在债券中寻找简单游戏

Looking for Easy

Markets are always evolving. One of the biggest changes in recent years has been the rise of index funds. These passive funds, which include traditional index funds and exchange-traded funds that follow specific rules, have received the majority of flows from investors in the past decade. While active managers continue to play a critical role in gathering information and reflecting it in prices, passive managers are growing their share of the market in stocks and bonds.

The most popular indexes in the U.S. are the S&P 500 for stocks and the Bloomberg Barclays US Aggregate Bond Index for bonds.

Passive funds can mimic the S&P 500’s returns closely and cheaply, but mimicking the returns for the Bloomberg Barclays US Aggregate Bond Index is substantially more challenging because it has 20 times the number of securities, higher turnover, and Games in Bonds

Th e difficulty in matching the Bloomberg Barclays US Aggregate Bond Index spells opportunity for active managers. Indeed, a much higher percentage of bond managers than stock managers beat their passive peers over time. One result is that U.S. active bond managers have received $0.9 trillion in inflows while active equity managers have suffered $1.1 trillion in outflows over the past decade. In this report, Michael reviews the shift from active to passive, discusses the role of skill, takes measure of the opportunities for active bond managers, considers what these developments mean for asset managers, and examines the history of pre-fee excess returns for bond managers. We always welcome your questions and comments.

Andrew Feldstein Chief Investment Officer April 16, 2019

more frequent rebalancing than the S&P 500.

BLUEMOUNTAIN INVESTMENT RESEARCH / April 16, 2019

Looking for Easy Games in Bonds The shift from active to passive management remains one of the hottest Michael J. Mauboussin Director of Research [email protected] issues in the investment industry. In this report we examine the shift from active to passive investing in bonds. The migration in stocks has received more attention than that in bonds because the move has been more dramatic. The transition in bonds has been less rapid because active bond investors have outperformed passive alternatives at a higher rate than equity managers have, even if much of that outperformance can be explained by exposure to risk premiums.

We document the shift by examining trends in assets under management and flows for active and passive taxable bond funds. Next, we describe the challenges that investors face in identifying skillful managers. We turn to methods for measuring easy games, including a comparison of popular bond and stock indexes and a consideration of the participants in the markets.

Finally, we review the implications for asset managers and conclude with a novel framework for gauging the value that active managers generate.

Introduction The growth in passive investing continues to be cumulative outflows in the past ten years, while an important topic for investors to consider. flows into passive funds have approached $2.4 Active managers select securities in an effort to trillion.

generate returns, adjusted for risk, that exceed Active managers in the bond market have not certain targets. Passive managers create fared as badly. The bottom panel of exhibit 1 portfolios that mirror indexes or follow specific reveals that cumulative flows have been positive rules. Examples of rules-based funds include those for both active and passive bond funds. Over the that attempt to capture premiums from past ten years, flows into active bond funds have fundamental factors such as size, value, been roughly $935 billion and flows into passive momentum, and volatility. These are commonly bond funds have been about $960 billion. The called “smart beta” funds. Passive funds can be trend has been clearly more favorable for passive in the form of traditional index mutual funds or managers since 2012.

exchange-traded funds (ETFs).1

The top panel of exhibit 1 shows that active equity funds in the U.S. have suffered $1.1 trillion in

Exhibit 1: Cumulative Flows from Active to Passive in U.S. Equities and Bonds

2,500 Equities 2,500 2,500 Bonds 2,500

2,000 2,000 2,000 2,000

Source: Morningstar Direct.

Note: 2009-2018; Monthly data for long-term, open-end mutual funds and ETFs domiciled in the U.S. that invest in U.S.

equity, sector equity, or international equity (“Equities”) and taxable bonds (“Bonds”).

Exhibit 2 provides a simple explanation for the credit exposure than their benchmark, allowing difference in flows. The percentage of actively- the intermediate-term bond managers to have a managed bond mutual funds that have higher batting average. outperformed their passively-managed peers is Research by the Vanguard Group, one of the much higher than that for equities for holding world’s largest investment advisers, suggests that periods of three or more years through 2018. On an investor can approximate the performance of the surface it is easy to see why investors have active managers benchmarked against the selected active managers for bonds and Bloomberg Barclays US Aggregate Bond Index shunned them for stocks.

with an 80 percent allocation to a total U.S. bond But interpreting this apparent difference in results market index fund and a 20 percent allocation to accurately requires a deeper appreciation of a corporate bond fund with low tracking error true excess returns and exposure to risk factors. and fees.2 That said, it is easy to understand why Most of the outperformance relative to passive an investor who wants to invest in bonds would peers is attributable to traditional risk premiums, select an active rather than a passive fund. including term, corporate credit, and emerging Not all asset classes within the bond universe are markets risk. Intermediate-term bond managers created equally. The dispersion of returns for generally have more credit exposure than their managers of government bonds or mortgage-benchmark, and high-yield managers have less backed securities funds is small, excluding the

Exhibit 2: Percentage of Active Mutual Funds That Outperformed Passive Peers, 2018

100 3 Years 100 10 Years

Percentage Outperforming Percentage Outperforming

Source: Morningstar.

amplifying effect of leverage, whereas the altogether. Active managers generate excess dispersion in returns for managers of emerging- returns by gathering information and reflecting it market or high-yield bonds is large. For instance, in the market. In so doing, they promote price the difference between the 75th and 25th discovery and provide liquidity.3 Active managers percentile managers, measured over three years, make money as they drive prices toward is roughly 5 times larger for emerging market debt efficiency and allow all market participants to get than for long-term government debt. Equity funds in and out of investments at a relatively low cost. are more volatile but the difference in dispersion These are good for society and are a positive between the high- and low-dispersion asset externality. Traditional index funds and ETFs get a classes is only two times. free ride on this externality. The open question is what percentage of the investment Notwithstanding the growth of passive investment management industry needs to be active to in stocks and bonds, it is important to underscore accrue these benefits.

that active management will not disappear

Documenting the Shift The assets under management (AUM) for the U.S. Exhibit 4 shows the flow of money into active and taxable bond mutual fund and ETF universe have passive taxable bond funds from 1993 through gone from about $380 billion in 1993 to roughly 2018. Flows into active funds were positive for 20 $3.8 trillion today (left panel of exhibit 3), and of the 26 years, although 3 of the 6 years of passive funds have gone from close to zero to outflows have occurred since 2012. Passive flows nearly one-third of the AUM (right panel). By way have been positive every year. of comparison, passive U.S. equity funds are now about one-half of all AUM for managed funds.4

Exhibit 3: AUM of U.S. Taxable Bond Mutual Funds and ETFs and Market Share of Passive, 1993-2018

Source: Morningstar Direct.

