September 2017

Demonstrating Best Execution Under MiFID II - Clifford Chance

Using the MarketAxess Composite Price to demonstrate Best Execution under MiFID II  


Simon Gleeson, Partner, Clifford Chance

Caroline Dawson, Senior Associate, Clifford Chance



Some firms have in place policies and procedures which only permit them to execute where they have at least 3 competing dealer quotes. Where clients trade bilaterally, they will not necessarily have access to three or more competing dealer quotes. To address this, MarketAxess provides clients with composite prices, to give them an idea of where the market was trading when they executed a particular trade.  

In this article, we explore the extent to which firms may rely on tools such as the MarketAxess Composite Price to demonstrate best execution particularly if there are fewer than 3 competing dealer quotes available or where the firm is a price maker in the MarketAxess Open Trading protocol.

Although a firm could use information on other dealer quotes in the process of fulfilling its requirements under MiFID II, there is no specific requirement under MiFID II for them to use other dealer quotes in this way, and it is clear that other data and tools can be used to satisfy these requirements. These other data and tools could include tools such as the MarketAxess Composite Price.


MarketAxess' composite price

What are the sources for the MarketAxess Composite Price?

Information contributing to the MarketAxess Composite Price is comprised of the following sources, TRACE data, Trax data, MarketAxess inquiry data and MarketAxess inventory data each described more fully below.  

TRACE data: The Trade Reporting and Compliance Engine (TRACE) was introduced in July 2002 in order to increase price transparency in the US corporate debt market. TRACE captures and disseminates consolidated information on secondary market transactions in publicly traded TRACE-eligible securities. Information is available both in an end-of-day format and also as real-time price and volume information.

MarketAxess inquiry data: Clients that submit orders to buy or sell bonds on the MarketAxess system receive firm prices back from interested dealers. Typically, the client will trade with the dealer providing the best price.

MarketAxess inventory data: Dealers provide MarketAxess continuous bid and offer price and size information on a range of bonds available for trading. These figures are indicative, meaning non-executable. If a client asked a dealer for a firm quote, they would typically get a better price. As a result, the spread is wider than that shown via permissioned RFQ inquiry responses or by actual trading.

Trax data: Pricing data from Trax includes traded prices from approximately 30,000 trades a day for over 7,000 instruments, which are inputs into the formation of the Global Composite Price. Trax estimates that it processes approximately 65% of all fixed income transactions in Europe as part of its post-trade service offering, and in 2015, Trax processed 13.3 million fixed income transactions on behalf of its user community which contribute towards its pricing products.

How is the MarketAxess Composite Price calculated?

MarketAxess has composite prices for over 14,000 ISINs, with updates every 15 seconds. It is aggregated by bond and segmented by client direction; e.g. client buys vs. client sells.  Meaning, composite bid prices are calculated separately from composite offer prices even on the same bond.

The source data provides the “inputs,” or individual prices, that ultimately form the composite prices. As might be expected, specific filters and outlier screens are applied to validate the inputs. For example, the methodology includes a minimum size associated with an input (for inquiry and inventory data only). Final composite prices are calculated as the mean of all surviving inputs, where the minimum number is 3.

The MarketAxess Composite Price can then be used to benchmark individual trades in client portfolios. As discussed further below, firms should then be able to use the Composite Price to aid in demonstrating best execution.


Can the MarketAxess Composite Price be used to demonstrate best execution?

As discussed above, there is no express obligation under MiFID I or II for firms to obtain at least three dealer quotes in order to demonstrate best execution, so a firm should be able to demonstrate best execution using any other appropriate information that may be available to them (such as the MarketAxess Composite Price and the information required to be published under Article 27(3) and (6) of MiFID II).

What is the scope of the best execution obligation under MiFID II?

Under Article 27 of MiFID II, Member States shall require that investment firms take all sufficient steps to obtain, when executing orders, the best possible result for their clients taking into account price, costs, speed, likelihood of execution and settlement, size, nature or other consideration relevant to the execution of the order.

Article 27(5) requires firms to establish and implement an order execution policy. This policy must include at least those venues that enable the investment firm to obtain on a consistent basis the best possible result for the execution of client orders.

