The considerable and growing interest among investors, issuers, regional broker/dealers and other market participants in a centralized order-driven liquidity mechanism is a manifest element of that natural selection process. Research indicates that such a neutral facility with full pre-trade and post-trade transparency — where all participants are equally empowered to access and offer liquidity and anonymously make markets — would materially increase liquidity and reduce transaction costs and market volatility.
Liquidity Resources
“Nature abhors a vacuum.” ―Aristotle
The growing liquidity vacuum in the corporate bond market is the result of a number of causes, including the market’s exclusive quote-driven OTC structure. While economic and regulatory pressures have diminished the capacity and appetite of dealers to source liquidity, whether obligated or opportunistic, the market’s singular quote-driven model precludes institutions with deep resources of alternative price-sensitive liquidity from directly filling that void.
Benefits abound in a centralized order-driven market, where all transactions are negotiated and settled anonymously. Institutional investors seeking higher yields, including those that traditionally buy and hold, could provide liquidity and profitably make markets in securities they hold. This model would also better position brokers and dealers to access and source liquidity and make markets with lower capital constraints.
The Growing Thirst for Innovation
“Innovation is the central issue in economic prosperity” ―Michael Porter
The author also cautions that "corporate bonds are not standardized” and "the core elements to successful CLOB trading are not present in the marketplace." That isn’t so. In fact, corporate bonds are better suited to anonymous matching than stocks, and those required core elements are already in place and poised for interaction.
Stocks are not readily fungible instruments and their trading requires exact matching of the potential counterparties’ interests (e.g., the likelihood of a seller of Stock A and a buyer interested only in Stock B matching is remote). Corporate bonds, however, are well-suited to standardization and grouping due to their unifying characteristics (coupon, maturity, rating, issuer, duration, convexity, etc.).
This homogeneous nature of congruent issues accommodates their relative pricing based on yield-spreads to more liquid instruments and optimizes the number of potential transaction counterparties. Rather than seeking a specific CUSIP, prospective bond buyers are generally interested in any order offering an issue that meets their desired characteristics and price/size requirements. Thus a transaction occurring between a buyer and one of an array of competing sellers (and vice versa) is much more likely.
The Coming Diversity of Options
“A horse never runs so fast as when he has other horses to catch up and outpace.” ―Ovid
History demonstrates that when multiple delivery models fairly compete, innovation and efficiency increase and the market benefits from that diversity. The number and diversity of available liquidity models in the corporate bond market will inevitably, and likely soon, increase.
Extensive informal polling of institutional investors indicates the most compelling model would be a fully transparent, order-driven network operated by a regulated, neutral third-party, through which all participants equally and anonymously:
-
Interact with an array of competing liquidity platforms and providers in both the primary and secondary market; and
-
Access and source liquidity and opportunistically make markets.
-
Most respondents also indicated that such a regulated market would likely:
-
Reduce transaction costs and elevate the Topic 820 classification of corporate bonds;
-
Promote investor and regulatory confidence;
-
Facilitate more effective and dynamic risk management; and
-
Assure every market transaction is executed at the best price, regardless of where the trade occurs.
Comments | Post a Comment
15 Comments to "Evolution, Liquidity Resources and the Growing Thirst for Innovation in the Corporate Bond Market ":
Anonymous
10 October 2012
Nice article on where the evolution is headed
Andy Nybo
10 October 2012
Larry -- You bring up a number of interesting points and I would agree the corporate market is ready for a move towards more automated trading. But I am baffled by your statements on standardization and fungibility.
Even if—and that is a big if—you assume that investors are truly indifferent to matching “like” corporate bonds with similar duration, credit and structural characteristics, the idea that there is sufficient buying/selling interest at any one point in time is suspect. It may be true in the days following a new issuance but over time bonds get held in portfolio and are not traded by (current) bond market investors like equities.
