19 June - FI Active Mgt - Credit Strategies (skippe

FIXED-INCOME ACTIVE MANAGEMENT: CREDIT STRATEGIES

by Campe Goodman, CFA and Oleg Melentyev, CFA.

Campe Goodman, CFA, is at Wellington Management (USA). Oleg Melentyev, CFA, is at Bank of America Merrill Lynch (USA).

LEARNING OUTCOMES

The candidate should be able to:

• describe risk considerations for spread-based fixed-income portfolios

• discuss the advantages and disadvantages of credit spread measures for spread-based fixed-income portfolios, and explain why option-adjusted spread is considered the most appropriate measure

• discuss bottom-up approaches to credit strategies

• discuss top-down approaches to credit strategies

• discuss liquidity risk in credit markets and how liquidity risk can be managed in a credit portfolio

• describe how to assess and manage tail risk in credit portfolios

• discuss the use of credit default swap strategies in active fixed-income portfolio management

• discuss various portfolio positioning strategies that managers can use to implement a specific credit spread view

• discuss considerations in constructing and managing portfolios across international credit markets

• describe the use of structured financial instruments as an alternative to corporate bonds in credit portfolios

• describe key inputs, outputs, and considerations in using analytical tools to manage fixed-income portfolios

INTRODUCTION

Most fixed-income instruments trade at a nominal yield to maturity (YTM) that lies above that for an equivalent government or benchmark bond of similar maturity. This yield spread or difference compensates investors for credit risk, or the risk that they may not receive interest and principal cash flows as expected, whether this arises from a financially distressed corporate borrower, a sovereign issuer unable (or unwilling) to meet scheduled payments or a deterioration in credit quality in an underlying pool of assets of a structured instrument such as an asset-backed security. A portion of the yield spread reflects the bid–offer cost of buying or selling a particular bond versus a government security, a liquidity premium that varies based on market conditions. Active managers of spread-based fixed-income portfolios take positions in credit and other risk factors that vary from those of an index to generate excess return versus passive index replication. Financial analysts who build on their foundational knowledge by mastering these more advanced fixed-income concepts and tools will broaden their career opportunities in the investment industry.

We begin by reviewing expected fixed-income portfolio return components with a particular focus on credit spreads. These spreads are not directly observable but rather derived from market information. Similar to benchmark yield curves, credit-spread curves are often defined by spread level and slope, and usually grouped by credit rating to gauge relative risk as well as to anticipate and act on expected changes in these relationships over the business cycle. We outline credit spread measures for fixed- and floating-rate bonds and quantify the effect of spread changes on portfolio value. Building blocks for active credit management beyond individual bonds include exchange-traded funds (ETFs), structured financial instruments, and derivative products such as credit default swaps (CDS). These tools are used to describe bottom-up and top-down active credit management approaches as well as how managers position spread-based fixed-income portfolios to capitalize on a market view.

KEY CREDIT AND SPREAD CONCEPTS FOR ACTIVE MANAGEMENT

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563573/key-credit-and-spread-concepts-for-active-management-1

CREDIT STRATEGIES

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563583/credit-strategies-1

Learning Outcomes

• discuss bottom-up approaches to credit strategies

• discuss top-down approaches to credit strategies

Bottom-Up Credit Strategies

As active fixed-income managers consider the selection process for spread-based bond portfolio investments, they must assess different ways in which to maximize excess spread across the fixed-income issuer types, industries, and instruments within their prescribed investment mandate. A fundamental choice these investors face is whether to engage in an individual security selection process or bottom-up approach; a macro- or market-based, top-down approach in pursuing this objective; or a combination of both.

Fundamental credit analysis covered earlier in the curriculum considers the basis on which a specific issuer can satisfy its interest and principal payments through bond maturity. Analysts often assess unsecured corporate bonds using factors such as profitability and leverage to identify the sources and variability of cash flows available to an issuer to service debt. These measures are usually chosen and compared relative to an industry and/or the jurisdiction in which the issuer operates. In the case of a sovereign borrower, the relevant metric is the economic activity within a government’s jurisdiction and the government’s ability and willingness to levy taxes and generate sufficient revenue to meet its obligations. Alternatively, for a special purpose entity issuer with bonds backed by mortgage-based or other securitized cash flows, a credit measure of both the residential borrowers and underlying collateral value as well as internal credit enhancements are among the primary factors considered in the assessment.

While individual bonds across all these issuer types are usually rated by at least two of the major credit rating agencies, active managers typically conduct their own credit assessment of individual borrowers rather than relying on ratings, which are frequently used to define a mandate (e.g., investment grade versus high yield), categorize, or benchmark investments of similar credit quality.

Defining the Credit Universe

A bottom-up approach typically begins with a manager defining the universe of eligible bonds within a mandate and then grouping the universe into categories that allow consistent relative value analysis across comparable borrowers. For example, a corporate bond portfolio manager is likely to divide eligible bonds into industry sectors, such as media and telecommunications and industrials, as well as into subsectors and/or firms located in different jurisdictions. Media and telecommunications subsectors include firms in the cable and satellite industries, internet media, and telecommunications carriers. Within each sector or subsector based on either industry classification methodologies or a customized approach, she can use relative value analysis to determine the bonds that are attractively valued.

EXAMPLE 13

Dividing the Credit Universe

1. An investor is conducting a relative value analysis on global bond issuers in the health care sector. He is trying to decide whether the global health care sector is a sufficiently narrow sector for his analysis. Through his research, he has determined the following:

   • Biotech and pharmaceutical companies are active globally across Europe, Asia, and the Americas.

   • Health care facilities are typically local in nature and tend to sell into only one of these three regions.

   • Medical equipment and devices is a more cyclical business, and many of these firms are part of multi-industry companies in which health care accounts for a smaller fraction of overall company sales.

   Describe considerations that the investor can use in determining how to best divide the health care sector into comparable companies.

Bottom-Up Credit Analysis

Once the credit universe has been divided into sectors and prospective bonds identified, the investor evaluates each issuer’s implied credit risk comparing company-specific financial information to spread-related compensation for assuming default, credit migration, and liquidity risks for comparative purposes.

Beyond the prospects within a company’s industry, its competitive position within that industry, and operating history, financial ratios are a valuable tool to compare creditworthiness across firms. Earlier lessons stressed the value of key ratios, including profitability and cash flow, leverage, and debt coverage, which are summarized in Exhibit 16.

Exhibit 16:

Key Financial Ratios for Bottom-Up Credit Analysis

Ratio	Description	Advantages	Disadvantages
EBITDA/ Total Assets	Profitability Cash flow as a percentage of assets	Combines operating income with non-cash expense	Ignores capital expenditures and working capital changes
Debt/ Capital	Leverage Fraction of company’s capital financed with debt	Direct measure of relative reliance on debt financing	More relevant for investment-grade than high-yield issuers
EBITDA/ Interest Expense	Coverage Cash flow available to service debt	Measures relative issuer ability to meet debt payments	Volatile measure for firms with high cash flow variability

While offering a relatively consistent basis for comparison across firms and over time, reliance on financial ratios based on publicly available accounting data alone is of limited value because of comparability issues across firms and industries as well as the historical nature of financial statements. Alternative measures combine several relevant financial ratios with market-based measures to establish a forward-looking approach to creditworthiness.

A previous lesson established that statistical credit analysis models to measure individual issuer creditworthiness can be categorized as either reduced form credit models or structural credit models. Reduced form models solve for default intensity, or the POD over a specific time period, using observable company-specific variables such as financial ratios and recovery assumptions as well as macroeconomic variables, including economic growth and market volatility measures. Structural credit models use market-based variables to estimate the market value of an issuer’s assets and the volatility of asset value. The likelihood of default is defined as the probability of the asset value falling below that of liabilities.

An early example of the reduced form approach is the Z-score established by Altman (1968), which combined liquidity (working capital/total assets), profitability (retained earnings/total assets), asset efficiency (EBIT/ total assets), market versus book value of equity, and asset turnover (sales/total assets) factors weighted by coefficients to form a composite score. Each composite, or Z-score, was used to classify manufacturing firms into those expected to remain solvent and those anticipated to go bankrupt. Similar to credit scoring models, this multiple discriminant analysis reduces the dimensionality of the input variables to a single cutoff Z-score that represents the default threshold, as shown in the following example.