Exhibit 4: Fund Flows into Actively and Passively Managed U.S. Taxable Bond Funds, 1993-2018 500 Actively Managed 400 Passively Managed

Source: Morningstar Direct.

Exhibit 5 shows how the $3.8 trillion in mutual funds Exhibit 6: AUM of Exchange-Traded Funds by and ETFs breaks down by asset class within Asset Class, 2018 taxable bonds. The largest percentage of the funds, 38 percent, are dedicated to intermediate- U.S. Equity term bond funds followed by 10 percent in short- International Equity term bond funds and 7 percent in high-yield bond Taxable Bond funds.

Sector Equity The first equity ETF was launched in 1993 and the first fixed-income ETF debuted in 2002. There are Commodities

roughly 400 U.S. taxable bond ETFs today. ETFs Alternative trade all day, allow investors to transact through Municipal Bond a broker, and can provide better tax efficiency Allocation than traditional mutual funds. ETFs represent about one-half of total passive bond AUM. While 0 400 800 1,200 1,600 much smaller than equities in absolute dollars, the Billions of U.S. Dollars proportion of ETF assets to the size of total Source: Morningstar Direct.

passively managed assets is similar (see exhibit 6).

The AUM for ETFs that own taxable bonds have Exhibit 7: AUM of Exchange-Traded Funds, U.S.

grown rapidly and are roughly ten times higher Taxable Bonds, 2001-2018 today than a decade ago (see exhibit 7). Passive 700 ETFs represent the bulk of AUM, but active ETFs have increased in recent years. Active ETFs were 600 10 percent of AUM in 2018, up from 0 percent ten

Billions of Dollars 500 years ago. Fixed-income ETFs can have larger 400 tracking error than equity ETFs because the underlying bonds tend to be less liquid than the 300 underlying stocks. In general, the less liquid the 200 assets comprising the ETF, the less accurately the ETF will track the asset value.5 100

0 Exhibit 5: AUM of U.S. Taxable Bond Mutual Funds

Source: Morningstar Direct.

Intermediate-Term Note: Active and passive ETFs.

Other

Short-Term

High-Yield

World

Ultrashort

Multisector

0 300 600 900 1,200 1,500 $ Billions

Source: Morningstar Direct.

Sorting Skill and Luck

Active management is a zero-sum game in the sense that there has to be symmetry between managers who generate positive excess returns and those who generate negative excess returns.

Investors face the challenge of identifying and selecting skillful managers.

In fact, there are really two issues. The first is judging which managers are likely to deliver attractive returns and the other is assessing the attractiveness of the opportunity to express skill. A common way to measure skill is to look at persistence. Results are persistent in activities where outcomes are due predominantly to skill rather than luck. Think of a world-class musician.

She will play beautifully night in and night out. The quality of the performance is lasting. Results are more random when luck primarily determines outcomes. Yesterday’s lottery winner has no better a chance to win tomorrow.

Consistent with the message from exhibit 2, the persistence in performance for bond mutual fund managers tends to be higher than that for equity mutual fund managers. 6 This means that bond funds that outperform the benchmark tend to do so with some consistency.

There are a couple ways to explain that result. It may be the case that there is a lot of variety in the skill of bond managers and that the better ones outperform the worse ones. It may also be the case that measuring relative results is trickier than it appears. For example, managers may have exposure to a risk factor that a simple return figure fails to reflect and therefore the outperformance is really compensation for risk.

The variation of skill is a crucial consideration in any competitive domain. One useful concept is the “paradox of skill,” which says that in activities where both skill and luck contribute to outcomes, as skill increases, luck can play a bigger role in determining results. 7 The key to the idea is to recognize that you can measure absolute and relative skill. In nearly all competitive domains, absolute skill has improved. Consider sports measured against a clock, such as running or swimming. Athletes today are faster than their predecessors as the result of advancements in technique, nutrition, coaching, and training.

At the same time that absolute skill has improved, relative skill has narrowed. That means that in a field of competition, the difference between the best and the average player is smaller today than it was in the past. As the level of skill becomes more uniform, the role of luck rises.

The paradox of skill is a big deal in investing. That manager results appear close to random is not because professional investors don’t know what they are doing. It’s actually the opposite.

Investors today are so good that asset prices do an effective, albeit imperfect, job of reflecting relevant information. 8 One way to measure the paradox of skill over time is to examine changes in the standard deviation of results. Think of it this way: the results for many activities follow a distribution that looks like a bell. Lots of participants are near the average at the top of the bell. The sides that slope down from the top represent the under-and outperformers. Standard deviation measures the width of the bell. A shrinking standard deviation reveals the paradox of skill, as it documents the gap between the average and the extreme performers.

The investment management industry measures results through excess returns adjusted for risk relative to an appropriate benchmark, or alpha.

The distribution of alpha doesn’t follow a perfect bell-shaped curve, but reality is close enough to make the point. Skillful active managers want wide distributions so they can earn positive alpha while less skillful investors suffer negative alpha. Exhibit 8 shows the three-year average standard deviation of single-factor alpha for U.S. large capitalization equity mutual funds from 1973 through 2018. There is a steady decline in standard deviation, save for a brief but explosive expansion during the dot-com bubble. That noted, the standard deviation has rebounded modestly since the financial crisis. In 2018, the standard deviation of alpha was 5.1 percent for more than 2,000 equity mutual funds.

Exhibit 8: Standard Deviation of Alpha for U.S. Large Capitalization Equity Mutual Funds

14 13 12

of Alpha (Percent)

Source: Morningstar Direct.

Note: Rolling three-year average.