Bond markets are institutional. Sure, individual investors do buy a few bonds every now and then but the vast majority of bonds are held by institutional accounts. Average bond trade sizes are bigger, there are more than 200,000 different issues and the total number of investors trading bonds is much lower. Compare this to equity markets where the total number of stocks traded is in the thousands, the number of firms and individuals trading equities is in the millions and equities are traded for capital appreciation (mostly) and dividend yield (increasingly).
In short, stocks have many more market participants trading a much smaller number of homogenous instruments. And they are completely fungible. Corporate bond markets have fewer market participants, a much larger number and diverse set of instruments and are not fungible. Matching works in the former and has not proven to work in the latter.
Add to this all of the activity in ancillary markets and strategies surrounding equity investing and you have a very liquid marketplace. Much of this is dependent on the fungibility of equities. Short-selling, options trading, ETFs and dare I mention low latency strategies all contribute to a very diverse marketplace with considerable flow that is perfectly suited for anonymous matching.
I am not sure the same is true for corporate bonds either now or in the foreseable future.
Andy
lfondren
10 October 2012
Andy – It’s not that institutional bond investors are “indifferent to matching”, but rather that their investment metrics are focused more on the downside and yield/risk characteristics of an issue than on the future upside of the issuer. Thus corporate bonds, particularly investment grade issues, are more conducive to relational interaction among an assortment of potential counterparties and congruent issues than stocks.
For example, an all-to-all order-driven market for institutional trading of corporate bonds is much more likely to develop an initial “potential-match” between the interests of a prospective buyer and one of an array of competing sellers (and issues) than a stock exchange - where only prospective counterparties that entered potentially matching contra-orders for the same stock are eligible for further interaction. Indeed, if it were not for the fungible nature of corporate bonds, they could not be effectively traded in the OTC market.
Most liquidity in the secondary OTC market is not the result of dealers buying new issues and holding them for future sale, but rather their knowing “where the bodies are buried” and opportunistically buying them in to satisfy the current demand of other participants. A transparent platform that continually informed all current holders of securities meeting the expressed interests of potential buyers (and vice-versa), and that enabled all to anonymously and directly compete, interact and trade would tend to optimize liquidity.
This fungibility of debt instruments makes the total number of outstanding issues less relevant as most CUSIPs in a given cohort of congruent issues perform as if they were a single “virtual” issue. Moreover, as corporate bond limit-orders are priced at spreads to the dynamically-adjusting “zero-risk” yield of a comparable U.S. Treasury, they have a considerably longer shelf-life than stocks, which accommodates a more substantive interaction among many competing participants. If your assessment that stocks “…are completely fungible” (and thus infinitely interchangeable) is correct, why do they not trade at prices relative to a common comparable benchmark?
Once corporate bonds trade in a fully-transparent exchange-like environment, the benefits of that market structure currently enjoyed by stocks will likely flow to bonds as well. Security-level and portfolio-level risk management will then become much more dynamic and effective, with liquidity exponentially growing as participants hedge, refine and optimize their current holdings, and bid for new issues in when-issued order-books and single-price auctions.
The future bond market will be a very fluid and interactive environment.
Larry
Comments (7)
Anonymous
11 October 2012
I'm sceptical. In this new order-driven market for bonds, what's the average lot size? is it small enough to draw in retail investors? if not, where does this new liquidity come from? I'm with andy on this one - most bond investors don't see value in turning over their portfolios. i could see a retail bond exchange/market emerging, with small lot sizes, allowing the average joe to add bonds to their portfolio, but that opens up a new can of worms ... does the average joe understand the nuances of bond valuation, or would they just see a nice little spread over treasuries and have absolutely no idea of risk from issuer default...
lfondren
11 October 2012
The envisioned market is institutional in scope, accommodating aggregation of liquidity flows by registered broker-dealers for and among their retail clients. As the matching functionality of the order-driven model considers all elements of each order, including its size, it brings together only those potential counterparties that have expressed complementary size requirements. As to the average size of transactions, the mean trade-size on the InterVest platform was approximately USD3 million.
Discussions with a considerable number of institutional investors indicates that opportunistically making markets in securities they hold, and capturing the resulting increase in portfolio yield, would be very appealing – provided they could view and anonymously interact in the entire market at all times. The benefits of that yield-enhancing opportunity are somewhat akin to those reaped by those same investors in the Repo market.