EXAMPLE 14

Z-Score Comparison of Two Firms

A United Kingdom–based financial analyst considers a Z-score model in evaluating two publicly traded non-manufacturing companies as follows:

Z-Score Model = 1.2 × A + 1.4 × B + 3.3 × C + 0.6 × D + 0.999 × E,

where

A is Working Capital/Total Assets

B is Retained Earnings/Total Assets

C is EBIT/Total Assets

D is Market Value of Equity/Total Liabilities

E is Sales/Total Assets

Firms with a Z-score greater than 3.0 are considered financially sound, those scoring between 3.0 and 1.8 are at greater risk of financial distress, and those with a Z-score below 1.8 are likely to face insolvency.

1. Calculate the Z-score for Firm 1 and Firm 2. Which has a higher likelihood of financial distress based on this measure?

   Financial Data (GBP thousands)/Firm	Firm 1	Firm 2
   Total Sales	23,110	15,270
   EBIT	6,910	2,350
   Current Assets	7,560	4,990
   Total Assets	36,360	23,998
   Current Liabilities	5,400	3,564
   Total Liabilities	9,970	10,050
   Retained Earnings	20,890	13,787
   Market Value of Equity	29,000	18,270

   Solution to 1:

   First, calculate the respective ratios for both firms as follows, noting that working capital is equal to current assets minus current liabilities:

   Z-Score Factors	Firm 1	Firm 2
   Working Capital/Assets	0.059	0.059
   Retained Earnings/Assets	0.575	0.575
   EBIT/Total Assets	0.190	0.098
   Market Value of Equity/Total Liabilities	2.909	1.818
   Sales/Total Assets	0.636	0.636

   Solving for the respective Z-scores, we find that Firm 1 has a Z-score of 3.883, while Firm 2 has a Z-score of 2.925. Firm 2 therefore has a greater likelihood of financial distress.

2. Evaluate the most likely reasons for the difference in creditworthiness between the two firms based on the Z-score model factors.

   Solution to 2:

   Comparing the respective Z-score ratios of Firm 1 and Firm 2, we find that Firm 2 has a far lower asset efficiency (EBIT/Total Assets of 9.8% versus 19% for Firm 1) and a lower relative equity market value (Market Value of Equity/Total Liabilities of 1.818 versus 2.909 for Firm 1) than Firm 1, while all other ratios are comparable.

Structural credit models used in practice include Moody’s Analytics Expected Default Frequency (EDF) and Bloomberg’s Default Risk (DRSK) models, both of which provide daily POD estimates for a broad range of issuers over a selected period. The EDF model estimates a forward-looking POD defined as the point at which the market value of assets falls below a firm’s obligations. The model uses asset volatility to determine the likelihood of reaching the default point and is calibrated for different industries, regions, and observed credit market dynamics.

Bloomberg’s DRSK model estimate for AbbVie Inc., as shown in Exhibit 17, includes a market-based asset value measure derived from equity market capitalization and equity volatility as well as a default threshold measured using the book value of liabilities. These and other DRSK model inputs in the left column of the screen can be defined by users and compared within and across industry sectors. In addition to the one-year POD estimate of 0.0413%, DRSK calculates a “model” CDS spread (upper left corner) which can be compared to the actual market CDS spread.

Exhibit 17:

Bloomberg DRSK Model Estimate for AbbVie Inc.

Source: Bloomberg

Both the EDF and DRSK approaches are sometimes referred to as “distance to default” models because a probability distribution is used to determine how far an issuer’s current market value of assets is from the default threshold for a given period.

EXAMPLE 15

“Distance to Default” Models

1. An active manager is weighing an investment in the bonds of two issuers in the same industry with identical PODs using a structural credit model. Which of the following changes to the model inputs for one of the issuers would lead the analyst to expect an increase in the POD for that issuer?

   1. An increase in the issuer’s coverage ratio

   2. An increase in the volatility of the issuer’s stock price

   3. A decrease in the issuer’s leverage ratio

   Solution:

   The correct answer is B. Higher equity volatility increases the likelihood that the market value of the issuer’s assets will fall below the default threshold. A higher coverage ratio in A implies higher cash flow as a percentage of assets, increasing the issuer’s ability to service its debt obligations. The decrease in the issuer’s leverage ratio in C represents a decline in the amount of debt versus equity, reducing the issuer’s likelihood of financial distress.

Bottom-Up Relative Value Analysis

Given two issuers with similar credit risk, the investor will typically choose bonds of the issuer with the higher yield spread, given the greater potential for excess returns. For issuers with different credit-related risk, the investor must decide whether the additional spread is sufficient compensation for the incremental exposure. The excess expected return calculation in Equation 10 captures the relationship between yield spreads and the components of credit risk, as seen in the following example.

EXAMPLE 16

Comparing Investments Using Expected Excess Return

A portfolio manager considers two industrial bonds for a one-year investment:

Issuer	Rating	EffSpreadDur	YTM	Z-Spread
A Rated Industrial	A2	5.0	4.0%	100 bps
B Rated Industrial	B2	7.0	6.5%	350 bps

The manager observes a historical annual default probability of 0.27% for A2 rated issuers and 3.19% for B2 rated issuers and assumes a 40% recovery rate for both bonds.

1. Compute the estimated excess return for each bond assuming no change in spreads, and interpret whether the B rated bond spread provides sufficient compensation for the incremental risk.

   Solution to 1:

   As per Equation 10,

   E [ExcessSpread] ≈ Spread0 −(EffSpreadDur × ΔSpread) − (POD × LGD).

   A rated expected excess return is 0.84% = 1% − (5 × 0) − (0.27% × 60%). B rated expected excess return is 1.59% = 3.5% − (7 × 0) − (3.19% × 60%). The B rated bond appears to provide sufficient compensation for the added risk.

2. Which bond is more attractive if spreads are expected to widen by 10%?

   Solution to 2:

   Recalculate Equation 10 with ΔSpread of 10 bps for the A rated bond and 35 bps for the B rated bond.A rated excess return is 0.34% = 1% − (5 × 0.1%) − (0.27% × 60%).B rated excess return is −0.86% = 3.5% − (7 × 0.35%) − (3.19% × 60%).The A rated bond is more attractive under this scenario.

In practice, bonds from different issuers usually also have various maturity, embedded call or put provisions, liquidity, and other characteristics, so these additional features should be taken into account during the security selection process. For example, structural differences such as callability or priority within the capital structure must be factored in because they affect valuation. Also, bonds recently issued in larger tranches by frequent issuers will tend to have narrower bid–offer spreads and greater daily transaction volume, allowing investors to buy or sell the bond at a lower cost. This feature is likely to be of greater importance to investors who expect short-term spread narrowing and/or have a relatively short investment time horizon. Note that relative liquidity tends to decline over time, particularly if the same issuer returns to the bond market and offers a price concession for new debt. If, on the other hand, an investor has a longer investment horizon with the flexibility to hold a bond to maturity, he might be able to increase excess return via a greater liquidity premium. Finally, other factors driving potential yield spread differences to be considered include split ratings or negative ratings outlooks, potential merger and acquisition activity, and other positive or negative company events not adequately reflected in the analysis.

When deciding among frequent issuers with several bond issues outstanding, investors might consider using credit spread curves for these issuers across maturities to gauge relative value.

EXAMPLE 17

Using Spread Curves in Relative Value Analysis

A United States–based issuer has the following option-free bonds outstanding:

Outstanding Debt	Term	Coupon	Price	YTM
2-year issue	2	4.25%	106.7	0.864%
5-year issue	5	3.25%	106	1.984%
15-year issue	15	2.75%	91	3.528%

Current on-the-run US Treasury YTMs are as follows:

Tenor	Coupon	Price
2y	0.250%	100
5y	0.875%	100
10y	2.000%	100
20y	2.250%	100

An investor conxsiders the purchase of a new 10-year issue from the company and expects the new bond to include a 10 bp new issue premium. What is the fair value spread for the new issue based on outstanding debt?