Exhibit 9 shows the three-year average standard key for both. But bond markets have more deviation of alpha for U.S. taxable bond mutual securities, are more costly to trade, and funds from 1978 through 2018. The standard experience greater turnover. As a consequence, deviation has bounced around, but unlike the it comes as little surprise that fewer equity equity chart there is no clear trend. While bond managers than fixed income managers managers have certainly become more skillful outperform their respective benchmarks. This is over time, the relative skill gap between the best consistent with the more rapid transition into and worst performers has not narrowed as it has passive funds in the equity markets. for equity managers. In 2018, the standard Generating positive alpha requires applying skill deviation of alpha was 3.3 percent for nearly through a robust process and an ability to find 1,300 U.S. taxable bond mutual funds.

inefficiently-priced assets. Skilled poker players All markets are highly competitive, but the describe the search for easy games, where they persistence and standard deviation of alpha can be confident that their skill will prevail over suggest that the opportunity to find managers time. The same is true of markets. You know you who may outperform is greater in bonds than in are playing an easy game when you can clearly stocks. Investment process and time horizon are identify your source of edge.

Exhibit 9: Standard Deviation of Alpha for U.S. Taxable Bond Mutual Funds

of Alpha (Percent)

Source: Morningstar Direct.

Note: Rolling three-year average.

Measures of Easy Games In assessing easy games, you consider the game for frequent rebalancing, which can be a itself and the nature of the other participants. The challenge because the provision of liquidity has game provides some context for gauging how become more expensive since the financial crisis active managers can perform relative to an in 2008.9 Further, pricing all of the securities index fund or ETF that tracks a benchmark. Part of accurately in real time is a challenge.10 the reason that active bond managers succeed in outperforming a benchmark at a higher rate The S&P 500 index is market-capitalization than equity managers is the nature of the two weighted and has 505 securities. The information markets.

technology, health care, and financials sectors The Nature of the Game: Bonds versus Stocks. The represent about one-half of the index’s market Bloomberg Barclays US Aggregate Bond Index is capitalization. Turnover is relatively infrequent market-value weighted and has more than and has averaged about four percent per year 10,300 securities. U.S. Treasuries and mortgage- over the last six decades. The main reason for backed securities (MBS) make up about two- companies to exit the S&P 500 is that they are thirds of the index, and corporate bonds acquired. Exhibit 10 compares the two leading represent another 25 percent. The maturity and benchmarks.

issuance of securities means there is a need

Exhibit 10: Comparison of Leading Bond and Equity Indexes

Bloomberg Barclays US Characteristic Aggregate Bond Index S&P 500 Index Asset Class Investment grade taxable U.S. bonds Large-capitalization U.S. equities

MBS (28%) Health Care (15%)

Industrial – Corporate (15%) Financials (13%)

Financial – Corporate (8%) Consumer Discretionary (10%) Agency (3%) Communication Services (10%) Commercial MBS (2%) Industrials (9%)

Utility – Corporate (2%) Consumer Staples (7%)

Supranational (1%) Energy (5%)

Sovereign (1%) Utilities (3%)

Local Authority (1%) Real Estate (3%)

Asset-Backed Securities (1%) Materials (3%)

Source: Bloomberg; S&P Dow Jones Indices; Federal Reserve; Kenneth R. French, “Presidential Address: The Cost of Active Investing,” Journal of Finance, Vol. 63, No. 4, August 2008, 1537-1573; and BlueMountain estimates.

Note: Number of securities as of 3/31/2019; ownership as of 12/31/2018; turnover is the 5-year average for 2014-2018 and for the Bloomberg Barclays index is measured as the lower of the entering turnover and exiting turnover; annualized volatility for 1989-2018; sector composition as of 3/31/2019.

To show how indexing to these benchmarks

differs in practice, we can compare the iShares Core U.S. Aggregate Bond ETF (AGG) and the iShares Core S&P 500 ETF (IVV). AGG had $57 billion in AUM and IVV had $149 billion at year-end 2018. We look at three metrics for each:

percent of index securities held, active share, and tracking error.

AGG owns 60 percent of the securities in the Bloomberg Barclays US Aggregate Bond Index.

The leading ETFs that track the index hold between one-quarter and three-quarters of the securities in the index. Passive bond managers must decide which securities to own. IVV holds 100 percent of its benchmark, as do all large equity index funds.

Active share reflects how much a fund differs from its index. 11 Active share is zero percent if a fund exactly matches the index and 100 percent if it is completely different. The calculation assumes no leverage or short sales. (See appendix for more detail.) The active share is 53 percent for AGG and 1 percent for IVV. A passive bond ETF is less similar to its benchmark than a comparable stock ETF.

Tracking error, the standard deviation of the difference between the returns of a portfolio and the index, is a measure of how closely a portfolio matches the index. AGG’s tracking error is four basis points, while IVV’s is one basis point. Further, the annualized volatility for the Bloomberg Barclays US Aggregate Bond Index is lower on average than that for the S&P 500 index, so the tracking error as a percentage of volatility is much higher for AGG than IVV.

Exhibit 11 summarizes the differences between AGG and IVV and clarifies why equity indexing is more straightforward than bond indexing.

Exhibit 11: Tale of the Tape: AGG versus IVV

Metric AGG IVV Percent of index securities held 60% 100% Active share 53% 1% Tracking error (basis points) 4 1 Source: Morningstar Direct and ActiveShare.info.

Note: Percent of index securities held is through 3/26/2019; Active share as of 3/26/2019 for AGG and 12/31/2016 for IVV; Tracking error based on monthly returns annualized over three years ended 2/28/2019. Exhibit 12: Tracking Error for Index Funds and ETFs Benchmarked to Various Indexes Tracking Error Category (Basis Points) S&P 500 2 Short-Term Government 5 Intermediate Government 9 Mortgage-Backed 18 Inflation-Protected 20 Intermediate Investment Grade 21 Emerging Markets 23 High Yield 47 Source: Morningstar Direct.

Note: Monthly returns annualized over three years ended 2/28/2019; Reflects average of 5 largest funds for each category except for mortgage-backed, which includes 3 funds based on data availability.

We selected AGG because it is large and can be reasonably compared to the Bloomberg Barclays US Aggregate Bond Index. Exhibit 12 shows that the average tracking error for passive bond funds ranges from 5 basis points for short-term government funds to 47 basis points for high yield funds. In each case, the tracking error is calculated using the benchmark noted in the fund prospectus.

We have seen that it is difficult to build a fund that closely tracks the Bloomberg Barclays US Aggregate Bond Index. This difficulty creates opportunity. Refinancing, maturities, and issuance present ways for active managers to beat the benchmark. This ability is inherently more limited for active equity managers.

