FINRA requests comment on the practice of internalizing customer trades in the corporate bond market after obtaining auction responses, commonly known as “pennying.” In particular, pennying involves a dealer, after receiving a customer order, initiating a bid or offer wanted auction process on behalf of a customer, reviewing the auction responses, and then executing the customer order itself at a price that either matches or slightly improves the best priced auction response. As discussed below, FINRA performed a review of corporate bond auctions conducted by retail firms on electronic trading platforms. FINRA found in the review that firms internalized executions at varying rates after initiating auctions, and that these internalized executions offered varying amounts of price improvement. Consistent with a recommendation from the Securities and Exchange Commission’s (SEC) Fixed Income Market Structure Advisory Committee (FIMSAC) and a similar request for comment published by the Municipal Securities Rulemaking Board (MSRB), FINRA is soliciting comment on when such executions reflect a practice of pennying, how pennying impacts market quality and whether further regulatory action would be appropriate.
Questions regarding this Notice should be directed to:
- Patrick Geraghty, Vice President, Market Regulation, at (240) 386-4973 or by email;
- Cynthia Friedlander, Senior Director, Fixed Income Regulation, Office of General Counsel (OGC), at (202) 728-8133 or by email; or
- Alex Ellenberg, Associate General Counsel, OGC, at (202) 728-8152 or by email.
FINRA encourages all interested parties to comment on the proposal. Comments must be received by October 16, 2020.
Comments must be submitted through one of the following methods:
- Online using FINRA’s comment form for this Notice;
- Emailing comments to [email protected]; or
- Mailing comments in hard copy to:Jennifer Piorko Mitchell
Office of the Corporate Secretary
1735 K Street, NW
Washington, DC 20006-1506
To help FINRA process comments more efficiently, persons should use only one method to comment on the proposal.
Important Notes: Comments received in response to Regulatory Notices will be made available to the public on the FINRA website. In general, comments will be posted as they are received. FINRA reserves the right to redact, remove or decline to post comments that are inappropriate for publication, such as vulgar, abusive or potentially fraudulent comment letters.
Parties should submit in their comments only personally identifiable information, such as phone numbers and addresses, that they wish to make available publicly. FINRA reserves the right to redact or edit personally identifiable information from comment submissions.
Background and Discussion
On June 11, 2019, the FIMSAC approved a Recommendation Regarding the Practice of Pennying in the Corporate and Municipal Bond Markets.1 FIMSAC recommended that: (1) the SEC make a statement disapproving of the use of pennying in either the municipal or corporate bond markets on any electronic trading venue; and (2) FINRA publish a request for comment on the use of pennying in the corporate bond market, similar to a request for comment on pennying in the municipal bond market published by the MSRB in September 2018. The Recommendation further noted that FINRA and the MSRB should coordinate any future regulatory action in response to the requests for comment.
The FIMSAC Recommendation defined “pennying” as a practice where a dealer initiates a bid or offer-wanted auction process on behalf of a customer, reviews the auction responses, and then executes the customer order itself at a price that either matches or slightly improves the best priced auction response. The Recommendation stated that the practice of pennying may appear to benefit a customer but may harm overall auction competitiveness over time.
The FIMSAC distinguished pennying from “last look,” which the Recommendation defined as a valid practice of a dealer reviewing auction responses as part of the dealer’s best execution process and internalizing the order with meaningful price improvement when appropriate.2 According to the Recommendation, given the competitive nature of the auction process, internalizing orders following a last look for best execution purposes should be a case-by-case occurrence, while pennying involves a systematic practice of internalizing orders that is not driven by best execution.
As the FIMSAC discussed in the Recommendation, the MSRB in 2018 published a request for comment on the practice of pennying in the municipal bond market.3 The MSRB request for comment noted the MSRB’s concern that a dealer’s use of the auction process solely for price discovery purposes, without a bona fide intent to trade, can harm the auction process by reducing bidders, thereby reducing the likelihood that the high bid in a bid-wanted will represent a bond’s fair market value. In its request for comment, the MSRB asked whether pennying should be considered an unfair practice or explicitly prohibited through rulemaking, and whether a firm could demonstrate it was not pennying by bidding blindly on auctions it initiates, or by providing “substantial” price improvement.