Article 27(7) requires investment firms to monitor the effectiveness of their order execution arrangements and execution policy in order to identify and, where appropriate, correct any deficiencies. In particular, they shall assess, on a regular basis, whether the execution venues included in the order execution policy provide for the best possible result for the client or whether they need to make changes to their execution arrangements, taking account of, inter alia, the information published under Articles 27(3) and (6).

Article 27 of MiFID II does not set out further guidance on how a firm should assess whether or not the execution venues included in its order execution policy provide for the best possible result for the client. There is some further guidance under Article 64 of the MiFID II Delegated Regulation, which indicates that a firm should take into account the characteristics of the client, the client order, the relevant financial instruments and the execution venues to which the order can be directed.

Article 65 sets out further guidance for investment firms providing the service of portfolio management, and require firms to provide clients with appropriate information about the entities chosen for execution. In particular, when a firm selects other firms to provide order execution services, it is required to summarise and make public the top five investment firms in terms of trading volumes where it transmitted or placed client orders for execution in the preceding year, and information on the quality of execution obtained. Article 65 goes on to state that this information shall be consistent with the information published in accordance with the technical standards in relation to Article 27(6) of MiFID II (i.e., the requirement for investment firms to publish information on their top five execution venues and information on the quality of execution obtained).

These requirements are set out in RTS 281 and include a requirement for firms to publish a summary of the analysis and conclusions they draw from their detailed monitoring of quality of execution, including:

  1. An explanation of the relative importance the firm gave to the execution factors of price, costs, speed, likelihood of execution or any other consideration including qualitative factors when making assessments of the quality of execution;
  2. A description of any close links, conflicts of interests, and common ownerships with respect to any execution venues used to execute orders;
  3. A description of any specific arrangements with any execution venues regarding payments made or received, discounts, rebates or non-monetary benefits received;
  4. An explanation of the factors that led to a change in the list of execution venues listed in the firm's execution policy, if such a change occurred;
  5. An explanation of how order execution differs according to client categorisation, where the firm treats such category of client differently and where it may affect the order execution arrangements;
  6. An explanation of when other criteria were given precedence over immediate price and cost when executing retail client orders and how these other criteria were instrumental in delivering the best possible result in terms of the total consideration to the client;
  7. An explanation of how the investment firm has used any data or tools relating to the quality of execution including any data published under RTS 27 [i.e., the requirement for trading venues and systematic internalisers to publish data on quality of execution];
  8. An explanation of how the investment firm has used, if applicable, output of a consolidated tape provider established under Article 65 of Directive 2014/65/EU.

As a result, in determining the quality of execution provided by a particular venue, a firm should determine whether there are any qualitative factors that it intends to take into account (and the relative importance that it intends to give to these qualitative factors) and consider whether or not to use any data or tools relating to quality of execution (including information published under RTS 27). Although a firm could use information on other dealer quotes in the process of fulfilling these requirements, there is no specific requirement under MiFID II for them to use other dealer quotes in this way and it is clear that other data and tools can be used to satisfy these requirements.

There is also further guidance in the Recitals to the MiFID II Delegated Regulation. Recital 100 states that where an investment firm selects a single entity for execution the firm must be able to show that this allows them to obtain the best possible result for their clients on a consistent basis and that they can reasonably expect that the selected entity will enable them to obtain results for clients that are at least as good as the results that they could reasonably expect from using alternative entities for execution. This reasonable expectation should be supported by relevant data published in accordance with RTS 27 and 28, or by internal analysis conducted by these investment firms.

Similarly, Recital 104 states that investment firms should gather relevant market data in order to check whether the OTC price offered for a client is fair and delivers on best execution obligations.

Again, while the requirements under these Recitals to conduct internal analysis or to gather relevant market data could be satisfied (at least in part) by obtaining at least three dealer quotes before executing a transaction, MiFID II does not specifically require this and it seems likely that these requirements could be satisfied by other data. 


Is there a requirement to obtain at least three dealer quotes in a firm’s assessment of quality of execution?

The FCA conducted a thematic review2 of best execution practices of UK firms and published its results in July 2014. This guidance was given in the context of MiFID I, but it is likely that the FCA will continue to apply a similar approach under MiFID II.

