Treasury Notes

 Examining Corporate Bond Liquidity and Market Structure

By: Jake Liebschutz and Brian Smith
7/7/2016


This blog post is the fourth in a series on fixed income market dynamics by the Department of the Treasury to share our perspective on the available data, discuss key structural and cyclical trends, and reiterate our policy priorities. This post examines the corporate bond market.

The corporate bond market plays a critical role in the U.S. economy.  Businesses tap the bond market to raise over $1 trillion in financing each year, and the more than $8 trillion of corporate bonds outstanding represent an important asset class for a variety of investors.  The corporate bond market, in contrast to the Treasury market discussed previously in this series, is highly diverse with tens of thousands of distinct securities. Accordingly, liquidity differs in the corporate bond market.  While certain bonds trade frequently, many rarely trade. 


Recently, some have argued that liquidity conditions in the corporate bond market have deteriorated.  Although there have been anecdotal reports of periods during which liquidity conditions have been challenging, the corporate bond market has always been less liquid than many markets, and the available data does not show convincing broad-based evidence of dwindling liquidity.  Nevertheless, the corporate bond market is undergoing an important transition.  After reviewing common measures of liquidity, we highlight several market trends that are transforming the structure of risk transfer in the marketplace.   


Trading Volumes: While traded volume does not reveal the cost of trading nor potential desired but unexecuted trading activity, it is a direct measure of the level of activity that the market accommodates.  Both investment grade and high yield corporate bond average daily trading volumes have trended upward, roughly doubling since the financial crisis.  Despite growing volumes, record-breaking issuance has resulted in a decline in the overall “turnover.”  While turnover is a commonly cited measure of liquidity, its application may be limited given a variety of factors affecting the corporate bond market, including the increase in recent years of smaller issuers whose bonds trade less frequently and the prevalence of long-term holders who naturally trade less frequently. 

blog1.PNG 
 
In addition, the emergence of exchange-traded funds (ETFs) that track corporate bonds has provided market participants with another avenue for trading corporate bond exposure.  Trading in corporate bond ETFs has increased substantially (to about $2.5B average daily volume), with high yield corporate bond ETFs in particular experiencing significant growth in comparison to the amount of trading in the underlying bonds.  However, corporate bond ETFs are relatively new and their effects on market dynamics through the credit cycle are not fully tested. 

 

 
blog2.PNG 

 

 
Bid-Offer Spread: The spread between the average price at which securities are bought and sold provides a measure of the transactions costs in corporate bonds.  Investment grade and high yield bond bid-offer spreads are generally consistent with or below recent historical levels.  It is important to note, however, that the level of bid-offer spread in this market has consistently been fairly high compared to other, more liquid fixed income markets.  Moreover, because these spreads are estimated from transaction data, they do not capture cases where investors may have forgone trading due to high bid-offer spreads.
 
blog3.PNG 
 
Trade Size: Corporate bond trade sizes declined through the financial crisis but have generally been increasing since then, though they remain lower than their pre-crisis peak.  The decline from the pre-crisis period may reflect the reduced willingness of intermediaries to bear certain risks associated with facilitating large trades. Anecdotally, market participants have reported more difficulty executing large “block” size trades compared to the pre-crisis period.  However, the changes in market structure discussed below may also be contributing to smaller trade sizes and the implications of the trend for overall liquidity are unclear.
 
blog4.PNG 
 
Price Impact of Trades: Another measure of liquidity is the impact that executed trades have on market prices.  In more liquid markets, executed trades of a given size generate less price impact than in less liquid markets, all else equal.  The below chart shows that the price impact per $100 million of corporate bonds traded has declined since the crisis and now sits below pre-crisis levels.
blog5.PNG 
 
Dealer Inventories: Many commentators have pointed to declining dealer inventories as evidence of reduced corporate bond liquidity.  However, the link between inventories and liquidity is not straightforward.  Historically dealers have held significant inventories to facilitate principal intermediation, while agency intermediation, discussed below, would not require such large inventories. 
 