Comments (7)
Anonymous
11 October 2012
Larry - Thank you for posting a thought provoking article on what is clearly the topic of the year. I share some of your views on the future of the bond market, but there must be differentiation between the future of the retail and institutional markets.
As we have seen with every other exchange based product (equities, listed options, futures), the exchange structure is great for smaller transactions and improves pricing efficiency for retail sizes. At this current time, one needs only to look at the TRACE tape to see that this type of market structure is desparately needed to help retail investors source corporate bond liquidity without being charged significant "access fees."
For large sized bond transactions, even in the most liquid cusips, a "fully-transparent exchange-like would not resolve block trading liquidity problems and would most likely adversly impact execution levels for end users. Using the example of exchange based products again, blocks trade OTC because large orders executed on an exchange create technical distortions that are taken advantage of by more sophisticated investors (see: High Frequency Trading). To quote Duncan Niederauer "For more than 1,200 widely held equities, more than 50 per cent of trades now occur “in the dark” – "
lfondren
11 October 2012
Thank you for your kind words. While the envisioned order-driven market is institutional in scope, it’s fully-transparent architecture enables all interested parties, including retail investors, to view the current price and size of all pending orders, and transactions as they occur, in real-time. All market participants can thus determine with confidence the current level of market pricing and the amount of fees and costs they are incurring.
Optimally, the envisioned market will be a hybrid model that incorporates both automated order-routing/matching functionality and a staff of neutral voice-brokers that facilitate confidential interaction among all participants. Those voice-brokers can (as can all market participants) also access and source liquidity in the OTC market, buying-in and offering-out where advantageous, to facilitate entry of firm and subject contra-orders on the order-driven market.
This hybrid model will also allow a large-block seller to employ smaller-sized “sounding” orders to test the current depth of demand and price-elasticity in the market and then, where appropriate, employ the voice-brokers to “work” the order by confidentially communicating to those initial buyers and others that a substantially larger supply is available. All resulting contra-orders are then matched and reported in real-time and the aggregate block trades without adverse market-impact.
Larry
Comments (7)
hoya1989
16 October 2012
Having authored the article Larry is referring to, I am elated to see his thoughtful and intelligent response. My original inspiration for writing such an article was to get the market thinking of the complexity of this discussion. As I mentioned at the conclusion of the OpEd, it is perplexing how some view this as a binary outcome, a struggle between two opposing market structures. My belief is that we will see a development of an OTC execution ecosystem, with a wide array of trading mechanisms.
As for some of the key elements of Mr. Fondren's article, I believe they are subject to the fundamental flaw that often plagues academics in this area of discussion - theory does not always translate well into practice. Having lived through the era of BondBook et al., it was very apparent that a limit order book type structure in the institutional corporate bond market was not going to work, at least at that time under those parameters, and thus it died a painfully quick death. That single episode in LOB failure does not suggest there is no room for LOB trading in the space, however. The voice-driven inter dealer market is essentially a functional LOB, and the quickly developing "live quote" driven retail ECN market place also resembles a LOB...NEITHER is a central limit order book, or CLOB, and it is highly unlikely that a CLOB will ever develop in this particular market place. CLOBs are more conducive to proprietary product markets such as futures, and even equities trade in a fragmented ecosystem of limit order books brought together by the specialization of OMS and EMS.
Finally, I have often heard the notion that bonds were made available to trade in the past on the NY Stock Exchange, and Mr. Fondren points this out. This notion of "if first you don't succeed, try and try again" is a great teaching lesson for parents to their children, but has no relevance in the discussion of product evolution. The market is the ultimate proving ground, and it is the market that determines the structure through which it will transact. Pundits can espouse models all day long, but in the absence of a forceful catalyst (such as regulatory mandates), models will always develop to suit the market structure of the day. As we enter this era of change, many want to suggest that a new "model" will lead to market structure change...many investments dollars that could have been put to good use have been wasted in pursuit of that business strategy.
