four times more liquid than the other. Buying the less liquid stock would be a bad decision. And while that may sound very obvious, for a manager trading hundreds or thousands of stocks without the use of pre-trade analytics, it is virtually impossible to avoid making such decisions.

What tools are available to bridge the differences between fixed-income and equity markets in support of measuring execution quality? 

David Weisberger: Unlike equities, fixed-income markets are generally not ‘order driven’. In practice, equity traders use orders for specific instruments at specific prices at specific points in time. In the fixed-income market, where the average instrument trades less than once a day and where a company might have a large number of similar bonds to trade, there are often no orders available. Fixed-income managers frequently react to bonds that are offered to them instead of placing orders. This means that using a ‘tape’ to evaluate how orders are handled is not typically helpful in measuring the fixed-income trading that occurs within the market. Therefore, unlike equities, the best way to evaluate trading quality in the bulk of fixed-income instruments is to compare trades to a third-party evaluative pricing service. Such analysis can demonstrate how trades relate to the quoted spreads, the benchmark instrument price, as well as the history of both quotes and trades. As ‘electronification’ increases in fixed-income trading, this analysis will become more accurate and will start approaching the sophistication of equities. Considering the breadth of the market, however, such a trend is only likely to occur in the most liquid instruments.

_{Markit is a leading global provider of financial information services. We provide products that enhance transparency, reduce risk and improve operational efficiency. Our customers include banks, hedge funds, asset managers, central banks, regulators, auditors, fund administrators and insurance companies. Founded in 2003, we employ more than 4,200 people in 13 countries. Markit shares are listed on Nasdaq under the symbol MRKT. For more information, please visit www.markit.com}

_{Contact: david.weisberger@markit.com}

Read/download the article in PDF format

Download the white paper, A Fresh Perspective: Enhancing Transaction Cost Analysis Effectiveness

Sponsored content