Bond Portfolio Management Strategies

Write about each of theses points:
Bond Portfolio Performance
Bond Portfolio Style
Bond Portfolio Strategies:
1. Passive Management Strategies
a. Buy and hold
b. Indexing
2. Active Management Strategies
a. Interest rate anticipation
b. Valuation analysis
c. Credit analysis +(Edward formulas: z and zeta)
d. Yield spread analysis
CHAPTER 19
Bond Portfolio Management Strategies
After you read this chapter, you should be able to answer the following questions:
• What are the five major classes of bond portfolio management strategies?
• How is the investment style box defined for fixed-income portfolios?
• What are the two main types of passive bond portfolio management strategies?
• What are the main active bond portfolio management strategies?
• How do active bond portfolio strategies differ from one another in terms of scope, scalability, and risk-adjusted  return potential?
• What is meant by core-plus bond portfolio management?
• What are the primary “plus” strategies in a core-plus approach to management?
• How does a matched-funding approach to bond portfolio management differ from an active or passive approach?
• How does bond immunization work and how does that strategy differ from a cash-matching approach to  managing a bond portfolio?
• What other dedicated management strategies are available to bond managers?
• What is meant by a contingent immunization approach to bond portfolio management?  In this chapter, we shift attention from bond valuation and

analysis to an examination of the most widely used bond portfolio management strategies. After a brief discussion of how bonds have performed as an

asset class in recent years and how fixed-income investment styles are typically classified, we will see that these strategies can be classi- fied into

one of five broad approaches: passive management, active management, core-plus management, matched-funding management, and contingent and structured

active management. In the following sections, we describe these approaches in more detail and give examples of how each is used in practice.  19.1 BOND

PORTFOLIO PERFORMANCE, STYLE, AND STRATEGY  The volatile pattern of interest rates prevailing during recent decades has provided increasingly at-

tractive returns to bond investors of all types. Active bond portfolio managers have found the fre- quent opportunities to realize capital gains that

resulted from those rate shifts to be especially attractive. However, despite the favorable economic climate that has prevailed for most of the last

quarter century, it remains the case that fixed-income portfolios generally produce both less return and less volatility than other asset classes (e.g.,

domestic equity, foreign equity, real estate). Exhibit 19.1 summarizes the average annual returns and standard deviations for several performance in-

dexes over the 20-year period ending in 2010, a time horizon that saw two major downturns in eq- uity markets. Bond portfolios—as represented by the

Citigroup U.S. Government Bond, World Government Bond, and U.S. Corporate Bond indexes—fall at the lower end of the risk-return spec- trum measured by

the capital market line, making them a conservative choice within an investor’s

691

692 Part 5: Analysis and Management of Bonds
Exhibit 19.1 Risk-Return Comparison between Bond Portfolios and Other Asset Classes

16.0 14.0 12.0 10.0
8.0 6.0 4.0
2.0 0.0
MSCI Emerging Mkt Equity
GS Commodity
25.0 30.0 35.0 40.0

FTSE NAREIT
EAFE

Citi US Hi Yld Bond Russell 1000 MSCI World Equity
Russell 2000 S&P 500

Citi World Govt Bond
US T-bills Inflation
Citi US Corp Bond Citi US Govt Bond
10.0 15.0

0.0
5.0
20.0
Volatility (%)

Source: Author calculations.
overall asset allocation strategy. On the other hand, the U.S. High Yield Bond Index exhibited risk and return dynamics that made it more comparable to

many of the equity indexes. Finally, notice that the relatively low historical correlation between fixed-income and equity securities—only 0.08 over

this time period—have made bond portfolios an excellent tool for diversifying risk as well.
In Chapter 16, we saw that it was useful to classify the investment style of equity portfolios along two dimensions: market capitalization and relative

valuation (i.e., value vs. growth). Simi- larly, the investment style of a bond portfolio can be summarized by its two most important char- acteristics:

credit quality and interest rate sensitivity. Exhibit 19.2 shows how the 3 × 3 style grid can be adjusted to accommodate these dimensions. The average

credit quality of the portfolio can be classified as high grade (e.g., government, agency, AAA-rated or AA-rated corporate bonds), me- dium grade (e.g.,

A-rated or BBB-rated), or low grade (e.g., below BBB-rated), based on the profile of the composite holding. We also established in the last chapter that

average duration is an effec- tive way to measure the portfolio’s price sensitivity to interest rate changes. The second dimension of the bond

portfolio’s investment style can be separated as short term (e.g., duration less than 3.0 years), intermediate term (e.g., duration between 3.0 and 6.5

years), or long term (e.g., duration more than 6.5 years). For example, the Barclays Capital U.S. Aggregate Bond Index, one of the most widely used

benchmarks, is purposely constructed to mimic the profile of the investment grade fixed-income security market in the United States, which typically

consists of 70–80 percent government, agency, or AAA-rated bonds. The BCA Index is structured to maintain an average duration of between 4.0 and 5.0

years. Thus, it would plot in the middle cell of the top row of the style grid in Exhibit 19.2 and would be classified as a high-grade/intermediate-term

portfolio.
Just as the inherent investment style of one bond portfolio compared to another can vary widely, so too can the underlying strategic approach adopted by

the managers who formed those portfolios. The nature of the investor’s problem usually dictates the way in which the manager will think about designing

the bond portfolio that solves that problem. Consequently,
Return (%)
an investor who desires a specific amount of cash to fund a financial obligation in the near

Chapter 19: Bond Portfolio Management Strategies 693
Exhibit 19.2 Fixed-Income Investment Style Grid

High-Grade
Credit Quality:
Low-Grade
High-Grade/ Short-Term
Medium-Grade/ Short-Term
Low-Grade/ Short-Term
BCA Index
High-Grade/ Intermediate-Term

Medium-Grade/ Intermediate-Term
Low-Grade/ Intermediate-Term
High-Grade/ Long-Term
Medium-Grade/ Long-Term
Low-Grade/ Long-Term