1. First, solve for the credit spreads for outstanding bonds as the difference in the YTM from an actual or interpolated government bond:5-year spread: 110.9 bps (= 1.984% − 0.875%)15-year spread: Solve for 10- and 20-year bond interpolation weights.10-year weight: w10 = 0.50% (= (20 − 10)/(15 − 10))20-year weight: w20 = 0.50% (= (1 − w10))15-year interpolated bond: 2.125% = (2.00% × 0.5) + (2.25% × 0.5)15-year spread: 140.3 bps (= 3.528% − 2.125%)

2. Derive the implied 10-year new issue spread by interpolating the 5- and 15-year credit spreads using the same interpolation weights as for Treasuries and adding the 10 bp new issue premium.10-year spread: 135.6 bps = 0.1% + (1.109% × 0.5 + 1.403% × 0.5)

Many issuers have several bond issues, each of which typically has a different maturity and duration. To reflect the various maturities, a spread curve can be developed for each issuer and can be useful in conducting relative value analysis. A spread curve is the fitted curve of credit spreads for similar bonds of an issuer plotted against the maturity of those bonds.

Exhibit 18 plots the Z-spread versus maturities for select outstanding bonds of two A2/A+ rated health care companies, Eli Lilly (LLY) and Bristol-Myers Squibb (BMS), which have similar probabilities of default.

Exhibit 18:

Spread Curves for Eli Lilly and Bristol-Myers Squibb

Source: Bloomberg

These spread curves are closely aligned except in roughly five-year and nearly 30-year maturities, where the BMS spreads are approximately 10 bps wider than those of LLY. If the bonds have similar features and liquidity, then a manager might conclude that the market perceives BMS credit risk to be slightly higher than that of LLY. However, if the manager believes that BMS is the stronger credit, several actions are possible depending on portfolio objectives and constraints. For example, if the investment mandate is to outperform a benchmark using long-only positions, the manager might overweight BMS bonds and underweight LLY bonds relative to the benchmark. If the objective is to generate positive absolute returns, underweighting or avoiding LLY bonds is less appropriate because such actions are meaningful only in the context of a benchmark. If permitted, the manager could also consider a long–short CDS strategy outlined later.

Once a manager has identified specific issuers and bond maturities to actively over- or underweight versus a benchmark, the next important step is to quantify and track these active investments in the context of the primary indexing risk factors identified in an earlier lesson in the active portfolio construction process. For example, if an investor chooses to overweight specific health care industry issuers versus the respective sector and spread duration contributions of the benchmark index, the difference in portfolio weights between the active and index positions establishes a basis upon which excess return can be measured going forward.

Top-Down Credit Strategies

A top-down approach to credit strategy focuses on a broader set of factors affecting the bond universe in contrast to the more detailed and issuer-specific bottom-up approach. Macro factors critical to credit investors include economic growth, real rates and inflation, changes in expected market volatility and risk appetite, recent credit spread changes, industry trends, geopolitical risk, and currency movements. Assessment of these factors guides investors in selecting credit market sectors with attractive relative value characteristics, with an increased bond allocation to more attractive sectors and an underweight (or possibly short bond positions in) less favorable sectors. Top-down investors frequently use broader sector distinctions than under a bottom-up approach. For example, a top-down investor expecting credit spreads to narrow might favor the relative value opportunity of high-yield bonds over investment-grade bonds.

GDP growth is critical to the credit cycle, as seen in Exhibit 19, which shows global speculative-grade default rates versus the real GDP growth rate among G7 countries from 1962 to 2019. Sharp declines in GDP growth are often associated with rising default rates.

Exhibit 19:

Global Speculative-Grade Default Rate and Real GDP Growth Rate for G7 countries, 1962–2019

Sources: Moody’s Investors Service, OECD (IHS Markit)

A portfolio manager or analyst might decide to factor this relationship into the investment decision-making process; for example, an above-consensus real GDP growth forecast might lead to an increased high-yield allocation if future defaults are expected to remain below market expectations.

Assessing Credit Quality in a Top-Down Approach

Active top-down and bottom-up credit managers frequently use public ratings to categorize and rank the credit quality of bonds within a portfolio. As investors compare investments across credit ratings, the fact that default risk rises more rapidly as ratings decline is important to consider. The use of weighted factors, such as those established by Moody’s based on the likelihood of credit loss over a specific period versus ordinal factors across the credit spectrum, enables managers to capture this effect more accurately, as demonstrated in Exhibit 20.

Exhibit 20:

Weighted Versus Ordinal Credit Rating Categories

Moody’s	S&P	Fitch	Ordinal	Weighted
Aaa	AAA	AAA	1	1
Aa1	AA+	AA+	2	10
Aa2	AA	AA	3	20
Aa3	AA-	AA-	4	40
A1	A+	A+	5	70
A2	A	A	6	120
A3	A-	A-	7	180
Baa1	BBB+	BBB+	8	260
Baa2	BBB	BBB	9	360
Baa3	BBB-	BBB-	10	610
Ba1	BB+	BB+	11	940
Ba2	BB	BB	12	1,350
Ba3	BB-	BB-	13	1,766
B1	B+	B+	14	2,220
B2	B	B	15	2,720
B3	B-	B-	16	3,490
Caa1	CCC+	CCC+	17	4,770
Caa2	CCC	CCC	18	6,500
Caa3	CCC-	CCC-	19	10,000
Ca	CC	CC	20

Source: Moody’s Investors Service

The impact of weighted ratings is best demonstrated using an example. For instance, assume a manager is assessing credit quality for a portfolio in which half of the bonds are rated A1/A+ and the other half are rated Ba3/BB-. Using an ordinal scale, the average portfolio credit quality score is 9 (= 50% × 5 + 50% × 13), which corresponds to an average rating of Baa2/BBB in Exhibit 20. However, using the weighted scale at the far right, the portfolio’s average credit quality score is 918 (= 50% × 70 + 50% × 1,766), or closer to Ba1/BB+, two levels (notches) below the average rating derived using an ordinal scale.

Earlier readings underscored the risks of relying on public credit ratings, in particular that ratings tend to lag the market’s pricing of credit risk critical to an active investor. In addition, one should note that S&P’s and Moody’s ratings capture different types of risks, with S&P ratings focused on the POD, while Moody’s focuses on expected losses, which could influence historical comparisons. The credit rating time horizon is also critical because ratings agencies issue both short-term and long-term ratings for specific issuers, which might warrant additional attention. For these reasons, active managers often prefer to use credit spread measures such as OAS to measure average portfolio credit quality. To calculate a portfolio’s average OAS, each bond’s individual OAS is weighted by its market value. A manager might also group bonds by OAS categories, which are sometimes mapped to public ratings for comparative purposes.

The use of spread-based rather than rating-based measures also facilitates the measurement of changes in portfolio value due to spread changes. As shown earlier, Equation 5 provides a framework to quantify portfolio value changes due to yield spread movements:

%∆PVSpread ≈ −(EffSpreadDur × ΔSpread) + (½ × EffSpreadCon × (ΔSpread)2)

Smaller yield spread changes are often estimated using the first term in Equation 5. This analytical duration approach provides a reasonable approximation of the price–yield spread relationship for investment-grade bonds with low credit spreads. However, for bonds with greater default risk further down the credit spectrum, changes to both the EffSpreadDur and the ΔSpread terms might be required to accurately reflect empirical observations of how credit risk changes affect overall portfolio value.

In isolating portfolio value changes due to yield spread changes using EffSpreadDur, Equation 5 implicitly assumes that government bond YTMs and credit spreads are uncorrelated, independent variables. However, empirical duration estimates using statistical models often diverge from analytical duration calculations over time and in different interest rate environments. For instance, under a “flight to quality” scenario, the macroeconomic factors driving government bond YTMs lower will cause high-yield bond credit spreads to rise as the result of an expectation of a greater likelihood and higher severity of financial distress, as shown in Exhibit 21 during the COVID-19 pandemic in early 2020.

