By Charles H. Webb, CFA
In recent years the inability of most active equity managers to outperform their benchmarks has resulted in a significant move toward passive or indexed equity management. This raises the question of whether passive management would be advantageous to fixed income management. For a plethora of reasons, the answer to this question is a resounding, “No.” Unlike equity indexation, true passive fixed income management is impossible. Equities have a manageable finite number of liquid stocks (30, 500, etc.) which can be owned in a portfolio. But in fixed income there are, for example, almost ten thousand issues in the Bloomberg Barclays Aggregate Index, which still represents less than half of the total fixed income universe, excluding about $21 trillion of investible securities. These underrepresented and excluded bonds include many attractive bond structures, such as non-agency commercial and residential mortgage backed securities, CMO obligations and all floating rate issues, which could come in handy in a rising rate environment. The Aggregate holds a record 13% of the index in BBB securities, but in the Bloomberg Barclays U.S. Corporate Index, it is almost 50% BBB rated. Such misrepresentation of the true investible bond universe hinders both diversification and performance, and brings to question why one would want to replicate such an arbitrary measure to begin with.
Within the indices, a substantial percentage of these issues are illiquid because they long ago were put away in portfolios and are rarely, if ever, seen in the open market. Hence, by virtue of arithmetic, almost any holding in a portfolio will be overweight versus the index, sometimes significantly so. Ninety percent of all corporate bonds trade less than five times a year, resulting in potentially gaping bid/ask spreads, a cost that the theoretical indices don’t incur. Securities come and go in the indices with no liquidity problems or transaction costs, which in bonds can result in percentage points lost versus mere basis points for stocks. S&P Dow Jones estimates the average implied transaction costs on bonds are 145 basis points vs 1 or 2 basis points for stocks. Should an issue get downgraded to non-investment grade, it magically disappears from the index at month end (e.g. WorldCom in 2002) while the active manager gets stuck with illiquid trash worth pennies on the dollar, thereby impairing performance. Because indices often add securities when debt is upgraded and prices rise and the inverse case in downgrades, passive Funds/ETFs are forced to buy after prices have gone up and forced to sell after prices have fallen. And large groups of bonds maturing within a year vanish from an index, again with no liquidity constraints or transaction costs, unlike the real world of active management.
Nevertheless, some market participants, mistakenly believing bond indexation is like equity indexation, have looked to bond index funds or ETFs in order to achieve the “market” return in fixed income. The following are some of the reasons why this concept is flawed and unlikely to achieve the desired results for the degree of risk taken. As a proxy for indexed fixed income funds and ETFs, the preeminent bond index fund, The Vanguard Total Bond Market Index Fund will be used. It is the oldest, largest, most popular, and least expensive bond index fund in existence, and its policies and operations are representative of passive bond management.
1) Indexed Bonds Funds overweight companies with the worst financials:
When indexing stocks, portfolio holdings are weighted based upon market cap, where theoretically superior companies are rewarded with higher share prices and thus a higher portfolio weighting. Conversely, bond indexation overweights inferior companies as weightings are based solely upon how much debt companies have issued. Hence, with passive fixed, the greater the debt and the worse the balance sheet, the more overweighted inferior issues are in the portfolio.
2) Indexed Bond Funds and ETFs are not passive indexes:
They are hardly passive as it is impossible to index fixed income when the indices hold thousands of issues. This mandates the use of statistical sampling and the use of derivatives in an effort to replicate an index’s return. However, the degree of sampling representation is startling, with many bond index funds or ETFs holding less than 2% of the issues in the index. Arithmetically, this means something must be overweighted. For instance, Vanguard fund managers nearly always make use of an option in the prospectus called "corporate substitution," for up to 15% of assets, whereby managers selectively overweight types of corporate bonds.
3) Indexed Bond Funds buy outside the index:
As the Vanguard bond index funds prospectus allows, it not only makes extensive use of statistical sampling from within the Aggregate index, but up to 20% of the holdings fall outside the index.
4) Indexed Bond funds employ active management:
It's astonishing how far beyond the indexes prospectuses allow their fund managers managerial latitude. The truth is that “indexed” managers tinker a lot with their portfolios. As the Vanguard bond index fund’s prospectus puts it, "To the extent that the funds invest outside of the index, they may employ active management.”