It is undeniable that certain areas of the corporate bond market are conducive to limit order book trading and the market place needs to see further development of these mechanisms. It is also undeniable that the current RFQ mechanism has grown tired and antiquated, and the expansion of the "liquidity providing pool" combined with current institutional market structure has made this mechanism less efficient than its original intent. Change is certainly needed, and the debate is extraordinarily constructive - AP
Comments (5)
lfondren
16 October 2012
1 - Kudos to the Tabb Forum for providing an equitable platform for constructive and anonymous interaction among all participants, and a gravitational alert system that facilitates the simultaneous convergence and interaction of many competing views in a transparent exchange. The corporate bond market could use such a centralized and anonymous interaction model.
2 - AP, you seem to agree that there is benefit in having more than one liquidity model in the corporate bond market, but limit the structure of any alternative models to only those that fit into the existing “OTC execution ecosystem”. Experience indicates the optimal market structure would include at least one, if not a series of dynamically interlinked, order-driven execution models that are regulated, fully transparent and anonymous. Through that diversity of models, the market would organically expand to encourage and enable all participants to contribute and draw upon the confluent liquidity flows within the expanded ecosystem.
Your assertion that some of the key elements of my article are “theory” and “subject to the fundamental flaw that often plagues academics in this area of discussion” is misguided. My observations are based on my personal experience as founder and CEO of the InterVest centralized limit-order platform and the successful matching of many anonymous orders entered by our investor, issuer and broker/dealer subscribers. A candid analysis of the history of the InterVest CLOB will reveal its demise was not the result of any lack of efficiency, liquidity, subscribers or market appeal.
While you admit that it’s “undeniable that the current RFQ mechanism has grown tired and antiquated”, you choose to limit the structure of alternative execution mechanisms to only other versions of a decentralized RFQ archetype. The commercial viability of a regulated centralized order-driven model has been demonstrated and, while you deem the “rush” toward that anonymous all-to-all alternative as “illogical”, it’s appeal among deep alternative resources of price-sensitive liquidity is both palpable and growing.
I agree with your assessment that, ultimately, “it is the market that determines the structure through which it will transact”, provided that market is truly free and sufficiently transparent to accommodate regulatory surveillance. I also agree that a “forceful catalyst” is likely needed to expand the diversity of the current corporate bond market. However, I believe that catalyst will likely arrive not in the form of “regulatory mandates”, but rather in the form of facilitating regulatory oversight of a single, or group of interlinked CLOBs that augment, enhance and potentially restore the withering levels of dealer-sourced liquidity.
Thanks again to the Tabb Forum for facilitating this discussion, as we clearly agree that “Change is certainly needed”, and that this ongoing dialogue is “extraordinarily constructive”.
Larry
Comments (7)
hoya1989
16 October 2012
Larry
TABB Forum is a wonderful marketplace for the exchange of ideas agreed! I wanted to clear up three items. First, I'm in no way suggesting you are an academic, only that several notions you espouse are excellent theoretical concepts, but have not been proven effective in the "Court of Practice." I'm sure you are a credible individual, and certainly very articulate and intelligent based on your piece. Second, I agree the market will benefit from a multitude of mechanisms and platforms, hence the "ecosystem" I refer to. I never implied these systems have to fit within existing market structure, but instead clearly stated that "changing behavior" does not always translate into changing market structure, and furthermore, no particular electronic trading model will change market structure unless the structure of the market place is desiring to change. Finally, I am not limiting the structure of alternative models to the decentralized RFQ archetype...in fact, the entire premise of my discussion is based around the notion that the discussion should not be a binary choice of RFQ versus CLOB.
At some point, those supporters of CLOBs are going to come to the realization that CLOBs and Limit Orders Books (LOBs) are two mechanisms separated by more than just semantics. Equities trade in a market structure of fragmented limit order books, voice-driven request for quote venues, and dark pools. This ecosystem of mechanisms is connected and supported by a vast array of order and execution management systems. Furthermore, whilst equity transactions can come in the form of sponsored access, they may also be consummated through direct access. This varies considerably from your standard execution of a proprietary 30 year futures contract traded on an exchange.