For example, the issuance of new securities is far more active in the bond market than in the stock market, creating a potential source of alpha for bond managers. There are more new bond issues because bonds mature whereas equities are perpetual. Institutional bond investors purchase new issues from bank underwriters, often at a discount to similar secondary transactions.

This new-issue yield premium compensates investors in corporate debt for bearing the uncertainty of what the eventual market price will be, as well as other risks. New issues equaled 17 percent of total debt outstanding for the corporate debt market in the decade ended 2018. But research suggests bond investors earn returns, or alpha, that these risks do not fully explain. One argument is that banks underprice new issues to benefit institutional investors. This effectively transfers wealth from issuers to investors. 12

The Nature of the Participants. Another aspect to to do better than individuals, reflecting greater assessing how hard it is to outperform a sophistication and the ability to pay lower costs. benchmark is a consideration of the other The model also reiterates why “searching participants. Nicolae Gârleanu and Lasse investors” can earn higher returns in markets that Pedersen, professors of finance, created a model are less efficient. A recent survey of returns to to reflect the reality that for there to be winners, active management found “that the value of there must be losers (see exhibit 13).13 The core active management depends on the efficiency idea is that some investors can gather information of the underlying market and the sophistication of that is relevant for discerning value, buy and sell the investor.”14 Returns to active managers accordingly, and profit from mispricing. These are depend on the composition of the market in the informed asset managers.

which they compete, and the bond and stock For some managers to outperform, others must markets are very different. underperform. These are the noise traders and Both the poker metaphor and the Gârleanu and noise allocators, who include retail investors and Pedersen model illustrate that part of finding an some issuers. Passive investors neither win nor lose.

easy game is understanding the other To the extent that noise traders transition to be participants. Expressing skill is easier when other passive investors, active management can get investors are less sophisticated or have even more difficult because the weak players motivations to trade that are not based solely on have left the game.

fundamentals. Exhibit 14 shows the composition The model shows why it makes sense for of ownership for U.S. Treasury securities, agency-unsophisticated investors to participate in most and government-sponsored enterprise (GSE)- markets via index funds or ETFs. Gârleanu and backed securities, and corporate bonds since Pedersen also find that institutional investors tend 1973, the year that what is now the Bloomberg Barclays US Aggregate Bond Index was created.

Exhibit 13: A Model of Efficiently Inefficient Markets

Searching Searching Noise Noise investors: investors:

allocators traders passive active

Search for informed Random managers allocations Active asset Active asset managers: managers:

informed uninformed

Uninformed Informed Uninformed Random trading trading trading trading

Security market

Source: Nicolae Gârleanu and Lasse Heje Pedersen, “Efficiently Inefficient Markets for Assets and Asset Management,” Journal of Finance, Vol. 73, No. 4, August 2018, 1663-1712.

Exhibit 14: Ownership of U.S. Treasuries, Agency and GSE-Backed Securities, and Corporate Bonds, 1973-2018

Treasury Securities 100% Other State/Local Governments 90% Federal Government 80% Banks/Insurance Co's/

40% Households/Nonprofits

30%

20% Rest of World

10%

0%

Agency- and GSE-Backed Securities 100% Other 90% GSE's

50% U.S.-Chartered

100% Other Financial Institutions Banks/Credit Unions 90% 80% Insurance Companies

70% Pension/Retirement Funds 60% 50% Mutual Funds/ETFs

Source: Federal Reserve.

Note: GSE: government-sponsored enterprises.

There are a couple of observations that are  The pattern of fund flows is different in relevant for a consideration of passive and active equities versus bonds, as we saw in exhibit 1. bond management: Active U.S. equity managers have realized outflows of $1.1 trillion in the past ten years,  Some large holders of U.S. Treasury and whereas active U.S. taxable bond managers agency- and GSE-backed securities may have seen inflows of $935 billion. While flows have incentives unrelated to the value of into passive bond funds have exceeded the bonds solely. For instance, the Federal flows into active bond funds, the story is not Reserve buys bonds to achieve quantitative as one-sided as in equities.17 Flows are easing and non-U.S. central banks buy important in investing because they create bonds to stabilize exchange rates.15 This is a tailwind for funds doing well and a important because foreign ownership of U.S. headwind for those performing poorly.

Treasury and agency- and GSE-backed Investors generally chase past performance securities has increased substantially in by investing in funds that have done well. recent decades.

Portfolio managers who receive inflows Further, banks are required to hold high-commonly purchase more of the securities quality liquid assets to ensure liquidity. they already own, boosting the Because the Bloomberg Barclays US performance of the portfolo.18 Aggregate Bond Index largely comprises Underperforming managers who suffer U.S. Treasuries and agency- and GSE-outflows must sell what they hold, backed securities, these non-economic compounding performance woes. Flows are owners can influence the price of the a short-term predictor of results.19 securities and hence the index’s returns.

These elements clarify the nature of the game  Rules or regulations constrain the bond active bond managers play as they seek to holdings of certain institutions.16 For deliver returns in excess of a benchmark. First, the example, an insurance company that owns Bloomberg Barclays US Aggregate Bond Index is an investment grade bond may need to sell markedly different than the S&P 500 in terms of it if it is downgraded to high yield. This leads the number of securities, liquidity, pricing, and to a lower price on the security than the pace of rebalancing. Active managers can more fundamentals justify. In both cases, active readily distinguish themselves from the managers can position themselves to benchmark in a way that improves their relative sidestep these challenges or take performance. Exhibit 15 summarizes some sources advantage of them. This makes it easier to of alpha for active bond managers.

beat the benchmark.

Exhibit 15: Sources of Alpha for Active Bond Managers

Source of Alpha Mechanism Nature of market Bond markets have more securities and turnover and are more costly to trade. Taking other side of forced sales or exploiting constraints of other parties:

• Central banks: pursue quantitative easing and stabilization of Taking other side of exchange rates;

non-economic trades

• Banks: required to hold high-quality liquid assets;

• Insurance companies: potential forced selling of downgraded bonds. Liquidity provider Active managers compensated for providing liquidity to index changes. Negotiation Trading over-the-counter versus on exchanges allows for negotiation.

Dispersion in spreads over Treasuries can present an opportunity as bonds with Dispersion in spreads the same credit rating may offer different yields.

New issue premium/wealth transfer Active managers can take advantage of underpricing at issuance.