During FIMSAC consideration of the Recommendation, FINRA committed to conducting a review of electronic auction and execution practices in the corporate bond market, to better inform a request for comment. Accordingly, FINRA gathered sample data on electronic Request for Quote (RFQ) auctions for corporate bonds conducted in the first quarter of 2019 by a set of firms that generally represent retail customers. FINRA evaluated this initial set of data and developed an aggregate view of execution and internalization rates following an auction; overall, FINRA found that of the RFQs in the sample, 70 percent resulted in an execution; of those executions, 84 percent occurred with external bids and 16 percent were internal executions by the firm that initiated the RFQ.
FINRA then requested more detailed underlying auction data on a subset of the initial sample, where auctions resulted in internal executions. FINRA performed further analysis on this more detailed underlying data to evaluate the number of bids received in response to each RFQ, the submission time for each bid, and, in cases where a firm submitted its own bid on an RFQ it initiated, whether the firm submitted its bid without knowledge of the other bids collected. The more detailed data in the subset indicated that internal executions generally resulted from bids being submitted or updated by the initiating firm after the firm received bids from external firms during the auction. In addition, in the detailed subset of internalized trades, based on a comparison of internal and external bid prices, FINRA observed that 28 percent were executed based on internal bid prices that did not improve the best external bid, 16 percent improved the best external bid by 5 basis points or less, 7 percent improved the best external bid by 5.01-10 basis points, 9 percent improved the best external bid by 10.01-25 basis points, and 40 percent improved the best external bid by more than 25 basis points.4 At the firm level, FINRA found that practices varied significantly among the firms in the subset sample, with firms improving the best external bid by 25 basis points or less between 23 percent and 87 percent of the time in internalized trades.5
Informed by this data, and as more specifically outlined in the questions below, FINRA is requesting comment on the practice of pennying, and its definition. During FIMSAC discussion of the Recommendation, and in response to the MSRB’s request for comment, a number of market participants observed the difficulty in prescribing an amount of price improvement that should be considered slight or nominal, as opposed to meaningful. In addition, as suggested by the FIMSAC Recommendation, a policy decision on pennying inherently requires consideration and evaluation of both price improvement in specific customer transactions and the broad impact on market quality over time. The practice of pennying is of particular concern in the debt markets because of its potential impact on the use of RFQs, which are an important pricing mechanism in the absence of reliable, widely available quotations for debt instruments. Accordingly, FINRA is interested in market participant views on the impact of pennying on market quality, and the extent to which such views can be supported by data analysis.
Request for Comment
FINRA requests comment on all aspects of this request for comment. FINRA requests that commenters provide empirical data or other factual support for their comments wherever possible. FINRA specifically requests comment concerning the following questions:
- Do you agree with the FIMSAC’s definitions of pennying and last look practices and the distinction between the two? If not, how would you propose differently to define or distinguish between the two practices?
- Do the results of FINRA’s sample study accurately represent the nature or frequency of practices you observe in the markets? Do the results of the sample study demonstrate that in the corporate bond market, last look is a common practice to achieve best execution, or the practice of pennying is prevalent, or both?
- Is there additional relevant data or evidence you can provide based on your observations or experiences in the markets?
- If you have observed pennying or last look practices, are they more prevalent on certain types of electronic trading platforms? If so, which ones? Are these practices limited to alternative trading systems or do they occur on other electronic trading platforms? Are certain types of RFQs more likely to result in pennying?
- Have you observed pennying or last look practices occurring in transactions with institutional customers? If so, is there any relevant data or evidence you can provide based on your observations? Do you believe that FINRA should consider a study of such practices in the institutional markets?
- If pennying is defined as a pattern or practice of internalization with no or slight price improvement after viewing prices obtained through an RFQ, what amount of price improvement should be considered meaningful and what level of regularity would constitute a pattern or practice?