As I said, a subset of the market may be conducive to LOB trading, but as my article pointed out, "the definition of insanity is doing the same thing over and over again expecting a different outcome." Trying to get all of these incongruent securities to trade in one centralized place has been no better example of that age old decree.
Comments (5)
lfondren
17 October 2012
AP
Thank you for your kind words, and for clarifying several points upon which we now seem to be in agreement.
It appears we agree that:
1) The ecosystem of the entire corporate bond market (including its constituent “OTC ecosystem”) will benefit from the diversity offered by one or more alternative execution platforms;
2) The “changing behavior” of market participants could include their use of both quote-driven RFQ models and non-fragmented order-driven CLOB models;
3) Any new CLOB or other alternative liquidity model must fairly compete with all other execution mechanisms and platforms in the “Court of Practice”;
4) The commercial success of any new alternative model will be largely dependent upon the desire or need of many market participants to change their behavior; and
5) The elements differentiating a Centralized Limit-order Book (CLOB) from a decentralized Limit-Order Book (LOB) include its accommodation of all-to-all access, 360-degree transparency, regulatory surveillance and every participant’s ability to both access and source liquidity and opportunistically make markets.
As for those areas in which we continue to hold opposing views, I suggest we agree to disagree. Indeed, without opposition, no atom, molecule, cell, organism, structure or universe could exist.
Larry
Comments (7)
Anonymous
19 October 2012
Larry - I read your article and the comments with interest.
I also checked out http://www.delphx.com, but it doesn't seem to offer much in terms of an explanation about what your company does. Can you elaborate on that?
lfondren
19 October 2012
Thank you for your interest DelphX. Formed in 2011, DelphX LLC is a privately owned company poised to become the innovation leader in anonymous and transparent fixed income markets. In early 2013, we’ll be launching the first regulated order-driven platform to provide market wide pre-trade and post-trade price transparency, real-time Topic 820 pricing of all eligible debt instruments (including those that rarely trade) and anonymous all-to-all interaction and settlement among our subscribers.
My article and comments above describe at a high level some of the features and benefits we’ve integrated within DelphX. I encourage you to visit www.delphx.com again early next month, when the first in a series of unveilings will occur, and to follow us via LinkedIn. Thanks again.
Larry
Comments (7)
bondcube
28 March 2013
I have arrived a little late for this discussion, but I did visit the website today. Is it possible to get an update comment on how things are going please?
Comments (7)
lfondren
28 March 2013
Hello Paul (bondcube).
The Tabb Forum article, " Revolution, Darwin and the Illogical Rush to a Centralized Corporate Bond Market", my reponding article, "Evolution, Liquidity Resources and the Growing Thirst for Innovation in the Corporate Bond Market", and the comments posted in response to each, dealt with the issue of whether the growing demand for a "CLOB" among market participants was justified and whether such a model could successfully restore liquidity in the corporate bond market. I'd be delighted to respond in this venue to any comments you have regarding that discussion.
For a comprehensive, concise and, I believe, prescient view of the current and emerging structure of the US corporate bond market, I encourage you to obtain a copy of the new Celent Report, "Innovation In Focus: Electronic Trading Platforms In US Corporate Bonds". http://www.celent.com/reports/innovation-focus-electronic-trading-platforms-us-corporate-bonds, , , , )
It is no secret that my firm is actively collaborating with other firms to implement a neutral SRO-like facility that will manage a CLOB consistent with that described in my article and related comments. I believe it's prudent, however, that announcements regarding the progress of that collaboration, and any unveiling of the resulting facility, should be exclusively facilitated through www.delphx.com and the sites of our Alliance partners.
As Bondcube and DelphX seem to have a common vision as to the value of transparency and the need to facilitate best executions in the global corporate bond market, I'd be delighted to discuss how we might explore a collaborative undertaking. If such a discussion is of interest, please contact me at lef@delphx.com.
Thanks.
Larry
Comments (7)