Short-Term
Average Duration:
Long-Term

Exhibit 19.3 Bond Portfolio Investment Strategies

1. Passive Management Strategies a. Buy and hold  b. Indexing
2. Active Management Strategies
a. Interest rate anticipation
b. Valuation analysis
c. Credit analysis
d. Yield spread analysis
e. Sector/country analysis
f. Prepayment/option analysis
g. Other (e.g., liquidity, currency, anomaly capture)
3. Core-Plus Management Strategies a. Enhanced indexing b. Active/passive “plus” sectors
4. Matched-Funding Strategies a. Dedicated: exact cash match b. Dedicated: optimal cash match c. Classical immunization d. Horizon matching
5. Contingent & Structured Strategies a. Contingent immunization b. Structured management

future will likely have a very different strategy for assembling a fixed-income portfolio than someone whose goal is to maximize capital gains resulting

from an anticipated shift in interest rates. Exhibit 19.3, which is based in part on the development in Leibowitz (1986a), indicates
that bond portfolio strategies can be divided into the five broad groups mentioned above.

694 Part 5: Analysis and Management of Bonds
Prior to the 1960s, only the first two strategic approaches—passive and active—were widely available, and most bond portfolios were managed on a buy-

and-hold basis with the intention of producing a steady stream of cash flow for the investor. The early 1970s saw a growing level of curiosity with

alternative active bond portfolio management approaches, while the late 1970s and early 1980s were characterized by record-breaking inflation and

interest rates as well as extremely volatile yields across all spectrums of the bond market. This led to the intro- duction of many new financial

instruments in response to the increase in rate volatility (e.g., adjustable-rate bonds and mortgages). Since the mid-1980s, matched-funding techniques,

core-plus strategies, and contingent bond management approaches have been developed to meet the increased needs of institutional investors, such as

pension funds and insurance com- panies. Finally, beginning in the mid-1990s, it has become increasingly common to see bonds combined with positions in

derivative securities in the management of sophisticated fixed- income portfolios; this topic will be explored in later chapters.
19.2 PASSIVE MANAGEMENT STRATEGIES
Two specific passive portfolio management strategies exist. First, is a buy-and-hold strategy in which a manager selects a portfolio of bonds based on

the objectives and constraints of the client with the intent of holding these bonds to maturity. In the second passive strategy— indexing—the objective

is to construct a portfolio of bonds that will be matched as closely as possible to the performance of a specified bond index, such as the Barclays

Capital U.S. Aggre- gate Bond Index described earlier.
19.2.1 Buy-and-Hold Strategy
The simplest fixed-income portfolio management strategy is to buy and hold. This approach involves finding securities with the desired levels of credit

quality, coupon rate, term to matu- rity or duration, and other important indenture provisions, such as call and sinking fund fea- tures. Buy-and-hold

investors do not consider active trading as a viable alternative to achieve abnormal returns but look for bond issue whose maturity/duration

characteristics approximate their investment horizon in order to reduce price and reinvestment risk. Many successful bond investors and institutional

portfolio managers follow a modified buy-and-hold strategy wherein they invest in an issue with the intention of holding it to maturity, but still look

for opportunities to trade into a more desirable position should the occasion arise. Of course, if the buy-and-hold approach is modified too much, it

becomes an active strategy.
Whether the manager follows a strict or modified buy-and-hold approach, the critical concept is finding investment vehicles that possess the appropriate

maturity, yield, and credit quality at- tributes. The strategy does not restrict the investor to accept whatever the market has to offer, nor does it

imply that selectivity is unimportant. Attractive high-yielding issues with desirable features and quality standards are actively sought. For example,

these investors recognize that agency issues or asset-backed securities generally provide attractive incremental returns relative to Treasuries with

little sacrifice in quality, or that various call and put features can materially impact the risk and realized yield of an issue. Thus, successful buy-

and-hold investors use their knowledge of market and security characteristics to seek out attractive realized yields.
Finally, recognize that there is an important fundamental difference between managing a bond portfolio and a stock portfolio on a buy-and-hold basis.

Since bonds eventually ma- ture with the passing of time whereas stock shares do not, the bond manager is faced with the need to periodically reinvest

the funds from a matured issue. However, the stock man- ager can employ a “pure” buy-and-hold strategy in which he never adjusts the portfolio’s

composition once it is formed. Fixed-income portfolio managers often address this concern by forming a bond ladder, in which they divide their

investment funds evenly across the

Chapter 19: Bond Portfolio Management Strategies 695
portfolio into instruments that mature at regular intervals. For instance, a manager with an in- termediate-term investment focus, instead of investing

all of her funds in a five-year zero- coupon security—which would become a four-year security after one year had passed—could follow a laddered approach

and buy equal amounts of bonds maturing in annual intervals be- tween one and nine years. The idea would then be to hold each bond to maturity, but to

rein- vest the proceeds from a maturing bond into a new instrument with a maturity at the far end of the ladder (that is, to reinvest a maturing bond in

a brand-new nine-year issue). In this way, the desired maturity/duration target for the portfolio can be maintained over time without having to

continually adjust the investment weights for the remaining positions.
19.2.2 Indexing Strategy
In the discussion of efficient capital markets earlier in the text, numerous empirical studies were cited that have demonstrated that the majority of

money managers have not been able to match the risk-adjusted return performance of common stock and bond indexes. As a result, many investors have

chosen to invest at least some of the funds dedicated to these asset classes on a passive basis. Rather than forming their own buy-and-hold portfolios,

many inves- tors prefer to hold a bond portfolio designed to mimic a selected fixed-income index. In such a case, the bond index manager is judged not

on the basis of his ability to produce abnormal returns, but by how closely his portfolio produces returns that match those of the index. When

describing similar concepts for stock index managers in Chapter 16, we saw that tracking error was a useful tool to judge how closely the returns of a

managed portfolio match (i.e., “track”) those of the targeted index. Recall that tracking error is measured as the standard deviation of the difference

in returns produced by the managed portfolio and the index over time. An annualized tracking error statistic of 1 percent or less usually indicates that

an index fund manager is doing a good job matching the performance of the index.
As with stock index funds, when designing a bond portfolio to mimic a hypothetical index, managers can follow two different paths: full replication or