There can be hundreds of bonds for a single issuer versus usually only one Relative value opportunities stock. This variety may present attractive relative value opportunities within the capital structure.

Source: BlueMountain Capital Management.

What It Means for Asset Managers Economies of scale are important for passive Exhibit 17 shows the trend in asset-weighted managers because the fees on passive funds are expense ratios in the U.S. taxable bond market for a fraction of those on active funds. Exhibit 16 active mutual funds, index funds, and index ETFs, shows the fifteen firms with the highest share of as well as the weighted average for all mutual passive AUM for U.S. taxable bond funds. funds. The expense ratio on index ETFs is about 10 Vanguard dominates the industry, with a market basis points higher than that for traditional index share of roughly 55 percent, followed by funds. Active expense ratios remain roughly 40 BlackRock (iShares), Fidelity, and State Street. The basis points higher than passive ones. same four investment managers have the largest Passive investors can pay lower expenses share in traditional index funds and ETFs for because they benefit from the positive equities, albeit in a different order.

externalities active managers create, including Expense ratios have been trending down in the price discovery and liquidity. Expense ratios on all investment management industry, reflecting the funds have declined roughly 35 basis points in the cost-saving impact of technology and the shift past two decades.

from active to passive AUM.20

Exhibit 16: Largest Passive U.S. Taxable Bond Fund Managers, 2018

Source: Morningstar Direct.

Note: Mutual Funds and ETFs.

Exhibit 17: Expense Ratios on U.S. Taxable Bond Mutual Funds and ETFs, 1997-2018

Source: James Duvall, “Trends in the Expenses and Fees of Funds, 2018,” ICI Research Perspective, Vol. 25, No. 1, March 2019.

Winning the Easy Game (Value Added by Active Managers )

Markets cannot be perfectly efficient, where price and value are the same, because there is a cost to gather information and reflect it in prices.

To compensate investors for assuming this cost, there must be a profit opportunity in the form of excess return. In a competitive market, it would be reasonable to assume that cost and benefit are in rough parity.

The trick is to quantify cost and benefit. The amount of fees that investors pay to active managers is one approach to quantifying cost, and available positive alpha captures benefit. While much of the investment industry reports results using percentage returns versus a benchmark, the crucial underlying question is how many dollars in excess returns a fund creates.

Jonathan Berk and Richard Green, professors of economics, created a framework to consider excess returns that captures both cost and benefit. 21 They assume that investors seek to give their money to funds that will generate excess returns but that the more money a skillful manager runs, the lower is his or her expected alpha. Their setup suggests that skillful managers end up with more money and expected alpha moves toward zero as a fund grows.

The Berk and Green framework is similar to economic profit, or a company’s economic return after considering the opportunity cost of capital. 22 Economic profit equals a firm’s return on invested capital (ROIC) minus its weighted average cost of capital (WACC) times its invested capital. For example, a firm with a 13 percent ROIC, an 8 percent WACC, and $1,000,000 in invested capital has an economic profit of $50,000 ([0.13 – 0.08] x $1,000,000).

The same analysis applies to investment funds. A fund’s gross economic profit equals its beta-adjusted gross return minus the benchmark return, multiplied by AUM. We call gross economic profit “gross profit” for short. For instance, a bond fund with average risk that has a gross return of 9 percent, a benchmark return of 4 percent, and $100,000,000 of AUM would produce a gross profit of $5,000,000 ([0.09 – 0.04] x $100,000,000). Five million dollars is the value the investment manager extracts from the market.

Berk and Green find that gross profit is a better reflection of skill than simple measures of return.

Here’s an example. Peter Lynch, who managed the Magellan Fund at Fidelity Investments from May 1977 to May 1990, generated an average annual return of more than 29 percent per annum. Alpha was large but the AUM was small in his first five years running the fund. Lynch’s average monthly gross profit was $2.1 million in 2018 U.S. dollars.

Positive alpha was small but the AUM was large in the last five years he ran the fund. Average monthly gross profit was roughly $38 million. Even though Lynch’s alpha was vastly larger in the early years than in the late years, he extracted nearly 20 times more value in his last 5 years as a result of the large increase in AUM. 23 Results for Bill Gross, the famous bond investor who spent the bulk of his career running the Total Return Fund at PIMCO, provide a noteworthy comparison. The average monthly spread between the fund’s risk-adjusted gross return and the benchmark return, gross alpha, was 0.10 percent over Gross’s first five years running the fund. His skill allowed him to extract about $1.9 million per month in 2018 U.S. dollars from the market.

In his last five years ended in September 2014, the average monthly spread was 0.18 percent, and the average fund AUM was $271 billion. Gross’s monthly value extraction was roughly $500 million. The magnitude of his alpha in the later years was roughly double what it was in the early years, leading to value extraction that was roughly 270 times larger because of the substantial increase in AUM.

We calculate gross profit before fees to accurately measure an investment manager’s skill. 24 Fees finance the manager’s ability to express skill by unearthing information and pricing it in securities. But fees also eat into investor returns. How best to split the fruits of skill between managers and investors is a topic of debate.

We use the Berk and Green calculation to analyze active mutual funds in the Morningstar Direct database that invest in U.S. taxable bonds from 1978 through 2018. Over the full period the data capture roughly 23,400 fund years and $35 trillion of AUM. The database is free of survivorship bias. We show all figures in 2018 U.S. dollars to reflect inflation.

The number of funds included in the analysis fund industry has large skewness. To illustrate, the increased from about 25 in 1978 to nearly 1,200 top 20 percent of bond funds manage just under today. The average fund size was roughly $300 85 percent of total AUM and the top 50 percent million in the late 1970s, peaked at $2.23 billion in control 97 percent of AUM. 2013, and is currently $2.17 billion. See exhibit 18.

The gross profit for the middle of the distribution of We calculate gross profit using alpha that is all funds, or the median, was $103,000 per month based on the capital asset pricing model. and $1.2 million per year. The median measures Research suggests the use of single-factor alpha the central tendency of a distribution with is reasonable given how investors behave.25 The skewness better than the average does. That the risk of the funds is measured relative to the average and the median gross profit are positive benchmark the funds specify in their prospectus. reveals mutual fund managers generated returns in excess of the market, after accounting for risk The average fund’s gross profit was $1.4 million but before fees. Exhibit 19 shows the frequency per month and $16.6 million per year over the distribution of gross profit for roughly 23,400 fund period we measure. The mean is misleading years from 1978-2018.

because the distribution of AUM in the mutual

Exhibit 18: Total Number of Funds and Average Fund Assets Under Management

2.5 Average Fund AUM Number of Funds 1,400

1,200

Average Fund AUM 2.0 1,000

Number of Funds 1.5 800

(2018 U.S. Dollars, Billions)

Source: Morningstar Direct.