Furthermore, some have argued that dealer inventories can serve as “shock absorbers” during times of stress, as dealers would expand their inventories by buying bonds.  However, even at their peak in 2006, dealer inventories represented less than half of 1 percent of the corporate bond market, and so were unlikely to provide significant shock absorption.  Moreover, dealers holding large inventories in times of stress may not have capacity to take on significant additional positions.  Instead, they may be motivated to reduce their own positions, often at the same time customers are attempting to do so, potentially adding to a sell-off.  As the following chart plotting changes in dealers’ corporate bond positioning against changes in financial conditions shows, there is little evidence that dealer inventories have expanded during periods of stress (i.e., there are few points in the upper right quadrant).  The points in the lower right quadrant highlight instances of dealers reducing their positions as financial conditions tighten (most dramatically during the financial crisis).
 
blog6.PNG 
 
Issuance: Primary issuance in the corporate bond market is important in and of itself as an indicator of financing activity, but it also provides an indirect indicator of the health of the secondary market; investors consider secondary market liquidity in evaluating whether to buy potential new issues.  Corporate bond issuance is influenced by a number of factors, but its continued strength to date suggests secondary market liquidity conditions have not been constraining issuance.  Investment grade issuance has set record highs for four straight years and the first quarter of 2016 was similarly strong.
 
blog7.PNG 
 
High yield issuance slowed over the past year as credit concerns in certain sectors increased, but overall issuance remains within historical ranges.
 
blog8.PNG 
While some measures, such as lower trade sizes and dealer inventories, are frequently cited as evidence of declining liquidity, the available evidence, when viewed holistically and in light of recent market trends, does not suggest a broad-based deterioration in liquidity.  Instead, most measures are within historical ranges.  Nevertheless, the corporate bond market is undergoing significant changes that are reshaping the nature of trading and liquidity provision, including: the shift from principal- to agency-based intermediation, growth in electronic trading, and the emergence of new trading platforms.
 
Intermediation Model: Buyers and sellers in the corporate bond market, with its large number of diverse securities that generally trade less frequently than standardized securities, have historically relied upon broker-dealers to intermediate transactions in a principal capacity.  Under this model, a potential buyer or seller calls a dealer to request a price for a bond.  If the dealer and the potential buyer or seller agree on a price, the dealer executes the transaction for its own account.  Increasingly, however, dealers have begun acting as agents, looking to match buyers and sellers directly, rather than holding positions on their own balance sheets, to reduce risk and required capital. 
 
One recent study suggested that for the most active dealers the share of agency intermediated trades has roughly doubled from 7 percent to 14 percent since the pre-crisis period; another recent study looking at all dealers estimated that the share could be as high as 42 percent.  Acting as an agent does not require a dealer to hold inventory and could result in tighter bid-offer spreads due to the dealer’s reduced risk.  However, some customers worry that it may take longer for a dealer to find another customer interested in taking the other side of the trade than it would for a dealer to execute as a principal, exposing the customer to intervening price movements. 
 
Electronic Trading: Electronic trading in the corporate bond market has increased significantly since the financial crisis, and an estimated 20 percent of investment grade corporate bond trading is now performed through electronic interfaces instead of by phone.   Electronic trading has had, and continues to have, a profound effect on market structure across many asset classes and is often associated with higher trading volumes and reductions in bid-offer spreads and trade sizes.  Unlike the growth of high-speed algorithmic electronic trading in Treasury markets highlighted in one of our previous blog posts, so far most of the electronic trading in corporate bonds uses the “Request for Quote” model that is used for phone-based trading.  This type of electronic trading can improve operational efficiency for dealers and customers as well as facilitate price competition by allowing potential buyers or sellers to efficiently gather quotes from multiple potential counterparties, which could reduce observed bid-offer spreads (though the likely effects in the corporate bond market are not as dramatic as in other markets with centralized electronic order books). 
 
New Trading Platforms: New corporate bond trading models and platforms are also being developed.  For example, “all-to-all” trading platforms, which generally permit any market actor—including customers, dealers, and principal trading firms—to participate, offer buyers and sellers the opportunity to transact directly without a dealer acting as an intermediary.  Other platforms focus on new methods of facilitating negotiation and trading protocols designed to protect customer information.  While still a small part of the market (with around 1 percent market share), these new models, if successful, could promote new intermediation patterns that meet the needs of different market participants and reduce reliance on dealer intermediation.  However, these platforms face significant hurdles to widespread adoption, in part due to the large number of corporate securities outstanding and the limited frequency with which many of these securities trade.
 
Given the importance of the corporate bond market to the U.S. economy and the significant transitions underway, we will continue to monitor conditions in both the primary and secondary markets for corporate bonds.


Jake Liebschutz is the Director of the Office of Capital Markets and Brian Smith is a senior policy advisor in the Office of Capital Markets at the U.S. Treasury Department.​​

Posted in:  Fixed Income
Bookmark and Share