stratified sampling. While it is quite common when constructing stock index funds to fully replicate the underlying index, the bond index fund manager

often follows a sampling approach, wherein a smaller number of instru- ments are held in the actual portfolio than appear in the index. One reason for

this is that bond indexes often contain several thousand specific issues and are adjusted frequently, mak- ing them both impractical and expensive to

replicate precisely in practice. The goal of the stratified sampling approach is to create a bond portfolio that matches the important charac- teristics

of the underlying index—such as credit quality, industry composition, maturity/dura- tion, or coupon rate—while maintaining a portfolio that is more

cost effective to manage. To the extent that the manager is not able to match these characteristics over time, the tracking error of the indexed

portfolio will typically increase.
When initiating an indexing strategy, the selection of an appropriate market index is clearly a very important decision, chiefly because it directly

determines the client’s risk-return results. Consequently, it is important for investors to be acquainted with the main characteristics (e.g.,

maturity/duration, credit quality) of their selected index. Reilly and Wright (1994, 1997, 2005) have examined many aspects of the major bond indexes,

such as their risk-return characteris- tics and the correlation between them over time. Also, Dialynas and Murata (2006) and Vol- pert (2001) discuss

how the characteristics of indexes affect their performance in different interest rate environments. Finally, recognize that the aggregate bond market

and the indexes change over time. Reilly, Kao, and Wright (1992) demonstrated that the market experienced significant shifts in composition, maturity,

and duration since 1975, which can significantly impact the tracking error performance of an indexed portfolio.1
1For further discussion on how bond market dynamics change over time, see Van Horne (2001). For more on how these changes impact the management of an

indexed portfolio, see Mossavar-Rahmani (1991) and Fabozzi (2007, Chapter 23).

696 Part 5: Analysis and Management of Bonds
19.2.3 Bond Indexing in Practice: An Example
To see how two actual managers have responded to the challenge of forming a bond portfolio designed to track one of the leading indexes, we consider the

several aspects of the Vanguard Total Bond Index Fund (ticker: VBMFX) and the iShares Barclays U.S. Aggregate (ticker: AGG) exchange-traded fund over a

recent investment period. Both of these portfolios were created to mimic the performance of the Barclays Capital U.S. Aggregate Index—whose formal

ticker symbol is LBUSTRUU—and represent the two methods widely used in practice to create indexed portfolios for retail investors (i.e., index mutual

funds and ETFs).
Exhibit 19.4 summarizes many of the most important structural characteristics for these funds as well as for the underlying index. It is interesting to

contrast the approaches these managers have adopted to replicate the index, which contains over 8,000 separate bond issues and would be difficult to

recreate exactly. The manager of VBMFX actually holds more posi- tions than the index (i.e., almost 14,000 names), whereas AGG’s manager follows a

stratified sampling method and attempts to mimic the index by holding only 710 distinct security posi- tions. Not surprisingly, AGG also has a

substantially higher portfolio turnover statistic than VBMFX—488 percent versus 80 percent—which is undoubtedly a by-product of trying to keep the ETF’s

portfolio composition aligned with that of the much larger index.
This difference in index replication approaches also leads to slight differences in the relevant investment characteristics of the portfolios. Generally

speaking, the index mutual fund maintains a bond portfolio that is closer to the index in terms of average duration (LBUSTRUU: 4.9, VBMFX: 4.8, AGG:

4.6), but both of the managed funds deviate from the index in terms of credit quality, as measured by the percentage of the portfolio carrying a rating

of AAA or higher (LBUSTRUU: 77.6, VBMFX: 75.9, AGG: 76.1). These discrepancies lead, in turn, to a sizeable difference in the tracking error statistic

produced by each manager, with VBMFX and AGG hav- ing annualized values of 0.46 and 1.80 percent, respectively. Thus, the exchange-traded fund’s

tracking error was roughly four times greater than that of index mutual fund and only the latter falls within the range of what is considered normal for

a passive approach to investing. Finally, the expense ratios for both indexed vehicles are quite similar (i.e., 22–24 basis points) and much smaller

than what would be typical for an actively managed bond portfolio.
Exhibit 19.4 Indexed Bond Investing: Index Fund vs. ETF, January 2011

Barclays U.S. Aggregate Index (LBUSTRUU)
Vanguard Total Bond Index Fund (VBMFX)
iShares Barclays U.S. Aggregate ETF (AGG)
Style Classification High Grade/ High Grade/ High Grade/ (Credit Grade/Duration) Intermediate-Term Intermediate-Term Intermediate-Term
# of Holdings Annual Turnover (%) Annual Yield (%) Avg. Duration (yrs.) Avg. Maturity (yrs.) Credit Quality (% of Port.):
8,216 13,812 710 — 80 488 3.0 3.4 3.5 4.9 4.8 4.6 7.1 6.6 6.4
Govt/Agency/AAA 77.6 AA 4.5 A 9.8 BBB 8.1 Other/Not Rated 0.0
Tracking Error (%/yr.): (1/07–12/10) — Expense Ratio (%) —
75.9 76.1 4.5 2.8 10.5 10.9 9.1 8.3 0.0 1.9
0.46 1.80 0.22 0.24
Source: Prepared by the authors using data from Morningstar, Inc., and Fidelity Investments.