Exhibit 19: Frequency Distribution of Annual Gross Profit, 1978-2018

12,000

10,000

8,000

Annual Gross Profit (2018 U.S. Dollars, Millions)

Source: Morningstar Direct.

Exhibit 20 documents the positive and negative of the less skillful. The financial crisis of 2008 gross profits by year from 1978 through 2018. Over marked the beginning of an era of larger positive the 41 years, the aggregate positive gross profit and negative gross profit figures. was about $660 billion and the aggregate Looking at the dollars of gross profit is informative, negative gross profit was roughly $270 billion, but the critical concept is opportunity compared netting to total gross profit of roughly $390 billion.

to the AUM. Over the full 41 years, the $390 billion This is the amount of value that bond mutual fund in gross profit was extracted from $35 trillion in managers wrested from the market before fees.

AUM. Thus the average gross profit was 1.1 Exhibit 21 smooths the short-term noise by percent of AUM. This exceeds the fees the funds showing the trailing five-year positive and charged during that period, reinforcing the negative gross profit. Investors with skill prefer notion that active bond funds offer an attractive periods with large quantities of positive and alternative to index funds or ETFs. For equity negative gross profit. This dispersion means there mutual funds over the same period, the gross are big winners and losers, which provides an profits roughly equaled the fees. opportunity for the skillful to thrive at the expense

Exhibit 20: Total Gross Profit (Annual), 1978-2018

150 125 100

2018 U.S. Dollars (Billions)

Source: Morningstar Direct.

Exhibit 21: Total Gross Profit (Sum of Trailing 5 Years), 1982-2018

320 280 240

2018 U.S. Dollars (Billions)

-80 Negative Gross Profit

-120

Source: Morningstar Direct.

Exhibit 22 traces the trailing 5-year gross profit and calculate the gross alpha and gross profit for yield, defined as gross profit divided by AUM, each. We then take the average of each of the from 1982-2018. The yield bounces around, with deciles over the 19 years. Most of the action is on upturns in the early 1980s, early 2000s, and the 5 the right side of the exhibit, as the largest five years after the financial crisis. The series declined deciles control the vast majority of the AUM. from the mid-1980s through the mid-1990s, and in A pair of patterns are visible. The first is a decline the past five years after peaking in 2013.

in annual gross alpha as you go from the first to Because the Berk and Green calculation is the eighth decile. Gross alpha then rises again for weighted for AUM, it captures a nuance of the largest funds. In general, the opportunity set performance assessment that is often absent in shrinks as funds get larger because of ownership the headlines about fund results. Exhibit 23 shows and liquidity constraints.26 But the reversal in the the relationship between fund size, gross alpha, trend for large funds suggests some sources of and gross profit. For each year from 2000 through edge. These include superior allocations to new 2018, we break the active U.S. taxable bond issues, better access to information, and more mutual fund industry into deciles based on AUM flexible mandates.

Exhibit 22: Gross Profit Yield (Trailing Five Years), 1982-2018

3

Gross Profit Yield 2

1

Trailing Five Years (Percent)

0

-1

Source: Morningstar Direct.

Exhibit 23: Mutual Fund Size, Gross Alpha, and Gross Profit, 2000-2018

Gross Alpha Gross Profit 1.6 125

Average Annual Gross Alpha (Percent)

1.4

(2018 U.S. Dollars, Millions)

Decile (Smallest to Largest AUM)

Source: Morningstar Direct.

The second pattern is a rise in annual gross profits as fund size increases. The jump is particularly large from the ninth to the tenth decile. The average annual gross profit for the largest 10 percent of funds is roughly $120 million per year. The next largest 10 percent, which today have 15 percent of the assets, is approximately $20 million per year. The smaller funds do well but play for lesser stakes. The larger funds have more modest gross alpha but play for higher stakes and hence extract more alpha from the market. This is especially true for the largest funds.

More often than not, headlines broadcasting the poor record of active managers rely on results where each fund has an equal weight. But if the Berk and Green framework is correct and more skillful investors manage more money than less

Exhibit 24: Equal-Weighted versus Asset-Weighted Returns

Investment-Grade Intermediate 5 Equal-Weighted Asset-Weighted (Percent) 4 Benchmark

Source: Aye M. Soe, Berlinda Liu, and Hamish Preston, “SPIVA® Indices Research, March 11, 2019.

skillful investors, returns weighted by AUM will provide a clearer picture of economic results.

Exhibit 24 reveals that asset-weighted returns are consistently higher than equal-weighted returns for investment-grade and high-yield bond mutual funds. For example, for the 10 years ended December 31, 2018, the asset-weighted returns for all investment-grade intermediate bond funds was 4.3 percent, roughly 70 basis points higher than equal-weighted returns.

Fees are generally lower for larger funds than for smaller ones, which is part of the reason that larger funds outperform smaller ones. But it is likely that differential skill also contributes to the gap between asset-weighted and equal-weighted returns.

High Yield 12 Equal-Weighted 10 Asset-Weighted (Percent) Benchmark

U.S. Scorecard: Year-End 2018,” S&P Dow Jones

Conclusion

This report explored the shift from active to passive investments in the bond market. Active investors have fared much better in bonds than in equities, in part because they have outperformed passive alternatives at a much higher rate than equity managers have. While a large part of that outperformance can be explained by exposure to traditional risk premiums, the fact remains that fund investors have realized higher returns owning active funds than they would have owning passive funds.

While there has been a large move into passive funds, active managers continue to contribute to price discovery and liquidity, which are vital. But active managers charge higher fees than do passive vehicles, including traditional index funds and ETFs. Passive managers charge lower fees because they rely on active managers to make markets efficient.

Mimicking indexes in bonds is not as straightforward as it is in stocks. One reason is that the most popular bond index, the Bloomberg Barclays US Aggregate Bond Index, includes more than 10,000 securities versus roughly 500 for the S&P 500 index. As a result, it is very difficult to create a passive fund that tracks the index with negligible tracking error.