Exhibit 21:

US Treasury Yields versus US Corporate BB Spreads, 2020

Source: Bloomberg

As for ΔSpread, recall the empirical observation that bonds trading at wider spreads usually experience larger spread changes, which are proportional to the DTS measure in Equation 8.

These greater changes in bond spread have an impact similar to that of the weighted Moody’s credit rating categories in Exhibit 18.

EXAMPLE 18

Top-Down Excess Returns

1. An investor has formed expectations across four bond rating categories and intends to overweight the category with the highest expected excess return over the next 12 months. Evaluate which rating group is the most attractive based on the information in the following table and assuming no change in spread duration:

   Rating Category	Current OAS	Expected ∆OAS	Expected Loss (POD × LGD)	EffSpreadDur
   A	1.05%	−0.25%	0.06%	5.5
   Baa	1.35%	−0.35%	0.30%	6.0
   Ba	2.45%	−0.50%	0.60%	4.5
   B	3.50%	−0.75%	3.00%	4.0

   Solution:

   The following table summarizes expected excess returns E [ExcessSpread] ≈ Spread0 − (EffSpreadDur × ΔSpread) − (POD × LGD) for each of the four rating categories. For example, expected excess return for rating category A is 2.37% (=1.05% − (5.5 × −0.25%) − 0.06%).

   Rating Category	Current OAS	Expected ∆OAS	Expected Loss (POD × LGD)	EffSpreadDur	E(Excess Return)
   A	1.05%	−0.25%	0.06%	5.5	2.37%
   Baa	1.35%	−0.35%	0.30%	6.0	3.15%
   Ba	2.45%	−0.50%	0.60%	4.5	4.10%
   B	3.50%	−0.75%	3.00%	4.0	3.50%

Given that the Ba category has the highest expected excess return, it is the most attractive rating category to overweight in the portfolio.

Sector Allocation in a Top-Down Approach

Industry sector allocations (or weightings) are an important part of a top-down approach to credit strategy. To determine which sector(s) to over- or underweight, an active portfolio manager usually begins with an interest rate and overall market view established using macroeconomic variables introduced earlier. This view is a key step in determining whether specific sectors of the economy are likely to over- or underperform over the manager’s investment time horizon.

Quantitative methods such as regression analysis are often used in making industry allocation decisions. For example, the average spread of bonds within an individual industry sector and rating category might be compared with the average spread of the bonds with the same rating but excluding the chosen industry sector. Alternatively, a portfolio manager might also use financial ratios in comparing sector spreads and sector leverage. Generally speaking, higher leverage should imply higher credit risk and thus wider spreads. A portfolio manager could therefore compare sectors on a spread-versus-leverage basis to identify relative value opportunities.

Sector- and rating-specific spread curves are a useful tool in guiding decision making for top-down sector allocations. A comparison of curves combined with an investor’s view could lead to credit portfolio positioning based on a view that a specific credit spread curve will flatten or steepen, or that two spread curves will converge or diverge. For example, Exhibit 22 shows the divergence in industrial versus health care spreads for BBB rated US issuers over the first half of 2020 during the COVID-19 pandemic. The flatter industrial credit spread curve reflects that sector’s relatively weak credit outlook versus health care over the period.

Exhibit 22:

US BBB Industrial versus Health Care Spreads (bps p.a.)

Source: Bloomberg

Factor-Based Credit Strategies

While the top-down approach to fixed-income portfolio construction outlined in the previous section grouped investment choices by sector and public ratings, active credit investors are increasingly turning to strategies based on style factors.

Key Factors Affecting Credit Spreads

Factor investing has long been applied in equity markets as noted in earlier lessons, but the application of systematic risk factors such as size, value, and momentum in fixed-income markets is relatively new. For example, Israel, Palhares, and Richardson (2018) established a framework for evaluating excess corporate bond returns based on a number of characteristics, evaluating their significance in explaining fixed-income returns. The authors found strong evidence of positive risk-adjusted returns to measures of carry, defensive, momentum, and value. Exhibit 23 lists these four factors, their rationale, and the measures used in their analysis.

Exhibit 23:

Selected Fixed-Income Factors

Factor	Rationale	Measures Used
Carry	Expected return measure if POD or aggregate risk premium is unchanged	OAS
Defensive	Empirical research suggests safer low-risk assets deliver higher risk-adjusted returns	Market-based leverage, gross profitability, and low duration
Momentum	Bonds with higher recent returns outperform those with lower recent returns	Trailing six-month excess bond and equity returns
Value	Low market value versus fundamental value indicates greater than expected return	Bond spread less default probability measure, which includes rating, duration, and excess return volatility

The returns represented diversification with respect to common market risk sources such as equity or credit risk premia and are similar in characteristic to those factors shown to be significant in equity markets, with some adjustments. Investigation of the source of returns suggested neither traditional risk exposures nor mispricing provided a comprehensive explanation for the excess returns.

Environmental, Social, and Governance Factors

The growing relevance of environmental, social, and governance (ESG) factors in active portfolio management is evidenced by growing adoption of the Principles for Responsible Investment. This independent body established in partnership with the United Nations to promote ESG factors in investing has more than 3,000 signatories worldwide with more than $100 trillion in assets under management.

Active credit investors usually incorporate ESG factors into portfolio strategies in one of three basic ways:

• The use of screens to either exclude specific industries with less favorable ESG characteristics, such as firearms, tobacco, or coal, or to rule out specific companies or sovereign issuers with ESG-specific ratings below a threshold

• Use of ESG ratings to target issuers within a given sector or rating category with relatively favorable ESG characteristics while matching a specific index risk and return

• Targeting fixed-income investments that directly fund ESG-specific initiatives

ESG-specific ratings for private and public issuers are a key element in the portfolio selection process. The wide range of quantitative and qualitative criteria used to measure ESG attributes and differences in methodology and weighting leads to greater dispersion in ESG versus credit ratings. That said, ESG and credit ratings tend to be positively correlated for two reasons. First, issuers with more financial resources are better able to meet more stringent ESG standards, while those with a greater likelihood of financial distress often face governance or other adverse risks. Second, major rating agencies now explicitly incorporate ESG risks into the traditional credit rating process. In 2019, Moody’s cited ESG risks as a material factor in one-third of its credit rating actions among private sector issuers.

Green bonds are fixed-income instruments that directly fund ESG-related initiatives such as those related to environmental or climate benefits. This rapidly growing segment of the fixed-income market includes corporate, financial institution, and public issuers where bond proceeds are directed to projects that reduce air pollution, recycle post-consumer waste products, underwrite environmental remediation projects, and invest in alternative construction materials for environmentally sustainable buildings. Issuers frequently agree to voluntary guidelines such as the International Capital Market Association’s Green Bond Principles (2018) to ensure that these securities meet investor ESG requirements. Although green bonds usually rank pari passu (or at the same level) with the issuer’s outstanding senior unsecured bonds and therefore reflect similar pricing, the favorable ESG characteristics often result in greater investor demand than for standard debt issues. For example, in October 2020, the European Union issued €17 billion in new 10-year and 20-year debt in its first-ever offering of social bonds to finance its COVID-19 pandemic-related job support program. At nearly 14 times the issuance size, the €233 billion in investor orders for the new bonds represented the largest demand ever for a primary bond issuance.

LIQUIDITY AND TAIL RISK https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563593/liquidity-and-tail-risk-1

Learning Outcomes

• discuss liquidity risk in credit markets and how liquidity risk can be managed in a credit portfolio

• describe how to assess and manage tail risk in credit portfolios

Liquidity Risk

The feasibility and cost of buying and selling fixed-income instruments are important considerations for active investors. Trading volumes and bid–offer costs vary widely across fixed-income markets and regions. For instance, sovereign bonds in large developed markets are highly liquid, usually offering institutional bid–offer spreads in secondary markets for on-the-run securities of less than one basis point during trading hours. Smaller, off-the-run corporate bonds or structured notes, on the other hand, might command bid–offer spreads of 10 bps or more and take days to execute, given that many outstanding bonds do not trade at all on a given trading day.