Chapter 19: Bond Portfolio Management Strategies 697

19.3 ACTIVE MANAGEMENT STRATEGIES
As we have seen with active equity portfolio management, the active fixed-income manager attempts to form a portfolio of securities that will outperform

her designated benchmark over time. That is, she will attempt to hold a collection of bonds that produce superior risk-adjusted returns (i.e., alpha)

compared to the index against which her investment performance is mea- sured. Of course, to beat a benchmark, the active manager must form a portfolio

that differs from the holdings comprising the index in a meaningful way. Thus, active bond management strategies are closely tied to the manager’s view

of what factors or market conditions will be the source of the incremental alpha returns she seeks.
Layard-Liesching (2001) analyzed the investment attributes of several potential sources of alpha for the active bond portfolio manager, all of which

depend on some structural barrier that prevents the bond market from being fully efficient. These characteristics are summarized in Exhibit 19.5. He

compares each active strategy on four dimensions: (1) scalability (i.e., how large a position can be taken); (2) sustainability (i.e., how far into the

future the strategy can be successfully employed); (3) risk-adjusted performance; and (4) extreme values (i.e., how exposed the strategy is to the

chance of a large loss). For instance, in the interest rate anticipation category, he argues that dura- tion-based active bets—in which the manager

increases or decreases the average duration level of the active portfolio on the belief that the yield curve will either shift down or up, respectively

—are highly scalable since they can be implemented with virtually any securities available in the market. However, they also offer the lowest chance of

sustainable performance as well as the worst risk- adjusted returns. By contrast, credit risk bets—where the manager takes a position in a bond that she

thinks has a substantially different default potential than has been priced in by the market—are a much more sustainable and reliable source of

potential alpha. Finally, notice that while valuation analysis offers the active manager reasonable alpha potential, it is a more limited strategy from

a scalability standpoint since it relies on identifying pricing errors in specific bond issues.2
Exhibit 19.5 Characteristics of Active Bond Portfolio Strategies

Risk-Adjusted Source Scalability Sustainability Performance* Extreme Values
Interest Rate Anticipation: Duration High Yield Curve Shape Low
Valuation Analysis: Security Selection Low Anomaly Capture Low
Credit Risk High Yield Spread Analysis:
Optionality Medium Prepayment Medium Liquidity Low
Global & Tactical: Sector Allocation High Country Allocation High Currency High
Very Weak Very Weak
Medium Weak Strong
Medium Medium Strong
Strong Strong Medium
1 Yes 3 No
5 No 7 Yes 8 Yes
7 Yes 6 Yes 3 Yes
6 No 5 No 2 Yes

*1 = Low, 10 = High (Note: This list is subjective; investors should make their own assessment of these criteria.) Source: Adapted from Ronald Layard-

Liesching, “Exploiting Opportunities in Global Bond Markets,” in Core-Plus Bond Man- agement (Charlottesville, VA: AIMR), 2001.

2Additional discussions of active bond portfolio strategies can be found in Boyd and Mercer (2010), Malvey (2005),

698 Part 5: Analysis and Management of Bonds
In the remainder of this section, we will explore four categories of active bond strategies— interest rate anticipation, valuation analysis, credit

risk, and yield spread analysis—in more de- tail, as well as describe bond swaps as a means to implement a specific active strategic view.
19.3.1 Interest Rate Anticipation
Interest rate anticipation is perhaps the riskiest active management strategy because it in- volves relying on uncertain forecasts of future interest

rates. The idea is to preserve capital when an increase in interest rates is anticipated and achieve attractive capital gains when inter- est rates are

expected to decline. Such objectives usually are attained by altering the duration structure of the portfolio (i.e., reducing portfolio duration when

interest rates are expected to increase and increasing the portfolio duration when a decline in yields is anticipated). Thus, the risk in such portfolio

restructuring is largely a function of these duration alterations. When durations are shortened, substantial income could be sacrificed and the

opportunity for capital gains could be lost if interest rates decline rather than rise. Similarly, portfolio shifts prompted by anticipated rate

declines are also risky. Assuming that we are at a peak in interest rates, it is likely that the yield curve is downward sloping, which means that bond

coupons will decline with maturity. Therefore, the investor is sacrificing current income by shifting from high-coupon short bonds to longer-duration

bonds. At the same time, the portfolio is purposely exposed to greater price volatility that could work against the portfolio if an unex- pected

increase in yields occurs. Note that the portfolio adjustments prompted by anticipation of an increase in rates involve less risk of an absolute capital

loss. When you reduce the matu- rity, the worst that can happen is that interest income is reduced and/or capital gains are forgone (opportunity cost).
Once future (expected) interest rates have been determined, the procedure relies largely on technical matters. Assume that you expect an increase in

interest rates and want to preserve your capital by reducing the duration of your portfolio. A popular choice would be high- yielding, short-term

obligations, such as Treasury bills. Although your primary concern is to preserve capital, you would nevertheless look for the best return possible

given the maturity constraint. Liquidity also is important because, after interest rates increase, yields may experi- ence a period of stability before

they decline, and you would want to shift positions quickly to benefit from the higher income and/or capital gains.
To illustrate this process, suppose that the yield curve for U.S. Treasury bonds is currently flat across all maturities at 4.75 percent. You have

observed the following “paired” transaction by an active bond portfolio manager:
Bond Transaction
1 Buy 2 Sell
Type
U.S. Govt. U.S. Govt.
Maturity Coupon (yrs.) Rate (%)
7 8 13 0
Modified Duration
5.438 12.698

What does this trade suggest about the manager’s view as to how the yield curve is likely to change in the future? First, by switching out of a long-

maturity, zero-coupon bond into an in- termediate-maturity, high-coupon bond, the manager has significantly shortened the modified duration of the

position and, presumably, of the entire portfolio. Thus, this trade is consistent with a view that the yields will rise in the future (i.e., the yield

curve will shift up). Exhibit 19.6 shows two potential situations for how this might happen. In Scenario #1, the manager fore- casts that all rates will

shift up by 50 basis points, keeping the yield curve flat. In Scenario #2, all future yields increase but rates on longer-term securities increase by

more, meaning that the shape of the curve moves from flat to upward sloping. In either case, the manager will benefit from replacing a bond with a

single cash flow paid out more than a decade in the future to one with a much shorter maturity that also makes payments every six months. Of course, the