One crucial question in investing is how much opportunity there is for active managers to capture alpha. We describe a framework, based on the work of professors Jonathan Berk and Richard Green, to gauge the value that active managers generate. We find that over the past four decades, cumulative gross profit and fees are roughly equal in equities but that gross profit exceeds fees in bonds. This suggests investors in bond funds have extracted more value from the market than equity fund investors have.

Gross profit allows us to get away from the misleading results of equal-weighted mutual fund returns and provides figures for both the cost and opportunity for active investment management.

Appendix: Notes on Methodology

Active Share. Active share reflects how much a fund differs from its index. Active share is zero percent if a fund exactly matches the index and 100 percent if it is completely different. This assumes no leverage or short sales. We calculate active share as 100 percent less the sum of the overlapping weights between the holdings of the fund and the index.

Here is a simple example for an equity fund (see exhibit 25). The fund holds four stocks (A, B, C, and D), and the index holds four stocks (A, B, C, and E). We see that A, B, and C are overlapping positions and D and E are not.

Exhibit 25: Active Share Example

Holdings in Either Overlapping

Active Share = 100% Minus Overlap = 20% Source: ActiveShare.info; Martijn Cremers, “Active Share and the Three Pillars of Active Management: Skill, Conviction, and Opportunity,” Financial Analysts Journal, Vol. 73, No. 2, Second Quarter 2017, 61-79.

For overlapping positions (A, B, and C), the overlapping weight is the lower of the two weights between the fund and the index. For non-overlapping positions (D and E), there is 0 percent overlapping weight. We sum the overlapping weights (80 percent) and subtract that sum from 100 percent to arrive at active share (20 percent). Therefore the more the holdings overlap, the lower is the active share.

Gross Profit. To calculate a fund’s gross profit, we use the alpha of the oldest share class and an aggregated size based on all share classes. The annual gross profit sums only include funds that have a gross alpha figure for a full calendar year. Based on data availability, our gross profit calculations for Peter Lynch and Bill Gross do not perfectly align with the months in which their tenures began.

 Lynch tenure: May 1977-May 1990; Analysis:

June 1977-May 1990;

 Gross tenure: May 1987-September 2014;

Analysis: October 1987-September 2014.

Endnotes 1 An exchange-traded fund (ETF) is an investment fund that trades on an exchange just as a stock

does.

2 “The Illusion of Active Fixed Income Alpha,” AQR Alternative Thinking, Fourth Quarter, 2018; Jordan

Brooks, Diogo Palhares, and Scott Richardson, “Style Investing in Fixed Income,” Journal of Portfolio Management, Vol. 44, No. 4, Special Issue 2018, 127-139; Gjergji Cici, Scott Gibson, Rabih Moussawi, “Explaining and Benchmarking Corporate Bond Returns,” SSRN Working Paper, June 2017; Paul M.

Bosse, Brian R. Wimmer, and Christopher B. Philips, “Active bond-fund excess returns: Is it alpha . . . or beta? The Vanguard Group, September 2013; and Richard Dewey and Aaron Brown, “Bill Gross’ Alpha: The King Versus the Oracle,” SSRN Working Paper, March 2, 2019.

3 Maureen O’Hara, “Presidential Address: Liquidity and Price Discovery,” Journal of Finance, Vol. 58,

No. 4, August 2003, 1335-1354 and Ekkehart Boehmer and Eric K. Kelley, “Institutional Investors and the Informational Efficiency of Prices,” Review of Financial Studies, Vol. 22, No. 9, September 2009, 3563- 3594.

4 Charles Stein, “Shift From Active to Passive Approaches Tipping Point in 2019,” Bloomberg, December

31, 2018.

5 Antti Petajisto, “Inefficiencies in the Pricing of Exchange-Traded Funds, Financial Analysts Journal, First

Quarter 2017, 24-54.

6 William G. Droms and David A. Walker, “Performance Persistence of Fixed Income Mutual Funds,”

Journal of Economics and Finance, Vol. 30, No. 9, Fall 2006, 347-355 and Aye M. Soe and Ryan Poirier, “Does Past Performance Matter? The Persistence Scorecard,” S&P Dow Jones Indices Research, January 2018.

7 Michael J. Mauboussin, The Success Equation: Untangling Skill and Luck in Business, Sports, and

Investing (Boston, MA: Harvard Business Review Press, 2012), 53-58.

8 Michael J. Mauboussin, “Who Is On the Other Side? You Need Good BAIT to Land a Winner,”

BlueMountain Investment Research, February 12, 2019.

9 Jens Dick-Nielsen and Marco Rossi, “The Cost of Immediacy for Corporate Bonds,” Review of Financial

Studies, Vol. 32, No. 1, January 2019, 1-41 and Lasse Heje Pedersen, “Sharpening the Arithmetic of Active Management,” Financial Analysts Journal, Vol. 74, No. 1, First Quarter 2018, 21-36. 10 Adriana Z. Robertson and Matthew Spiegel, “Better Bond Indices and Liquidity Gaming the Rest,”

SSRN Working Paper, November 15, 2017.

11 K. J. Martijn Cremers and Antti Petajisto, “How Active Is Your Fund Manager? A New Measure That

Predicts Performance,” Review of Financial Studies, Vol. 22, No. 9, September 2009, 3329-3365; Antti Petajisto, “Active Share and Mutual Fund Performance,” Financial Analysts Journal, Vol. 69, No. 4, July/August 2013, 73-93; and Martijn Cremers, “Active Share and the Three Pillars of Active Management: Skill, Conviction, and Opportunity,” Financial Analysts Journal, Vol. 73, No. 2, Second Quarter 2017, 61-79.

12 Robert S. Goldberg and Ehud I. Ronn, “Quantifying and Explaining the New-Issue Premium in the

Post-Glass-Steagall Corporate Bond Market,” Journal of Fixed Income, Summer 2013, 43-55. 13 Nicolae Gârleanu and Lasse Heje Pedersen, “Efficiently Inefficient Markets for Assets and Asset

Management,” Journal of Finance, Vol. 73, No. 4, August 2018, 1663-1712.

