Hotchkis & Wiley - The Advantages of Limiting Assets
The investment industry has certainly demonstrated a tendency to repeat the same mistakes. This newsletter will explore one of the most commonly repeated mistakes, particularly in high yield—asset bloat. Allowing assets under management to swell excessively is almost universally acknowledged as detrimental to investors, yet it has become such commonplace that it is often just accepted as standard practice. This newsletter will explore the advantages of limiting high yield assets to responsible levels, or equivalently, the perils of allowing assets to grow uninhibited.
Small Caps Comprise a Large Portion of the High Yield Market
Let’s first dissect the high yield market by size because it has important ramifications for a strategy’s capacity. “Size” refers to the size of the issuer (i.e. the company) as opposed to the size of the issue (i.e. the bond), which is consistent with customary high yield vernacular. For example, Widget Corporation has two forms of debt outstanding: 1) $300 million of senior secured bonds with a 9% coupon due June 2015 and 2) $200 million of senior unsecured bonds with a 12% coupon due June 2017. In this example, the size of the issuer (i.e. Widget Corporation) is $500 million.
Chart 1 decomposes the high yield market, as defined by the BofA Merrill Lynch US High Yield Master II Index, into various size ranges. We generally refer to issuers over $1 billion as “large cap” issuers, those between $500 million and $1 billion as “mid-cap” issuers, and those under $500 million as “small cap” issuers. Our clients have told us they are surprised how many small and mid-cap issuers comprise the market: 540 issuers (54% of all issuers) are small cap, 221 (22%) are mid-cap, and only 234 (24%) are large cap.
[ Enlarge Image ]
Rather than present the high yield index by number of issuers, Chart 2 presents the index by market value (i.e. the larger issuers get a larger weight). By market value, the 234 large cap issuers (from Chart 1) comprise $691 billion, or 69% of the approximately $1 trillion high yield market. Large high yield managers are often limited to this large cap universe because it possesses enough dollars to absorb their massive asset levels. We have learned from our experience that a universe of 234 securities is not sufficient to identify interesting opportunities for genuine credit pickers. These mega managers may decide that amassing more positions is a better alternative, but this is a dilution of resources/best ideas and likely still involves an outsized allocation to the large cap subset.
[ Enlarge Image ]
Small Caps Have a Yield Advantage
Small cap issues have typically exhibited higher yields than large cap issues. We believe there are two primary reasons for this dynamic, one justified and one not. The justifiable reason is that small issues are often less liquid than large issues, so investors demand higher yields to compensate for increased liquidity risk. The unjustifiable reason is that small cap issues are often overlooked by large high yield investors simply because they have massive levels of assets under management. It is difficult to ascertain how much of the yield advantage is due to additional liquidity risk and how much is due to market inefficiency, but we believe that a meaningful portion is due to the latter—a dynamic that we attempt to exploit through our research process.
Chart 3 shows the yield-to-worst for the high yield market divided into quartiles based on issue size. As is typical, smaller issues have higher yields.
[ Enlarge Image ]
More Assets Equals Fewer Opportunities
The idea that excessive asset growth poses challenges for investment managers is a rather intuitive concept. Less intuitive, however, is the actual asset level/range at which investing becomes cumbersome. It is clear to us that putting $500 million to work is easy and putting $500 billion to work is hard, but this range is too large for practical use. To provide practical perspective (i.e. narrow this range) let's walk through a simplified capacity analysis.
First, we need to make two assumptions. The first assumption is the strategy's average position size. We chose 1%, which implies a portfolio of 100 issuers. In our view, a portfolio with considerably more than 100 positions is a dilution of resources and consequently, a dilution of best ideas. The second assumption is the percentage of an issuer that the strategy is willing to own. We chose 20% because our experience has shown us that owning more than 20% of a single issuer often results in liquidity problems. There are exceptions to this of course, but we believe 20% is a reasonable assumption.
We applied these two assumptions in Chart 4, which graphs the percentage of the high yield market that is investable and the percentage that is not investable at various levels of assets under management. As the level of assets managed increases (as you move to the right), the investable percentage of the market decreases. This is one of the tools that we use in estimating the capacity for our high yield product. We believe that we can manage somewhere in the neighborhood of $5 billion while maintaining the integrity of the strategy (the gold line). This is considerably less than the largest actively-managed high yield strategies, which average $17 billion in assets (the red line).
[ Enlarge Image ]
Consider a high yield strategy with $17 billion in assets under management. This manager has three options, all of which can be detrimental to performance.
1) Only choose from the investable portion of the market 2) Shrink the average position size to less than 1% 3) Purchase a larger percentage of the issuer
We believe each of these options poses problems and is damaging to the strategy's investors. Option 1 limits the selection universe, Option 2 dilutes resources/best ideas, and Option 3 increases liquidity risk. We prefer a fourth option: Limit the strategy's assets under management. This is the simplest solution to ensure a sufficient investable universe from which to construct a high conviction portfolio without assuming undue risk.
Asset Growth and Performance: A Case Study
The $17 billion average asset level of the 20 largest high yield strategies translates into roughly $340 billion in total assets, which represents nearly one-third of the entire high yield market. We presumed that good investment performance was an important factor in attracting such sizable asset flows. If this presumption was accurate, we wanted to determine whether good investment performance was sustained as assets ballooned.
In Chart 5, we took the largest 20 high yield strategies as of June 2012 and plotted their asset growth over the past 15 years. The average asset level has grown from approximately $2 billion 15 years ago, to approximately $17 billion today.