Chapter 19: Bond Portfolio Management Strategies 699
Exhibit 19.6 Anticipated Yield Curve Shifts Consistent with a Duration-Reducing Active Bond Trade
Yield to Maturity
6.00% 5.60%
5.25% 4.75%
+35 bp +50 bp
7 yrs
+75 bp +50 bp
13 yrs
Scenario #2
Scenario #1
Original Curve
Maturity

trade will be the most profitable (in present value terms) under Scenario #2, since the 7-year bond will be subjected to a smaller yield increase than

the 13-year security. Finally, notice that by using one Treasury security to replace another, the manager has not introduced any credit risk into the

portfolio that might conflict with his interest rate anticipation view.
An alternative way to shorten maturities is to use a cushion bond—a high-yielding, long- term obligation that carries a coupon substantially above the

current market rate and that, due to its current call feature and call price, has a market price lower than what it should be given current market

yields. As a result, its yield is higher than normal. An example would be a 10-year bond with a 12 percent coupon, currently callable at 110. If current

market rates are 8 percent, this bond (if it were noncallable) would have a price of about 127; because of its call price, however, it will stay close

to 110, and its yield will be about 10 percent rather than 8 percent. Bond portfolio managers look for cushion bonds when they expect a modest increase

in rates because such issues provide attractive current income and protection against capital loss. Because these bonds are trading at an abnormally

high yield, market rates would have to rise to that abnormal level before their price would react, as described by Homer and Leibowitz (2004, Chapter

5).
A totally different posture is assumed by investors who anticipate a decline in interest rates. The significant risk involved in restructuring a

portfolio to take advantage of a decline in in- terest rates is balanced by the potential for substantial capital gains and holding period returns. When

you expect lower interest rates, you should increase the duration of the portfolio because the longer the duration, the greater the positive price

volatility. Also, liquidity is important because you want to be able to close out the position quickly when the drop in rates has been completed.
Because interest rate sensitivity is critical, it is important to recall that the higher the quality of an obligation, the more sensitive it is to

interest rate changes. Therefore, high-grade securi- ties should be used, such as Treasuries, agencies, or corporates rated AAA through BBB. Finally,

you want to concentrate on noncallable issues or those with strong call protection be- cause of the substantial call risk discussed in Chapter 18 in

connection with the analysis of duration and convexity.

700 Part 5: Analysis and Management of Bonds
19.3.2 Valuation Analysis
With valuation analysis, the portfolio manager attempts to select bonds based on their intrinsic value, which is determined based on their

characteristics and the average value of these characteris- tics in the marketplace. As an example, a bond’s rating will dictate a certain spread

relative to com- parable Treasury bonds: Long maturity might be worth an added 60 basis points relative to short maturity (i.e., the maturity or term

spread); a given deferred call feature might require a higher or lower yield; a specified sinking fund would likewise mean higher or lower required

yields. Given all the characteristics of the bond and the normal cost of the characteristics in terms of yield, you would determine the bond’s required

yield and, therefore, its implied intrinsic value. Doing this for a number of bonds, you would compare these derived bond values to the prevailing

market prices to determine which bonds are undervalued or overvalued. Based on your confidence in the characteristic costs, you would buy the

undervalued issues and ignore or sell the overvalued issues.
Success in valuation analysis is based on understanding the characteristics that are impor- tant in valuation and being able to accurately estimate the

yield cost of these characteristics with the understanding that these yield costs change over time.
19.3.3 Credit Analysis
A credit analysis strategy involves detailed analysis of the bond issuer to determine expected changes in its default risk. This involves attempting to

project changes in the credit ratings as- signed to bonds by the various rating agencies discussed in Chapter 17. These rating changes are affected by

internal changes in the entity (e.g., changes in important financial ratios) and by changes in the external environment (i.e., changes in the firm’s

industry and the economy). During periods of strong economic expansion, even financially weak firms may survive and prosper. In contrast, during severe

economic contractions, normally strong firms may find it very difficult to meet financial obligations. Therefore, historically there has been a strong

cycli- cal pattern to rating changes: typically, downgrades increase during economic contractions and decline during economic expansions.
To use credit analysis as a bond management strategy, it is necessary to project rating changes prior to the announcement by the rating agencies. This

can be quite challenging be- cause the market adjusts rather quickly to bond rating changes, especially to downgrades. Therefore, you want to acquire

bond issues expected to experience an upgrade and sell or avoid those bond issues expected to be downgraded.
Credit Analysis of High-Yield (Junk) Bonds One of the most obvious opportunities for credit analysis is the analysis of high-yield (junk) bonds. As

demonstrated by several studies, the yield differential between junk bonds that are rated below BBB and Treasury securities ranges from about 250 basis

points to over 1,500 basis points. Notably, these yield differentials vary substantially over time, as shown by a time-series plot in Exhibit 19.7.

Specifically, the average yield spread ranged from a low of less than 300 basis points in 1985 and 1997 to a high of almost 1,550 basis points during

late 2008.
Although the spreads have changed, a study by Mody and Taylor (2003) indicated that the average credit quality of high-yield bonds also changed over

time as indicated by interest cov- erage changes over the business cycle. Also, the credit quality of bonds within rating categories changed over the

business cycle as demonstrated by Reilly, Wright, and Gentry (2009).
These changes in credit quality make credit analysis of high-yield bonds more important, but also more difficult. This means that bond analysts and

portfolio managers need to engage in detailed credit analysis to select bonds that will survive. Given the spread in promised yields, if a portfolio

manager can—through rigorous credit analysis—avoid bonds with a high probability of default or downgrade, high-yield bonds will provide substantial

rates of return for the investor; see Vine (2001) and Fabozzi (2005a).

Chapter 19: Bond Portfolio Management Strategies 701
Exhibit 19.7 Monthly Yield Spread History, Citigroup High-Yield Market Index versus Constant-Maturity 10-Year Treasury
1600
1400
1200
1000
800
600
400
200
0
Year

Source: Author calculations from Federal Reserve Board and index data.
In summary, substantial opportunity for generating high risk-adjusted returns can be derived by investing in high-yield bonds if you do the credit

analysis required to avoid defaults, which occur with these bonds at substantially higher rates than the overall bond market, as shown by Asquith,