- Should price improvement be considered on a percentage or total dollar basis, or some combination of the measures? Does the answer depend on the size of the transaction? For example, should price improvement of 25 basis points on a $100,000 transaction, or $250, be considered more meaningful than price improvement of 25 basis points on a $10,000 transaction, or $25?
- Should the same amount of price improvement be considered more or less meaningful depending on the competitiveness of an auction? For example, would price improvement of 25 basis points be more meaningful after an auction with seven responses than an auction with one or two responses?
- How do transaction costs affect current measures of price improvement? Should such transaction costs be considered when comparing internal and external bid prices?
- Do firms apply the same mark-ups and mark-downs to external and internal bid prices to arrive at a final reported price? Are there reasons why firms do or would apply different mark-ups and mark-downs to internal and external bids?
- What level of regularity would signify a systematic business practice? How should regularity of occurrence be considered alongside the amount of price improvement? Should a less regular pattern with lower price improvement be considered similar to a more regular pattern with higher (but still slight) price improvement?
- What are the market quality and economic consequences of pennying? Does or will pennying harm overall auction competitiveness over time, for example by causing fewer firms to provide bids in response to auctions, or by causing responding firms to bid less aggressively? How can the impact of pennying be measured?
- Are there ways to measure the impact of pennying through data analysis?
- Are there different impacts for firms depending on whether the firms are believed by other market participants to engage in pennying? Are there ways for initiating firms to make it known to potential auction responders that the initiating firms do not engage in pennying? Should firms be required to disclose to customers information about their auction practices?
- Please provide examples of any impact you have observed caused by a practice of pennying.
- During FIMSAC discussion of the Recommendation, there was some support for a requirement that dealers “bid blind” in response to auctions their firm initiates. Under this kind of requirement, dealers would need to bid on auctions initiated by their firm on a blind, competitive basis during the auction period, the same as any other firm, without the opportunity to review other firms’ auction responses before entering the firm’s own order.
- Would a blind bidding requirement be an appropriate regulatory approach?
- If so, would a blind bidding approach need to allow for “last look” and an opportunity for meaningful price improvement?
- If not, should blind bidding still be considered as a best practice guide for firms?
- Are there any drawbacks to a blind bidding approach? Would a blind bidding approach impose any costs on customers?
- Are there any reasons why firms could or should not bid blindly into an auction they initiate on behalf of a customer? In other words, are there any reasons for initiating firms to first review external auction responses before providing their own bid?
- Would a blind bidding requirement be an appropriate regulatory approach?
- As FINRA continues to coordinate on pennying with the MSRB, consistent with the FIMSAC Recommendation, are there any differences between the corporate and municipal bond markets for which FINRA and the MSRB should account?
- See Recommendation Regarding the Practice of Pennying in the Corporate and Municipal Bond Markets (June 11, 2019) (“Recommendation”).
- Under Rule 5310, firms have an obligation for customer transactions to use reasonable diligence to ascertain the best market for the subject security, and to buy or sell in such market so that the resultant price to the customer is as favorable as possible under prevailing market conditions. See also Regulatory Notice 15-46 (November 2015) (providing guidance on best execution obligations in the fixed income markets).
- See MSRB Notice 2018-22 (September 2018).
- When measured more broadly as a percentage of overall RFQ auctions or executions (internal or external) by the firms in the detailed subset, the number of internalized executions with 25 basis points or less of price improvement represents roughly 9 percent of the total number of RFQs initiated by those firms, and 12 percent of those RFQs resulting in executions.
- For purposes of this analysis, FINRA compared internal and external bid prices on a pre-mark-up or mark-down basis. While mark-ups and mark-downs are included in reported execution prices to TRACE, FINRA understands that mark-ups and mark-downs are applied similarly to both internal and external bid prices; accordingly, FINRA did not include mark-ups and mark-downs in its calculations when comparing the internal and best external bid prices. A question on this issue is included below.