14 Alexander Dyck, Karl V. Lins, and Lukasz Pomorski, “Does Active Management Pay? New

International Evidence,” Review of Asset Pricing Studies, Vol. 3, No. 2, December 2013, 200-228. 15 Francis E. Warnock and Veronica Cacdac Warnock, “International Capital Flows and U.S. Interest

Rates,” NBER Working Paper 12560, October 2006.

16 Andrew Ellul, Chotibhak Jotikasthira, and Christian T. Lundblad, “Regulatory Pressure and Fire Sales in

the Corporate Bond Market,” Journal of Financial Economics, Vol. 101, No. 3, September 2011, 596-620; Vikram Nanda, Wei Wu, and Xing Zhou, “Investment Commonality across Insurance Companies: Fire Sale Risk and Corporate Yield Spreads,” Finance and Economics Discussion Series 2017-069, Washington: Board of Governors of the Federal Reserve System, June 2017; Matthew Spiegel and Laura Starks, “Institutional Rigidities and Bond Returns around Rating Changes,” Working Paper, November 19, 2016; and Giovanni Cespa and Thierry Foucault, “Illiquidity Contagion and Liquidity Crashes,” Review of Financial Studies, Vol. 27, No. 6, June 2014, 1615-1660.

17 Eric Balchunas and Athanasios Psarofagis, “Bond Managers Sidestep Active Outflows,” Bloomberg

Intelligence, January 18, 2018.

18 Yong Chen and Nan Qin, “The Behavior of Investor Flows in Corporate Bond Mutual Funds,”

Management Science, Vol. 63, No. 5, May 2017, 1271-1656 and Jimmy Shek, Ilhyock Shim, and Hyun Song Shin, “Investor Redemptions and Fund Manager Sales of Emerging Market Bonds: How Are They Related?” Review of Finance, Vol. 22, No. 1, February 2018, 207-241.

19Joshua Coval and Erik Stafford, “Asset Fire Sales (and Purchases) in Equity Markets,” Journal of

Financial Economics, Vol. 86, No. 2, November 2007, 479-512.

20 Patricia Oey, “Fund Fee Study,” Morningstar Research Services, April 26, 2018 and Greggory Warren,

“Asset Manager Observer,” Morningstar, March 2019.

21 Jonathan B. Berk and Richard C. Green, “Mutual Fund Flows and Performance in Rational Markets,”

Journal of Political Economy, Vol. 112, No. 6, December 2004, 1269-1295.

22 Tim Koller, Marc Goedhart, and David Wessels, Valuation: Measuring and Managing the Value of

Companies, Sixth Edition (Hoboken, NJ: John Wiley & Sons, 2015), 28-29.

23 Jonathan B. Berk and Jules H. van Binsbergen, “Measuring Skill in the Mutual Fund Industry,” Journal

of Financial Economics, Vol. 118, No. 1, October 2015, 1-20.

24 Jonathan B. Berk, “Five Myths of Active Portfolio Management,” Journal of Portfolio Management,

Spring 2005, 27-31 and Jonathan B. Berk and Jules H. van Binsbergen, “Active Managers are Skilled: On Average, They Add More Than $3 Million Per Year,” Journal of Portfolio Management, Vol. 42, No. 2, Winter 2016, 131-139. Also Jonathan B. Berk and Jules H. van Binsbergen, “Mutual Funds in Equilibrium,” Annual Review of Financial Economics, Vol. 9, No. 1, 2017, 147-167.

25 Jonathan B. Berk and Jules H. van Binsbergen, “How Do Investors Compute the Discount Rate? They

Use the CAPM,” Financial Analysts Journal, Vol. 73, No. 2, Second Quarter 2017, 25-32.

26 Ĺuboš Pástor, Robert F. Stambaugh, Lucian A. Taylor, “Scale and Skill in Active Management,”

Journal of Financial Economics, Vol. 116, No. 1, April 2015, 23-45.

Disclaimers:

This report is provided for informational purposes only and is intended solely for the person to whom it is delivered by BlueMountain Capital Management, LLC (“BlueMountain”). This report is confidential and may not be reproduced in its entirety or in part, or redistributed to any party in any form, without the prior written consent of BlueMountain. This report was prepared in good faith by BlueMountain for your specific use and contains a general market update and information concerning looking for easy games in bonds.

This report does not constitute an offer to sell or the solicitation of an offer to purchase any securities of any funds or accounts managed by BlueMountain (the “Funds”). Any such offer or solicitation may be made only by means of the delivery of a confidential offering memorandum, which will contain material information not included herein and shall supersede, amend and supplement this report in its entirety. Information contained in this report is accurate only as of its date, regardless of the time of delivery or of any investment, and does not purport to be complete, nor does BlueMountain undertake any duty to update the information set forth herein.

This report should not be used as the sole basis for making a decision as to whether or not to invest in the Funds or any other fund or account managed by BlueMountain. In making an investment decision, you must rely on your own examination of the Funds and the terms of the offering. You should not construe the contents of these materials as legal, tax, investment or other advice, or a recommendation to purchase or sell any particular security.

The returns of several market indices are provided in this report as representative of general market conditions and that does not mean that there necessarily will be a correlation between the returns of any of the Funds, on the one hand, and any of these indices, on the other hand.

The information included in this report is based upon information reasonably available to BlueMountain as of the date noted herein. Furthermore, the information included in this report has been obtained from sources that BlueMountain believes to be reliable; however, these sources cannot be guaranteed as to their accuracy or completeness. No representation, warranty or undertaking, express or implied, is given as to the accuracy or completeness of the information contained herein, by BlueMountain, its members, partners or employees, and no liability is accepted by such persons for the accuracy or completeness of any such information.

This report contains certain “forward-looking statements,” which may be identified by the use of such words as “believe,” “expect,” “anticipate,” “should,” “planned,” “estimated,” “potential,” “outlook,” “forecast,” “plan” and other similar terms. Examples of forward-looking statements include, without limitation, estimates with respect to financial condition, results of operations, and success or lack of success of BlueMountain’s investment strategy or the markets generally. All are subject to various factors, including, without limitation, general and local economic conditions, changing levels of competition within certain industries and markets, changes in interest rates, changes in legislation or regulation, and other economic, competitive, governmental, regulatory and technological factors affecting BlueMountain’s operations, each Fund’s operations, and the operations of any portfolio companies of a Fund, any or all of which could cause actual results to differ materially from projected results.