Mullins, and Wolff (1989) and Altman (1992).
Exhibit 19.8 lists the cumulative average default rates for bonds with different ratings and for various time periods after issue. Over 10 years—the

holding period that is widely used in prac- tice for comparative purposes—the default rate for BBB investment-grade bonds is only 4.76 percent, but the

default rate increases to over 16 percent for BB-rated bonds, to over 32 percent for B-rated bonds, and to over 53 percent for CCC-rated bonds. These

default rates do not mean that investors should avoid high-yield bonds, but they do indicate that extensive credit analysis is a critical component for

success within this risky sector of the fixed-income market.
Investing in Defaulted Debt Beyond high-yield bonds that have high credit risk and high default rates, a new set of investment opportunities has

evolved—investing in defaulted debt. While this sector requires an understanding of legal procedures surrounding bankruptcy as well as economic

analysis, the returns have generally been consistent with the risk, as demon- strated by Altman (1993), Altman and Simon (2001), Ward and Griepentrog

(1993), and Reilly, Wright, and Altman (1998).
Credit Analysis Models The credit analysis of high-yield bonds can use a statistical model or basic fundamental analysis that recognizes some of the

unique characteristics of these bonds. Altman-Nammacher (1987) suggest that a modified Z-score model used to predict the probability of bankruptcy

within two years can also be used to predict default for these high- yield bonds or as a gauge of changes in credit quality. The Z-score model combines

traditional financial measures with a multivariate technique known as multiple discriminant analysis to
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Basis Points

Exhibit 19.8 Cumulative Average Default Rates for Corporate Bonds: 1981–2009 (%) —Time horizon (years)—
Rating
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
AAA AA+ AA
0.00 0.03 0.14 0.26 0.39 0.51 0.58 0.68 0.74 0.82 0.86 0.90 0.94 1.04 1.14 (0.00) (0.20) (0.35) (0.44) (0.58) (0.69) (0.75) (0.83) (0.84) (0.85) (0.86)

(0.00) (0.87) (0.94) (1.02) 0.00 0.06 0.06 0.13 0.19 0.26 0.33 0.33 0.33 0.33 0.33 0.33 0.33 0.33 0.33 (0.00) (0.32) (0.33) (0.67) (0.98) (1.37) (1.82)

(1.85) (1.89) (1.93) (1.96) (2.00) (2.05) (2.09) (2.14) 0.02 0.04 0.06 0.16 0.25 0.31 0.41 0.50 0.58 0.66 0.71 0.74 0.84 0.87 0.91
AA− A+ A
(0.08) (0.13) (0.16) (0.23) (0.25) (0.30) (0.38) (0.42) (0.42) (0.56) (0.59) (0.58) (0.52) (0.52) (0.52) 0.04 0.12 0.24 0.35 0.46 0.61 0.71 0.79 0.00

0.98 1.08 1.20 1.24 1.34 1.39 (0.10) (0.16) (0.35) (0.73) (0.86) (1.08) (1.33) (1.55) (1.62) (1.60) (1.56) (1.83) (1.83) (1.92) (2.06) 0.07 0.13 0.29

0.48 0.64 0.78 0.96 1.13 1.33 1.55 1.76 1.97 2.23 2.56 2.84 (0.15) (0.27) (0.35) (0.46) (0.55) (0.56) (0.53) (0.48) (0.51) (0.62) (0.82) (0.00) (1.00)

(1.29) (1.35)
A− BBB+ BBB BBB− BB+ BB BB− B+
(0.14) (0.26) (0.34) (0.40) (0.46) (0.51) (0.62) (0.74) (0.85) (0.95) (1.05) (1.11) (1.05) (0.97) (0.82) 0.09 0.25 0.41 0.60 0.87 1.17 1.59 1.90 2.15

2.37 2.55 2.75 2.94 3.07 3.16 (0.20) (0.34) (0.48) (0.65) (0.90) (1.10) (1.52) (1.71) (2.03) (2.30) (2.29) (2.30) (2.28) (2.29) (2.34) 0.17 0.48 0.84

1.20 1.63 2.13 2.50 2.89 3.35 3.75 4.13 4.37 4.72 5.28 5.92 (0.32) (0.65) (0.96) (1.04) (1.30) (1.57) (1.90) (1.91) (1.75) (1.78) (1.50) (1.45) (1.56)

(1.97) (2.21) 0.24 0.59 0.91 1.42 1.98 2.52 3.04 3.58 4.18 4.76 5.41 5.98 6.51 6.70 7.06
B
(0.84) (1.77) (2.89) (3.59) (4.19) (4.35) (4.06] (3.84) (3.38) (3.20) (2.78) (3.13) (3.21) (3.43) (3.70) 1.34 4.12 7.02 9.76 12.14 14.51 16.60 18.72

20.55 22.03 23.19 24.07 25.11 26.12 27.05 (1.81) (3.65) (5.26) (5.98) (6.15) (6.32) (6.59) (6.66] (6.72) (6.81) (7.18) (7.37) (7.74) (7.95) (8.05) 2.70

7.22 11.54 15.35 18.29 20.55 22.66 24.53 26.22 27.93 29.36 30.50 31.62 32.63 33.59 (2.10) (4.60) (6.20) (7.46) (8.03) (8.05) (7.78) (7.79) (8.02) (5.97)

(5.79) (5.11) (4.84) (5.06) (5.29) 6.26 13.32 18.75 22.51 25.09 27.61 29.12 30.32 31.26 32.26 33.26 34.12 34.98 35.77 36.64
B−
(4.54) (7.39) (8.35) (9.01) (9.46) (8.84) (8.28) (7.74) (7.56) (7.21) (6.62) (6.09) (5.21) (4.88) (4.97) 9.86 17.94 23.95 28.04 31.05 32.96 34.84 35.93

36.83 37.45 38.15 38.78 39.12 39.49 40.09 (7.92) (12.34) (14.08) (14.63) (14.61) (14.96) (14.30) (13.86) (14.18) (14.26) (14.20) (14.19) (13.74) (13.88)

(14.14)
CCC/C
27.98 36.95 42.40 45.57 48.05 49.19 50.26 51.09 52.44 53.41 54.32 55.33 56.38 (12.90) (13.28) (13.62) (14.37) (14.30) (12.73) (12.33) (12.38) (11.74)

(10.47) (10.85) (11.73) (11.61)
57.28
57.28
Investment grade Speculative grade All rated
0.13 0.35 0.60 0.91 1.24 1.58 1.90 2.20 2.50 (0.12) (0.28) (0.40) (0.51) (0.61) (0.66) (0.70) (0.73) (0.80) 4.44 8.68 12.42 15.46 17.90 19.96 21.72