Consider, for example, the US corporate bond market, wherein a single major issuer might have dozens of outstanding debt tranches of varying tenor, currency, or other feature, each separately traded and identifiable via a specific CUSIP or ISIN (International Securities Identification Number). As mentioned earlier in the curriculum, individual bond issuance and trading has historically taken place in over-the-counter (OTC) markets as opposed to on an exchange. OTC market liquidity rests with individual dealers, their specific portfolio and depth of inventory, and appetite to supply liquidity at a cost. Corporate bonds are traditionally traded on a request-for-quote basis, in which investors reach out to multiple dealers to request a fixed price quote for a specific trade size. The use of electronic trading platforms for bond trading has grown because higher regulatory capital requirements reduced bond inventories among dealers after the 2008–09 global financial crisis. While electronic trading platforms comprised less than one-third of individual corporate bond trading volume as of 2020, trading in bond portfolios and bond ETFs, addressed later in this lesson, has grown in importance.

Transaction cost estimates in bond markets vary significantly from those in equity markets because of market structure differences. Price discovery for infrequently traded individual bonds often begins with matrix pricing (or evaluated pricing) techniques introduced earlier in the curriculum using bonds from similar issuers and actively traded government benchmarks to establish a bond’s fair value. For bonds quoted actively on a request-for-quote system by individual dealers, the effective spread transaction cost statistic introduced in an earlier lesson and shown in Equation 13 provides an estimate of trading cost.

Tradesize×{Tradeprice−(Bid+Ask)/2forbuyorders(Bid+Ask)/2−TradepriceforsellordersTrade size×{Trade price−(Bid+Ask)/2for buy orders(Bid+Ask)/2−Trade price for sell orders13

However, the effective spread is an inadequate gauge of trading costs for positions that are traded in smaller orders over time and/or whose execution affects market spreads. A separate, ex-post liquidity gauge specific to the US corporate bond market is the TRACE (Trade Reporting and Compliance Engine) reporting system introduced in 2002 to track real-time price and volume reporting for bond transactions. Portfolio managers will often review recent TRACE trading activity to gauge the estimated cost of trading a bond position.

Active portfolio managers take several steps in managing the liquidity risk of bond portfolios, given the significant market risk involved in trading less liquid positions. First, active managers will usually favor on-the-run government bonds or most recently issued corporate or other bonds for short-term tactical portfolio positioning, while reserving relatively illiquid positions for buy-and-hold strategies or strategic positioning to minimize expected return erosion due to trading costs. Second, active managers might consider liquid alternatives to individual bond trades to close portfolio gaps where active management adds little value, or to react quickly to rapidly changing markets. These alternatives include CDS outlined later and bond ETFs.

Fixed-income ETFs are liquid, exchange-traded bond portfolios that create and redeem shares using an OTC primary market that exists between a set of institutional investors (or authorized participants) and the ETF sponsor. These ETF shares trade in the secondary market on an exchange, overcoming the liquidity constraints of individual OTC-traded bonds. Bond ETFs have enjoyed significant growth and are available across the credit spectrum as well as for different maturities and in different markets. Although the underlying cash flow exposures are similar, ETFs usually neither mature nor experience duration drift (with the exception of target maturity ETFs) as do individual bonds. As ETF sponsors target a specific index or profile, ETFs offer relatively constant portfolio duration and pay variable monthly interest based on the overall portfolio. Active credit managers use ETFs to quickly and efficiently overweight or underweight exposures in rapidly changing markets and to take on strategic exposure in segments of the market where individual or bottom-up bond selection is less of a focus.

When relatively illiquid bond positions are purchased or sold over longer periods, portfolio managers might consider hedging strategies such as asset swaps to mitigate the benchmark risk of a portfolio position as outlined in the following example.

EXAMPLE 19

Using Asset Swaps to Manage Liquidity Risk

1. Recall the earlier example of a United States–based issuer with the following option-free bonds outstanding:

   Outstanding Debt	Term	Coupon	Price	YTM
   2-year issue	2	4.25%	106.7	0.864%
   5-year issue	5	3.25%	106.0	1.984%
   15-year issue	15	2.75%	91.0	3.528%

   Assume the investor instead holds a US$50 million face value position in the outstanding 15-year bond. Historical TRACE data suggest an average $5 million daily trading volume in the 15-year bond. Which of the following statements best describes how the issuer might use an asset swap to manage the benchmark interest rate risk associated with liquidating this bond position?

   1. The investor should enter into an asset swap where he receives fixed and pays floating, unwinding the swap position once the bond position is sold.

   2. The investor should enter into an asset swap where he pays fixed and receives floating, unwinding the swap position once the bond position is sold.

   3. The investor should enter into an asset swap where he pays fixed and receives floating, unwinding the swap position over time in proportion to the amount of the bond sold.

   Solution:

   The correct answer is C. Because the investor’s bond position represents a long position (i.e., long both spread duration and benchmark duration), the best hedge would be a short-duration (or pay-fixed swap) position rather than A. As for B, the hedge unwind occurs once the bond position is sold rather than over time, which exposes the investor to benchmark interest rate risk for the portion of the bond sold. The proportional swap unwind in C ensures that the offsetting swap position matches the benchmark interest rate risk of the bond.

Tail Risk

Extreme adverse outcomes that exceed those to be expected from a normally distributed portfolio are often referred to as tail events. In the context of active fixed-income management, the measurement and management of tail risk involves stress testing a portfolio’s value based on the key fixed-income returns factors in Equation 1. In an earlier lesson on measuring and managing market risk, value at risk (VaR) was introduced as a measure of the minimum portfolio loss expected to occur over a given time period at a specific confidence level. For example, a 5% daily VaR of €8.7 million implies that a portfolio manager should expect a daily portfolio loss of at least €8.7 million on 5% of all trading days. Assuming normally distributed portfolio returns, the 5% confidence level translates to an outcome at least 1.65 standard deviations below the mean, while a 1% confidence interval lies at least 2.33 standard deviations below the mean. Risk managers often use expected returns, volatilities, and correlations to estimate parametric VaR in addition to either historical simulation or Monte Carlo methods. The following example shows a simple parametric VaR calculation for a bond position.

EXAMPLE 20

Fixed-Rate Bond VaR

1. Consider the earlier case of an investor holding $50 million face value of a 15-year bond with a semiannual coupon of 2.75%, a current YTM of 3.528%, and a price of 91 per 100 of face value. What is the VaR for the full bond price at a 99% confidence interval for one month if annualized daily yield volatility is 1.75% (175 bps) and we assume that interest rates are normally distributed?

   Solution:

   First, we must adjust the annualized yield volatility to reflect a one-month period instead. The time interval under consideration is 1/12th of a year, and therefore the volatility measure is 0.00505 (1.75% ×, which for a 99% confidence interval equals 117.7 bps = (0.00505 × 2.33 standard deviations. We may quantify the bond’s market value change using either a duration approximation or the actual price change as follows. We may use the Excel MDURATION to solve for the bond’s duration as 12.025. We can therefore approximate the change in bond value using the familiar (-ModDur x ∆Yield) expression as $6,439,808 = ($50 million x 0.91 x (-12.025 x .0177)). We can also use the Excel PRICE function to directly calculate the new price of 79.132 and multiply the price change of 11.868 by the face value to get $5,934,133.

The simplicity and transparency of VaR can be misleading if it is used as a tool for quantifying tail risk for several reasons. First, VaR tends to underestimate the frequency and severity of extreme adverse events. It also fails to capture the downside correlation and liquidity risks associated with market stress scenarios. Finally, although VaR addresses minimum loss for a specific confidence level, it fails to quantify the average or expected loss under an extreme adverse market scenario. Conditional value at risk (CVaR), or expected loss, measures the average loss over a specific time period conditional on that loss exceeding the VaR threshold. While computationally more complex and beyond the scope of this lesson, CVaR is often measured using historical simulation or Monte Carlo techniques. Two related measures of portfolio VaR include incremental and relative measures. For example, an analyst seeking to measure the impact of adding or removing a portfolio position might use an incremental VaR (or partial VaR) calculation for this purpose. As mentioned in an earlier lesson, an investor could use relative VaR to measure the expected tracking error versus a benchmark portfolio by calculating VaR (or CVaR) based on a portfolio containing the active positions minus the benchmark holdings under a market stress scenario.