5) Indexed Fixed Income Funds typically underperform their indices:
According to Morningstar, over the past 10 years ending 6-30-17, the median passive bond manager underperformed their indices net of fees by 17 basis points annualized. Another example: the above Vanguard proxy has underperformed the stated Bloomberg BCAG benchmark over the last 15 years by 16 basis points annualized, and in 11 of those last 15 calendar years. Another prominent bond index fund, The Fidelity Spartan US Bond Index, has underperformed the Bloomberg BCAG index by 37 basis points annualized over the past 10 years. Most surprising is the fact that these two prominent bond index funds, both among the largest, have had an annual difference in performance by up to 162 basis points. How can this occur when both are supposedly “indexed” to the same Aggregate bond index? The answer is that bond indexation only exists in theory and is not applicable in the real world. There is simply no such thing as bond indexation in the sense of equity indexation. Regarding bond ETFs, it has been well documented that the problems of daily compounding (recalculation) of indexed ETFs, especially those using leverage, and in rolling derivative contracts in the futures market, can cause ETF returns to deviate substantially from the underlying index or commodity. (Not to mention liquidity crises such as the August 2015 ETF flash crash). True “buy and hold” fixed indexation (if found) would also prevent the possibility of outperformance by not permitting the managers to exploit market risks and inefficiencies.
6) Indexed Fixed Income Funds and passive typically underperform active bond managers:
According to Morningstar, over the past 10 years ending 3-31-17 the median active bond manager outperformed the median passive manager by 49 basis points annualized net of fees. In 2017, 61% of active managers outperformed passive peers while also outperforming on both 3- and 5-year periods.
7) Indexed Bond Funds fees are not significantly less expensive:
Only the expense ratio for the Vanguard Fund above, long regarded as the cheapest fund, at 6 basis points, is significantly less than active separate account fixed management. The above referenced Fidelity US Bond Index Fund is up to 22 basis points, which is competitive with active management. Bid-ask spreads in fixed income ETFs are not insignificant, as they widen whenever bond volatility rises. Long after the most recent financial crisis the spread on many popular fixed indexed ETFs still showed gaps in bid-ask prices of more than $8 according to NYSE Arca. Nearly 200 ETFs have had bid-ask spreads of over 50 basis points. Add to this the effect of higher trading costs in illiquid securities and it results in the average ETF “charging” 56 basis points, up from 40 basis points in the 2000’s.
8) Indexed Bond Funds managers can't customize client preferences, or allow for the elective taking of gains and losses:
Passive management does not screen social investment criteria such as specific “sin bonds.” Many feel that bond index fund managers are ill-equipped to manage credit quality, that is, to select the highest quality bonds or anticipate which are likely to be upgraded. Also, active managers can potentially better control call risk and are probably better positioned to invest only in bonds that cannot be called, thereby avoiding having to reinvest assets at lower interest rates when higher-yielding bonds are redeemed. Bond Index Funds cannot take selective gains or losses on holdings for reporting or tax purposes or for specialized accounting needs such as maximizing return on book, projecting or managing income, managing profits and losses, etc., for insurance or similar companies. Moreover, indexation eliminates bonds being pledged, dedicated, immunized or loaned.
9) Indexed fixed income management’s extensive use of derivatives may violate client policy / guidelines:
Passive managers routinely use up to 20% in derivative securities to manage cash flow or to take advantage of arbitrage opportunities, such as when T-bond futures prices are significantly lower than the cash index.
10) Fixed indexation, mutual funds and ETF’s have a “perpetual duration” and don’t ever mature:
e.g.: An individual 10-year issue assures 100 cents on the dollar paid upon maturity. Because 10 yr funds / ETFs never mature and keep a near-constant duration there is increased interest rate risk and no assurance of return of your entire principal, and you could easily lose money if interest rates rise over the next 10 years. And unlike real bonds, coupon payments from bonds in a ETF get converted into periodic dividend payments that aren’t always predictable.
This is not to say that there isn’t a time and place for fixed income indexation, primarily for smaller entities or individual investors who lack the critical mass to get liquidity or broad diversification. Attempting to actively manage and diversify a small fixed income portfolio within the inefficient and non-transparent corporate and mortgage backed bond markets without sacrificing a large portion of one’s capital to retail bond markups and sales fees, is very difficult, even for experienced pros. But for larger institutions, active separate account management is preferable, given the surreptitious inherent bets and tracking error that fixed income indexation entails.