[ Enlarge Image ]
In Chart 6, we took the same 20 high yield strategies and plotted their average return, gross of fees, relative to the BofAML US High Yield Master II Index. The chart plots the average 5 year rolling return of the 20 strategies minus the 5 year rolling return of the index. Over the past 15 years the aggregate excess return generated by these managers has dwindled. We cannot make the assertion that the deteriorating performance is due to asset growth, but we find the relationship dubious.
[ Enlarge Image ]
High Yield Exchange-Traded Funds ("ETFs")
Over the past several years, ETFs have become an increasingly popular means of obtaining high yield exposure. Chart 7 demonstrates this popularity by exhibiting high yield ETF net fund flows since December 31, 2008 (inflows minus outflows).
[ Enlarge Image ]
Investors have been attracted to ETFs because they are simple, inexpensive, and a way to gain diversified exposure to high yield. No doubt, these are attractive qualities. We believe, however, that ETFs possess other attributes that warrant careful consideration.
In equity markets, most major indices can be replicated with ease. The market is liquid, transaction costs are minimal, and turnover is modest. This is not the case for many bond markets, particularly high yield. The most commonly followed high yield indices contain numerous illiquid securities, the bid/ask spread can be up to 1% in normal environments (even higher in extreme cases), and bonds enter/exit the indices frequently (calls, tenders, new issues, etc.).
Recognizing these obstacles, ETF providers decided against tracking (or shall we say, attempting to track) the most commonly followed high yield indices. Instead, they track modified indices that only include bonds considered highly liquid. The iShares ETF (ticker: HYG) and the Barclays SPDR ETF (ticker: JNK) comprise over 90% of high yield ETF market. These two ETFs track the iBoxx Liquid High Yield Index and the Barclays High Yield Very Liquid Index, respectively.
The trouble with these tailored indices, in our view, is similar to the issues faced by mega managers—both are ignoring an important segment of the high yield market. Small and mid-cap issuers, which we believe present disproportionate opportunities, are vastly underrepresented. Consequently, ETFs have lagged the broad/commonly-used indices considerably as shown in Chart 8.
[ Enlarge Image ]
To be clear, Chart 8 is comparing ETF performance to broad indices and not to the "liquid" indices that the ETFs actually attempt to track. The objective of Chart 8 is to highlight the potential performance sacrifice from shunning the small and mid-cap segment of the high yield market.
Chart 9 compares the performance of the ETFs to the actual indices they attempt to track. Performance differences have been substantial at times, despite the "easier-to-track" liquid indices. A portion of this deviation can be attributable to the nature of all exchange trades funds, which can trade at a premium or discount to the underlying net asset value. The primary sources of disparity, however, are the challenges of replicating a high yield index that we discussed earlier: liquidity, transaction costs, and turnover.
[ Enlarge Image ]
Active managers can structure the portfolio to manage the ebb and flow of contributions and withdrawals to minimize trading costs. Bid/ask spreads can widen during extreme periods to as much as 4%. ETF managers often have little choice but to trade irrespective of the wide bid/ask spreads. Considering the combination of high trading costs and expense ratios around 0.5%, total ETF fund costs can easily surpass total costs of modestly-priced active managers that trade efficiently.
In the investment management industry, excessive asset growth occurs with astounding recurrence despite overwhelming evidence that it is harmful to investors. Asset bloat limits the selection universe and/or forces a strategy change, both of which can impede performance. In the high yield market, inflated asset levels often preclude a manager from investing in small and mid-cap issuers, which comprise a large and important portion of the market. Small and mid-cap credits typically exhibit higher yields and often present disproportionate opportunities for individual credit selection. Many investors have gravitated toward high yield ETFs as a simple solution, but they too suffer from the same pitfalls as the mega managers. At Hotchkis & Wiley, we are determined to avoid making the same mistake that many in our industry have made over and over. Our approach is absurdly simple and highly pragmatic—commit to limiting assets to a responsible level in the first place.
Mark Hudoff, Portfolio Manager
Ray Kennedy, Portfolio Manager
Patrick Meegan, Portfolio Manager
Ryan Thomes, Portfolio Analyst
All references to high yield based on BofAML US High Yield Master II Index.
Data source(s): Charts 1, 2, 3: Bloomberg; Charts 4, 6: Bloomberg, Wilshire Analytics; Chart 5: Wilshire Analytics; Chart 7: JPMorgan; and Charts 8, 9: Bloomberg. The indices assume reinvestment of dividends and capital gains, and assume no management, custody, transaction or other expenses.
All investments contain risk and may lose value. Investing in high yield securities is subject to certain risks, including market, credit, liquidity, issuer, interest-rate, and inflation risk. Lower-rated and non-rated securities involve greater risk than higher-rated securities. U.S. Treasuries, equities, high yield bonds, and other asset classes have different risk profiles. High yield securities have greater price volatility and credit and liquidity risks (presenting a greater risk of loss to principal and interest) than U.S. Treasuries and other higher-rated securities.
Any discussion or view on a particular company or fund are not investment recommendations, should not be assumed to be profitable, and are subject to change. Managed portfolios and H&W employees may or may not invest in the companies mentioned, and H&W has no obligation to disclose purchases or sales in these securities. Specific company examples do not represent all of the securities purchased, sold, or recommended for advisory clients.
©2012 Hotchkis & Wiley. All rights reserved. Any unauthorized use or disclosure is prohibited. This presentation is circulated for general information only, and does not have regard to the specific investment objectives, financial situation and particular needs of any specific person who may see this report. The research herein is for illustration purposes only. It is not intended to be, and should not be, relied on for investment advice. The opinions expressed are subject to change and any forecasts made cannot be guaranteed. H&W has no obligation to provide revised opinions in the event of changed circumstances. Information obtained from independent sources is considered reliable, but H&W cannot guarantee its accuracy or completeness.