EXAMPLE 21

VaR Measures

1. An active fixed-income manager is considering increasing an overweight portfolio allocation to BBB rated health care issuers versus a targeted index. Which of the following VaR measures is the most appropriate to evaluate the impact of this decision on overall portfolio VaR?

   1. Incremental VaR

   2. Relative VaR

   3. CVaR

   Solution:

   The correct answer is A. Incremental VaR measures the impact of a specific portfolio position change on VaR, while relative VaR in answer B evaluates all active portfolio positions versus the benchmark index and could be important for an active fixed-income mandate that aims to beat an index once the portfolio change has been made. CVaR in C measures a portfolio’s average loss over a specific time period conditional on that loss exceeding the VaR threshold.

Tail risk assessment is typically conducted using one of the three methods summarized in Exhibit 24.

Exhibit 24:

Methods to Assess Portfolio Tail Risk

Method	Description	Advantages	Disadvantages
Parametric Method	Uses expected value and standard deviation of risk factors assuming normal distribution	Simple and transparent calculation	Not well suited for non-normally distributed returns or option-based portfolios
Historical Simulation	Prices existing portfolio using historical parameters and ranking results	Actual results, accommodates options, with no probability distribution assumed	Highly dependent on historical period and repetition of historical market trends
Monte Carlo Analysis	Involves generating random outcomes using portfolio measures and sensitivities	Randomly generated results from a probability distribution, accommodates options	Highly dependent on model assumptions and less transparent

Hypothetical scenario analyses are often used to supplement these three methods of analysis to test portfolio vulnerabilities to specific portfolio parameter changes over time.

In addition to portfolio measures of duration and convexity as a basis for portfolio value changes, analytical models often rely on implied volatility parameters for benchmark interest rates and currencies, such as swaption volatility or currency option volatility, respectively, while reduced form or structural credit models incorporating CDS or equity volatility can be used to model expected spread volatility. Finally, term structure models introduced in an earlier lesson that incorporate interest rate volatility and drift in an equilibrium or arbitrage-free framework are frequently incorporated to simulate term structure changes over time.

Once tail risk under an extreme market scenario has been quantified, it is important to weigh this exposure against other binding portfolio constraints and to take steps to manage the downside risk. For example, a leveraged portfolio might face forced liquidation of certain bond positions beyond a certain tail risk threshold. Alternatively, a defined-benefit pension fund manager might be required to increase plan contributions if extreme market moves cause plan funding status to fall below a statutory minimum. Finally, a bank treasury officer could face increased regulatory capital requirements if adverse market changes under a stress test show significant portfolio losses.

A fixed-income portfolio manager can reduce tail risk by establishing position limits, risk budgeting, or using similar techniques designed to reduce portfolio concentration or to cap portfolio risk exposure to certain issuers, credit ratings, or regions. Alternatively, a portfolio manager might consider the use of derivatives to protect against downside portfolio risk. For example, the manager could consider purchasing a swaption (or the right to enter an interest rate swap at a pre-agreed rate in the future) or a credit default swaption (the right to purchase credit protection on an issuer or index at a strike rate in the future) to protect against the risk of benchmark YTM changes or credit spread changes, respectively. However, each of these strategies requires an upfront premium that will reduce excess portfolio spread over time. In addition, establishing these hedges in a distressed market will greatly increase hedging cost because of higher option volatility, so the manager must weigh these hedging costs against a risk mitigation strategy to determine the best course of action.

SYNTHETIC CREDIT STRATEGIES

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563603/synthetic-credit-strategies-1

Learning Outcome

• discuss the use of credit default swap strategies in active fixed-income portfolio management

As outlined in an earlier lesson, a CDS is the basic building block for strategies to manage credit risk separately from interest rate risk. CDS are often more liquid than an issuer’s underlying bonds, enabling investors to take long or short positions, access maturities, and establish other exposures unavailable in cash markets with a smaller cash outlay than direct bond purchases.

Exhibit 25 shows CDS contract mechanics under which a protection “buyer” purchases credit protection from a protection “seller.” Each contract references a specific issuer (or issuers) as well as credit event terms that, when triggered, lead to a settlement payment equal to the LGD multiplied by the contract notional amount from the seller to the buyer.

Exhibit 25:

CDS Mechanics

CDS contracts are usually quoted on an issuer’s CDS spread, which corresponds to a price equal to the present value difference between the CDS spread and a fixed coupon rate on the notional amount over the contract life. Fixed CDS coupon rates of 1% for investment-grade issuers and 5% for high-yield issuers were established when the International Swaps and Derivatives Association standardized CDS market conventions following the 2008 financial crisis. CDS pricing models discount future payments by the swap zero curve multiplied by the hazard rate, or the likelihood that an issuer credit event will occur given that it has not already occurred in a prior period.

The CDS price at contract inception or on a coupon payment date as a percentage of notional can be approximated using Equation 14:CDS Price ≈ 1 + ((Fixed Coupon − CDS Spread) × EffSpreadDurCDS)14where CDS Spread is the issuer’s current CDS market spread and EffSpreadDurCDS is the CDS contract’s effective spread duration (sometimes referred to as CDS DV01). At contract inception, the protection buyer must either make a payment to or receive a payment from the protection seller equal to the CDS contract price difference from par, as shown in Exhibit 26.

Exhibit 26:

Upfront Payment at CDS Contract Inception

Description	Upfront Premium
CDS Spread = Fixed Coupon	None
CDS Spread < Fixed Coupon	Protection buyer receives ((Fixed Coupon - CDS Spread) × EffSpreadDurCDS)
CDS Spread > Fixed Coupon	Protection buyer pays ((CDS Spread - Fixed Coupon) × EffSpreadDurCDS)

CDS contracts have similarities to both bonds and interest rate swaps. As with a cash bond priced at a discount when its coupon is below current market rates, the protection seller is entitled to an upfront payment in exchange for accepting a fixed coupon below the CDS market spread. As with a standard interest rate swap, a CDS contract priced at par has a zero net present value, and the notional is not exchanged but rather serves as a basis for spread and settlement calculations.

EXAMPLE 22

CDS Price and Price Changes

An investor seeks to purchase credit protection under a five-year CDS contract at a CDS market spread of 0.50% p.a. for an investment-grade issuer with an estimated effective spread duration (EffSpreadDurCDS) of 4.75.

1. Determine whether the investor must pay or receive an upfront amount upon CDS contract inception and calculate the difference from par.

   Solution to 1:

   Because investment-grade CDS contracts have a fixed coupon of 1.00% p.a. versus the 0.50% p.a. CDS market spread, the investor buying protection should receive the difference from par upfront in exchange for paying an “above market” coupon under the contract. Calculate the estimated difference using Equation 14 ((Fixed Coupon − CDS Spread) × EffSpreadDurCDS)) with CDS Spread of 0.50%, Fixed Coupon of 1.00%, and EffSpreadDurCDS equal to 4.75.

   Upfront premium: 2.375% of CDS notional (= (1.00% − 0.50%) × 4.75).

2. Calculate the change in contract price if the CDS spread rises to 0.60% p.a. and interpret the impact of the change on the protection buyer.

   Solution to 2:

   Calculate the upfront premium using Equation 14 and a 0.60% spread.

   Upfront premium: 1.90% of CDS notional (= (1.00% − 0.60%) × 4.75).

   The protection buyer realizes a mark-to-market gain equal to 0.475% (2.375% − 1.90%) of the CDS contract notional because of the wider CDS spread.

CDS price changes for a given CDS spread change can be quantified using the contract’s effective spread duration:

∆(CDS Price) ≈ − (∆(CDS Spread) × EffSpreadDurCDS)15

Active fixed-income portfolio managers buy or sell CDS protection across issuers, maturities, and/or sectors to alter portfolio exposure, as illustrated in the following example.