23.25 24.67 (2.82) (4.58) (5.70) (6.29) (6.45) (5.99) (5.29) (4.75) (4.42) 1.63 3.23 4.67 5.89 6.90 7.79 8.55 9.23 9.00 (1.07) (1.83) (2.40) (2.77)

(2.93) (2.85) (2.67) (2.55) (2.48)
2.00 (0.84)
3.08 (0.86)
3.31 (0.85)
3.55 (0.76)
(9.61) 3.78 (0.70)
(9.75) 4.04 (0.67)
0.09 0.22 0.36 0.51 0.68 0.91 1.14 1.38 1.65 1.98 2.25 2.42 2.56 2.66 2.95
(0.35) (0.54) (0.73) (0.93) (1.09) (1.21) (1.41) (1.69) (1.76) (1.85) (1.79) (1.85) (1.00) (1.80) (1.46) 0.41 1.21 2.14 3.26 4.38 5.43 6.38 7.33 8.10

8.96 9.79 10.54 11.25 12.39 13.18 (0.47) (1.19) (1.77) (2.21) (2.75) (3.49) (3.74) (3.54) (3.58) (3.22) (3.23) (3.31) (3.41) (3.24) (3.50) 0.53 1.49

2.81 4.21 5.51 6.88 8.09 8.86 9.97 11.10 11.90 12.74 13.43 13.91 14.79 (0.97) (1.98) (3.12) (4.06) (4.84) (5.12) (5.64) (6.05) (6.57) (6.96) (7.12)

(6.97) (6.84) (6.94) (7.03) 0.82 2.55 4.91 7.09 9.22 11.11 12.71 14.15 15.40 16.43 17.48 18.44 19.00 19.34 19.73
Source: Adapted from 2009 Annual Global Corporate Default Study and Rating Transitions (New York: Standard & Poor’s, March 17, 2010): p. 60. Reprinted

with permis- sion; numbers in parentheses are standard deviations.
25.96 (4.07)
27.08 (3.81)
28.02 (3.55)
28.91 (3.47)
29.68 (3.62)
30.45 (3.65)
10.45 (2.30)
10.97 (2.06)
11.40 (1.77)
11.82 (1.74)
12.20 (1.96)
12.60 (2.14)

Z = the overall credit score
Chapter 19: Bond Portfolio Management Strategies 703
derive a set of weights for the specified variables. The result is an overall credit score (Z) for each firm. The model is of the form:
Z=a0 +a1X1 +a2X2 +a3X3 + +anXn
where: X1:::Xn =theexplanatoryvariablesðratiosandmarketmeasuresÞ
a0 :::an =the weightings or coefficients The final model used in this analysis included the following seven financial measures:
X1 = profitability: earnings before interest and taxes ðEBITÞ=total assets ðTAÞ X2 = stability of profitability measure: the standard error of estimate

of EBIT=TA
ðnormalized for 10 yearsÞ X3 = debt service capabilities ðinterest coverageÞ: EBIT=interest charges X4 = cumulative profitability: retained

earnings=total assets X5 = liquidity: current assets=current liabilities X6 = capitalization levels: market value of equity=total capital ðfive-year

averageÞ X7 = size: total tangible assets ðnormalizedÞ
As an example of this process, Exhibit 19.9 illustrates the Z-score calculations for two compa- nies whose bonds were rated below investment grade as of

January 2011: Ford Motors (Ba2 rat- ing), a multinational automobile manufacturer, and Kronos Worldwide (B2), a specialty chemical producer. Z-scores

typically range from −5.0 to +20.0, with higher scores (i.e., above +3.0) indicating that bankruptcy over the next two years is unlikely, while lower

scores (i.e., be- low +1.8) suggest an increased potential for business failure. In this case, Kronos’s Z-score shows that it is in stronger financial

condition than Ford (i.e., +2.25 vs. +0.75), despite having the lower bond credit rating. It is important to note, however, that these scores are best

interpreted as they change over time, rather than as a single observation. In that respect, the charts at the bottom of each panel of Exhibit 19.9 show

that both companies’ Z-scores have improved substantially since the stock market crisis in late 2008.3 Active bond managers following a credit analysis

strategy might use this as a tool to help predict rating upgrades and downgrades before they occur. Jons- son and Fridson (1996) discuss predicting

default rates for high-yield bonds in more detail.
In contrast to using a model that provides a composite credit score, many analysts simply adapt their basic corporate bond analysis techniques to the

unique needs of high-yield bonds, which have characteristics of common stock as shown by Reilly and Wright (1994, 2001). Fabozzi (2005, Chapter 32)

claims that the analysis of high-yield bonds is the same as with any bond except that the following areas of analysis should be expanded:
1. What is the firm’s competitive position in terms of cost and pricing? This can be critical to a small firm.
2. What is the firm’s cash flow relative to cash requirements for interest, research, growth, and periods of economic decline? Also, what is the

firm’s borrowing capacity that can serve as a safety net and provide flexibility?
3. What is the liquidity value of the firm’s assets? Are these assets available for liquidation (are there any claims against them)? In many cases,

asset sales are a critical part of the strategy for a leveraged buyout.
4. How good is the total management team? Is the management team committed to and capable of operating in the high-risk environment of this firm?
5. What is the firm’s financial leverage on an absolute basis and on a market-adjusted basis (using market value of equity and debt)?

3Exhibit 19.9 also reports a double prime Z-score for each company, which is just a modification of the Z-score that
uses fewer of the financial measures in the discriminant analysis procedure described earlier.