Two relevant issues addressed by this thematic review were the extent to which firms monitored best execution and the fact that some firms did not consider that they had a duty of best execution to a client that chose to deal on a quote.

The concerns around clients that chose to deal on a quote arose particularly in relation to contracts for differences, but the FCA indicated that similar issues had arisen in relation to other markets. In particular, the FCA commented that clients' ability to shop around for quotes may be limited if they cannot maintain multiple accounts with different providers. Even where quotes are visible, it is difficult to compare them in markets where prices move rapidly and quote information is not consolidated. One example that the FCA gave of good practice to address this concern was a situation where one firm made available to clients additional information relevant to underlying market infrastructure or liquidity providers used to construct reference prices. Where this data was not available, the firm instead applied a transparent spread to the last actual trade in the underlying financial instrument or sought prices from a market maker with a history of providing two-way prices on a consistent basis in order to manufacture a price. This indicates that the FCA considers that provision of information similar to that provided by the MarketAxess Composite Price would be helpful in providing best execution to a client. The FCA has also indicated that firms should price transactions in OTC markets transparently (e.g., based on benchmarks or other publicly available pricing data).

The FCA also commented on the use of price benchmarks in relation to its review of firms' monitoring practices. In particular, the FCA noted that firms' monitoring was commonly based on a comparison of order execution against inadequate or simplistic benchmarks and data, or using unsuitably wide tolerances, rules and alerts.

The examples provided of "inadequate or simplistic benchmarks" included situations where firms were unable to explain why they were using particular benchmarks in their monitoring programme, why they were thought to be adequate, how they had been developed and when they had last been reviewed. The FCA goes on to say that it has deliberately avoided prescribing specific benchmarks or tolerances because of the diversity of potential approaches.

The FCA's comments indicate that if a firm uses the MarketAxess Composite Price to assist in its best execution assessment and monitoring that firm should consider why it is using the index, how it helps in the assessment and monitoring process and ask MarketAxess for information on how it was developed and updates on when it was last reviewed. However, this would not prevent a firm from using the MarketAxess Composite Price and if the firm was relying instead on obtaining multiple dealer quotes it would also have to go through a similar process to demonstrate that this process was appropriate to assess and monitor best execution.

ESMA also discussed this briefly in the context of its MiFID II consultation process. In particular, ESMA mentions in its December 2014 consultation paper that it "believes that order driven markets should report additional elements on execution quality that rely on full pre- and post-trade transparency data. For instance, benchmark prices such as average and realised spreads, best bid and offer, depth weighted spreads as well as more simply metrics like open and close prices and measures such as the high and the low price of the day could be utilised".

MarketAxess does not operate an order driven market, but ESMA's comments indicate that firms are expected to make use of data published by markets in order to assess and monitor best execution, including benchmark data.


Would the MarketAxess Composite Price be considered to be an adequate benchmark for the purposes of assessing and monitoring best execution?

Although the FCA commented on the use by firms of inadequate or simplistic benchmarks, it did not provide any detailed guidance on the features of benchmarks that it considered to be inadequate or simplistic. However, it seems likely that in order to be considered adequate, a benchmark pricing service would need to fulfil the following criteria:

  • Draw data from more than one source;
  • Include, wherever possible, real trade data rather than solely rely on indicative or derived pricing;
  • Provide for checks to ensure that inappropriate information does not distort the output (e.g., excluding data for unusually large trades); and
  • Involve an element of back-testing (e.g., periodically testing the output against other published sources to ensure that it is generally aligned with these).

Based on the methodology and variety of component sources described above, it seems likely that the MarketAxess Composite Price should be considered to be an adequate benchmark for the purposes of assessing and monitoring best execution.

Email the authors: or


This publication does not necessarily deal with every important topic or cover every aspect of the topics with which it deals. It is not designed to provide legal or other advice.


1 RTS 28 has now been adopted by the European Commission, but not yet published in the Official Journal. The comments above relate to the version of the RTS adopted by the Commission, but we have continued to refer to "RTS 28" as these technical standards do not yet have an Official Journal reference number.

2 TR14/13