EXAMPLE 23

Credit Underweight Using CDS

1. A European-based fixed-income manager intends to underweight exposure to a BBB rated French media and telecommunications issuer. She observes that the issuer’s current on-the-run five-year CDS contract is trading at a spread of 110 bps p.a. with an EffSpreadDurCDS of 4.595. Which position should she take in the CDS market? Calculate the result if spreads widen to 125 bps for a €10 million notional position.

   Solution:

   The manager can underweight the issuer’s credit by purchasing protection in the CDS market. This short risk position will realize a gain if the issuer’s spreads widen. For example, if the issuer’s credit spreads widen from 110 bps p.a. to 125 bps p.a., we can estimate the change in CDS contract value by multiplying (−∆(CDS Spread) × EffSpreadDurCDS) from Equation 15 by the CDS notional to get €68,925 (=−€10,000,000 × (−0.15% × 4.595)).

While CDS contracts are available across maturities, the five-year tenor is generally the most frequently traded contract. Exhibit 27 summarizes the most common CDS strategies used in practice.

Exhibit 27:

Credit Derivative–Based Alternatives to Corporate Bonds

Instrument	Description	Targeted Return	Portfolio Impact
Single-Name CDS	Protection buyer pays premium to seller in exchange for payment if credit event occurs	Buyer gains and seller loses if single-name credit spread widens or credit event occurs	Short (buyer) or long (seller) single-name credit spread exposure
Index-Based CDS	Protection buyer pays premium in exchange for partial payment if credit event occurs for index member	Buyer gains and seller loses if index member spreads widen or if credit event occurs	Short (buyer) or long (seller) index-based credit spread exposure
Payer Option on CDS Index	Option buyer pays premium for right to buy protection (“pay” coupons) on CDS index contract at a future date	Max (Credit spread at expiration − CDS Credit Spread Strike, 0) − Option Premium	Short CDS index-based credit spread exposure
Receiver Option on CDS Index	Option buyer pays premium for right to sell protection (“receive” coupons) on CDS index contract at a future date	Max (CDS Credit Spread Strike − CDS Credit Spread at expiration, 0) − Option Premium	Long CDS index-based credit spread exposure

Single-name reference entities include both private corporations and sovereign borrowers. Several CDS indexes are available across regions and often also offer subindexes covering a particular sector or borrower type. For example, the Markit CDX North American Investment Grade index consists of 125 equally weighted CDS contracts on entities, including six subindexes (High Volatility, Consumer Cyclical, Energy, Financials, Industrial, and Telecom, Media, and Technology).

CDS strategies are commonly used by active fixed-income portfolio managers to over- or underweight credit spread exposure to individual issuers, specific sectors, or borrower types. As with benchmark yield curves, CDS portfolio positioning strategies are usually based on expected changes in the credit curve level, slope, or shape. The credit curve referred to here is the CDS curve, or the plot of CDS spreads across maturities for a single reference entity or index, rather than the fitted credit spread curves addressed earlier. This might involve an investor taking a long or short CDS position in one issuer or issuer type, or a long or short position overweighting one reference entity or group of entities and underweighting another. Investors using CDS strategies to hedge bond portfolios must always consider the potential impact of basis changes on the strategy over the investment horizon.

Fixed-income ETFs offer derivatives such as futures and options that are different from CDS contracts. As with bond futures, ETF futures are a contract to take future delivery of an ETF and trade on a price rather than a spread basis. Because underlying ETF prices are derived from all-in bond yields held by the fund, ETF derivative prices change with changes in both benchmark rates and credit spreads.

CDS long–short strategies based on spread level are appropriate for both bottom-up and top-down approaches. Assume, for example, that an investor believes that issuer A’s credit spreads will likely narrow versus those of issuer B. To capitalize on this view in the cash market, the investor would first source A’s individual bonds for purchase and then seek a duration-matched amount of issuer B’s bonds to borrow and sell short. The existence of a liquid single-name CDS market for both issuers allows the investor to simply sell protection on A and purchase protection on B for the same notional and tenor.

EXAMPLE 24

CDS Long–Short Strategies

1. Consider the investor from the prior example who sought to underweight a French media and telecommunications issuer. Assume instead that the investor seeks to maintain a constant media and telecommunications credit allocation by overweighting a BBB rated German media and telecommunications competitor. CDS contract details are as follows:

   Issuer	Tenor	CDS Spread	EffSpreadDurCDS
   French Media & Telecoms Issuer	5 years	110 bps	4.697
   German Media & Telecoms Issuer	5 years	130 bps	4.669

   Describe an appropriate long–short CDS strategy to meet this goal, and calculate the investor’s return if the French issuer’s spreads widen by 10 bps and those of the German issuer narrow by 25 bps based on €10 million notional contracts.

   Solution:

   The manager purchases protection on the French issuer and simultaneously sells protection on the German issuer. Use (−∆(CDS Spread) × EffSpreadDurCDS) from Equation 15 multiplied by the CDS notional to solve for changes in the short and long risk positions:

   Short risk (French issuer): €46,970 (= −€10,000,000 × (−0.10% × 4.697))

   Long risk (German issuer): €116,725 (= €10,000,000 × (−(−0.25%) × 4.669))

   The total gain on the long–short strategy is €163,695 (= €46,970 + €116,725).

A similar long–short strategy can be applied under a top-down approach. For example, an investor might overweight (underweight) a specific sector given an expectation of narrower (wider) spread levels versus the total portfolio by selling (buying) protection on a CDS subindex contract. Alternatively, assume an active manager expects a weaker economy and a widening of high-yield versus investment-grade credit spread levels. The manager can capitalize on this view by buying five-year protection on a high-yield CDS index and selling protection on an investment-grade CDS index for the same tenor. Standardized CDS contracts eliminate the impact of duration difference, liquidity, and other factors that arise under a similar strategy in the cash bond market.

CDS long–short strategies based on expected credit curve slope changes involve CDS curve trades. For example, an upward-sloping credit curve implies relatively low near-term expected default probability that rises over time. An investor might expect an issuer’s CDS curve to steepen if its near-term default probability declines as a result of higher than expected profits and stable leverage. This investor can capitalize on this view by selling short-term protection using a single-name CDS contract and buying long-term protection on that same reference entity. As shown in the following example, capitalizing on spread changes for different maturities requires duration matching of the positions, as in the case of benchmark yield curve strategies.

EXAMPLE 25

Duration-Weighted Single-Name CDS Curve Steepener

1. Returning to our earlier example of the German media and telecommunications issuer, the investor decides instead to position her portfolio for a steepening of the issuer’s credit curve using the CDS market. Details of on-the-run 5- and 10-year CDS contracts outstanding are as follows.

   Issuer	Tenor	CDS Spread	EffSpreadDurCDS
   German Media & Telecoms Issuer	5 years	130 bps	4.669
   German Media & Telecoms Issuer	10 years	175 bps	8.680

   Describe an appropriate long–short CDS strategy to meet this goal assuming a €10,000,000 10-year CDS contract notional. Calculate the investor’s return if the 5-year spreads rise 10 bps and the 10-year spreads rise 20 bps.

   Solution:

   A steeper credit curve implies that ((CDS Spread)10yr − (CDS Spread)5yr) will increase. The appropriate long–short strategy to position for this change is to purchase protection based on the 10-year, €10,000,000 notional and to sell protection on an equivalent duration 5-year CDS contract.

   1. Calculate the 5-year CDS contract notional that matches the basis point value (BPV) of a 10-year, €10,000,000 CDS (BPV10yr = EffSpreadDur10yrCDS × notional) using the effective spread duration ratio of 1.859 (EffSpreadDur10yrCDS/EffSpreadDur5yrCDS = 8.68/4.669) multiplied by €10,000,000 to get €18,590,000.