704 Part 5: Analysis and Management of Bonds
Exhibit 19.9 Altman’s Z-Score Analysis

A. Ford Motor
<HELP> for explanation.
Ford Motor Co – F US Financial Data Input Reference Data Tangible Assets Working Capital Retained Earnings
Earnings Before Int & Taxes Market Value of Equity Total Liablities Sales to Tangible Assets Total Shareholders’ Equity Financial Health Assessment and

Outlook Altman’s Z-score
EquityAZS Altman’s Z-Score Model
9/2010 User Input 180207.00 180207.00 –18558.00 –18558.00 –19998.00 –19998.00
8437.00 8437.00 42644.16 42644.16 182070.00 182070.00

0.73 –1740.00
0.73 –1740.00
0.75 –0.73
2010

Altman’s Double Prime Z-score Z-score History 0.80 0.70
0.60 0.50 0.40 0.30 0.20
0.10 Altman’s Z-score 0.7484
0.75 –0.73
2008 2009
–0.50 –1.00 –1.50 –2.00

0.00 Altman’s Double Prime Z-score –0.7327 –0.10

2004
Australia 61 2 9777 8600 Japan 81 3 3201 8900
2005
Brazil 5511 3048 4500 Singapore 65 6212 1000
B. Kronos Worldwide
<HELP> for explanation.
Kronos Worldwide Inc – KRO US Financial Data Input Reference Data Tangible Assets
Working Capital Retained Earnings
Earnings Before Int & Taxes Market Value of Equity Total Liablities Sales to Tangible Assets Total Shareholders’ Equity Financial Health Assessment and

Outlook Altman’s Z-score
9/2010 1329.20 363.90 –655.20
123.90 1951.27 921.90 1.04 407.30
2.25 1.28
2008
2006 Europe 44 20 7330 7500
2007 U.S. 1 212 318 2000
Hong Kong 852 2977 6000 SN 335716 H263-281-3 04-Jan-2011 14:41:03
EquityAZS Altman’s Z-Score Model
User Input
1329.20 363.90 –655.20
123.90 1951.27 921.90 1.04 407.30
2.25 1.28
Altman’s Z-score
Altman’s Double Prime Z-score
2010
Germany 49 69 9204 1210 Copyright 2011 Bloomberg Finance L.P.

Altman’s Double Prime Z-score Z-score History
2.00 1.50 1.00 0.50
2.2529 1.50 1.2793
1.00 0.50 0.00 –0.50

Australia 61 2 9777 8600 Japan 81 3 3201 8900
Germany 49 69 9204 1210 SN 335716 H263-281-3 04-Jan-2011 14:44:16
2005
2006
2007
2009
Brazil 5511 3048 4500 Singapore 65 6212 1000 U.S. 1 212 318 2000 Copyright 2011 Bloomberg Finance L.P.
Europe 44 20 7330 7500
Hong Kong 852 2977 6000

Chapter 19: Bond Portfolio Management Strategies 705
In addition to the potentially higher financial risks, Sondhi (1995) and Squires (1998b) point out several factors that can also impact business risk.

An increase in business risk may exist if the firm sells off some operations that have favorable risk characteristics with the remaining operations,

such as a division that has low correlation of earnings with other units of the firm. Further, a change in management operating philosophy could have a

negative impact on operating earnings. Asset divestiture plans often are a major element of a leveraged buyout because they provide necessary capital

that is used to reduce the substantial debt taken on as part of the acquisition. Therefore, it is important to examine the liquidity of the assets,

their estimated selling values, and the timing of these programs. If the divestiture program is successful wherein the prices received are above normal

expectations and the assets are sold ahead of schedule, this can be grounds for upgrading the debt. Finally, it is necessary to constantly monitor the

firm’s refinancing flexibility. Specifically, what refinancing will be necessary, what does the schedule look like, and will the capital suppliers be

receptive to the refinancing?
19.3.4 Yield Spread Analysis
As discussed in Chapter 18, spread analysis assumes normal relationships exist between the yields for bonds in alternative sectors (e.g., the spread

between high-grade versus low-grade industrial or between industrial versus utility bonds). When an abnormal relationship occurs, a bond manager could

execute various sector swaps. The crucial factor is developing the back- ground to know the normal yield relationship and to evaluate the liquidity

necessary to buy or sell the required issues quickly enough to take advantage of the temporary yield abnormality.
Dialynas and Edington (1992) consider several specific factors that affect the aggregate spread. The generally accepted explanation of changes in the

yield spread is that it is related to the economic environment. The spread widens during periods of economic uncertainty and recession because investors

require larger risk premiums (i.e., larger spreads). In contrast, the spread will decline during periods of economic confidence and expansion. The

authors contend that a more encompassing factor is the impact of interest rate (yield) volatility, which will affect the spread via three effects: (1)

yield volatility and the behavior of embedded op- tions, (2) yield volatility and transactional liquidity, and (3) the effect of yield volatility on the

business cycle.
Recall that the value of callable bonds is equal to the value of a noncallable bond minus the value of the call option. Therefore, if the value of the

option goes up, the value of the callable bond will decline and its yield will increase. When yield volatility increases, the value of the call option

increases, which causes a decline in the price of the callable bond and a rise in the bond’s yield and its yield spread relative to Treasury bonds.

Similarly, an increase in yield volatility will raise the uncertainty facing bond dealers and cause them to increase their bid- ask spreads that reflect

the transactional liquidity for these bonds. This liquidity will have a bigger effect on nongovernment bonds, so their yield spread relative to Treasury

bonds will increase. Finally, interest rate volatility causes uncertainty for business executives and consu- mers regarding their cost of funds. This

typically will precede an economic decline that will, in turn, lead to an increase in the yield spread.
It is possible to have a change in yield spread for reasons other than economic uncertainty. If there is a period of greater yield volatility that is

not a period of economic uncertainty, the yield spread will increase due to the embedded option effect and the transactional liquidity ef- fect. This

analysis implies that when examining yield spreads, you should pay particular atten- tion to interest rate (yield) volatility.
You can use the book as source chapter 19. here is the link, I couldn’t upload the file:
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&cad=rja&uact=8&ved=0ahUKEwiH05S2yaHMAhVKQyYKHSfIA9UQFggrMAI&url=https%3A%2F

%2Fdocs.google.com%2Ffile%2Fd%2F0B8pig2KdTaOBd3BTYUtkSXR4SW8%2Fedit&usg=AFQjCNFk93UUyhdqBPlsqlMSR6fl3hg3vA&sig2=l_YBeMWulhmverkpyksENw