      Confirm this equivalence by comparing BPV5yr and BPV10yr:BPV5yr: €8,680 = €18,590,000 × 4.669/10,000BPV10yr: €8,680 = €10,000,000 × 8.68/10,000

   2. Calculate portfolio return for a 10 bp increase in 5-year CDS spreads and a 20 bp increase in 10-year CDS spreads using Equation 15 (−∆(CDS Spread) × EffSpreadDurCDS) multiplied by the CDS notional.5 year (long risk): −€86,800 (= €18,590,000 × (−0.1% × 4.669))10 year (short risk): €173,600 (=−€10,000,000 × (−0.2% × 8.68))Net portfolio gain: €86,800 = €173,600 − €86,800

The same curve strategy just described applies to expected credit curve slope changes for a CDS index or subindex. For instance, an investor who believes the economy is nearing the end of a growth cycle might expect the CDS curve for industrial issuers to flatten amid rising near-term credit spreads. Under this expected scenario, an investor purchases short-term CDS subindex protection on industrials and sells long-term protection on the same subindex to capitalize on a flattening view. Alternatively, an investor taking a top-down approach who shares a similar bearish economic view might consider a flattening trade for an entire CDS index.

Additional CDS strategies seek to either capitalize on the basis difference between CDS and cash bonds or take advantage of specific events that affect CDS spreads and curves. As noted earlier, basis differences arise from a number of factors but are also due to differences in liquidity across derivative and cash markets, a detailed treatment of which is beyond the scope of this lesson. Corporate events that influence CDS spreads by affecting bondholders differently from shareholders include mergers and acquisitions and leveraged buyouts, both of which are addressed elsewhere in the curriculum.

CREDIT SPREAD CURVE STRATEGIES

Learning Outcome

• discuss various portfolio positioning strategies that managers can use to implement a specific credit spread view

Earlier in the lesson, we established that the credit cycle is a key driver of credit spread changes across maturities and ratings. The probability of issuer default and severity of loss over the cycle must be considered within the context of an overall market view. For example, positively sloped credit spread curves suggest relatively low near-term default probability, a view consistent with stable or rising future inflation and relatively strong expected economic growth. Investor demand for higher credit spreads for assuming the risk of downgrade or default for longer periods also contributes to a positive slope.

The level and slope of credit curves change over the economic cycle. Early in an expansion, as profits rise and defaults remain high, high-yield spreads remain elevated and well above investment-grade spreads, which often exhibit a flat to inverted spread curve. As an expansion progresses, lower defaults and increased profits cause short-term high-yield and investment-grade spreads to decline and credit spread curves to steepen. Credit curve steepening continues as economic growth peaks amid higher leverage and inflation expectations. As economic growth slows or the economy enters a recession, credit spreads rise, and spread curves flatten, with the high-yield curve inverting in some instances amid falling profitability and rising defaults. Although no two credit cycles are alike, Exhibit 28 presents a stylized view of these credit spread curve level and shape changes for investment-grade (IG) and high-yield (HY) issuers over the economic cycle.

Exhibit 28:

Credit Spread Curves over the Economic Cycle

Active credit managers often incorporate the credit cycle into economic growth and inflation forecasts, which are then translated into sector- and issuer-specific views driving specific credit curve level and slope expectations using the bottom-up or top-down approaches outlined earlier. If these forecasts coincide with current credit spread curves, managers will choose active credit strategies consistent with static or stable credit market conditions. However, if an investor’s views differ from what today’s credit curve implies about future defaults and the severity of credit loss, they will position the portfolio to generate excess return based on this divergent view within investment mandate constraints using the cash and derivative strategies outlined in the following section.

Static Credit Spread Curve Strategies

An active credit manager might believe that current credit spreads are reasonably priced and that credit curves will remain stable or unchanged over an investment horizon while credit defaults and annual loss rates remain low. Exhibit 29 shows that a manager could position a portfolio to generate excess return in this scenario by either lowering the portfolio’s average credit rating or adding credit spread duration by investing in longer-dated bonds with a similar rating to the current portfolio.

In the first case, a portfolio tilt toward lower-rated bonds will increase expected spread return, as seen in Equation 10 (E [ExcessSpread] ≈ Spread0 − (EffSpreadDur × ΔSpread) − (POD × LGD)) if Spread, POD, and LGD remain stable. The shift from an average A rated to BBB rated portfolio in Exhibit 29 is an extension of an earlier case (Example 10) that quantified corporate versus government bond roll-down returns as the YTM difference assuming constant spreads and default rates.

Exhibit 29:

Buy-and-Hold Strategies under a Static Credit Curve

The increase in credit spread duration in the second case involves a “buy and hold” or “carry and roll down” approach familiar from the earlier yield curve strategies lesson. The first involves buying risky bonds with durations above the benchmark without active trading during a subsequent period. If the relationship between long- and short-term credit spreads remains stable over the investment horizon, the manager is rewarded with greater return from the higher spread duration. “Rolling down” the credit curve not only generates incremental coupon income (adjusted over time for any price difference from par) due to wider spreads but also adds return from the passage of time and the investor’s ability to sell the shorter-maturity position in the future at a lower credit spread at the end of the investment horizon. The following example illustrates this second case, shown in Exhibit 29.

EXAMPLE 26

Adding Credit Duration under a Static Credit Curve

1. A Sydney-based investor notes the following available option-free bonds for an A rated Australian issuer:

   Debt Term	Coupon	YTM	Price
   5 years	1.00%	1.00%	100
   10 years	1.35%	1.25%	100.937
   15 years	2.00%	1.95%	100.648

   The 5-year, 10-year, and 15-year Australian government bonds have YTMs and coupons of 0.50%, 0.75%, and 1.10%, respectively, and both corporate and government bonds have a semiannual coupon. As an active manager who expects stable benchmark yields and credit spreads over the next six months, the investor decides to overweight (by AUD50,000,000 in face value) the issuer’s 15-year versus 10-year bond for that period. Calculate the return to the investor of the roll-down strategy in AUD and estimate the returns attributable to benchmark yield versus credit spread changes.

   Solution:

   To estimate credit curve roll-down returns, we must solve for the first two return components from Equation 1 (Coupon income +/− Roll-down return) and separate the impact of benchmark yield versus credit spread changes.

   Calculate price appreciation using the difference between current bond prices and those in six months using the Excel PV function (= −PV(rate, nper, pmt, FV, type)) where “rate” is the interest rate per period (0.01225/2), “nper” is the number of periods (19), “pmt” is the periodic coupon (1.35/2), “FV” is future value (100), and “type” (0) involves payments made at the end of each period.

   10-year: Initial price: 100.937

   Price in six months: 101.118 (= −PV (0.01225/2, 19, 1.35/2, 100, 0))

   Price appreciation: $89,660 (= (101.118 − 100.937)/100.937 × $50 million)

   Because the yield spread curve is flat at 0.50%, the full $89,660 price change in the 10-year is benchmark yield curve roll down.

   15-year: Initial price: 100.648

   Price in six months: 101.517 (= −PV (0.0188/2, 29, 1, 100, 0))

   Price appreciation: $431,700 (= (101.517 − 100.648)/100.648 × $50 million)

   Because the 0.07% decline in YTM is estimated to be equally attributable to benchmark yield and yield spread changes, each is assumed equal to $215,850.

   1. Solve for the respective 5-year, 10-year, and 15-year bond credit spreads. Yield spread and G-spread are reasonable approximations because the bonds are option-free, with maturities closely aligned to par government securities.5-year spread: 0.50% (= 1.00% − 0.50%)10-year spread: 0.50% (= 1.25% − 0.75%)15-year spread: 0.85% (= 1.95% − 1.10%)

   2. Solve for 6-month expected returns of the 10-year versus 15-year bond:

      1. Incremental coupon income = $162,500 (= (2.00% − 1.35%)/2 × $50 million)

         Debt Tenor	Coupon	Yield Spread	Total Coupon Income	Coupon (Benchmark Yield)	Coupon (Credit Spread)
         10 years	1.35%	0.50%	$337,500	$187,500	$150,000
         15 years	2.00%	0.85%	$500,000	$275,000	$225,000

         Divide incremental coupon into benchmark and credit spread components:Income due to benchmark yields: $87,500 = $275,000 − $187,500Income due to credit spreads: $75,000 = $225,000 − $150,000

      2. Price appreciation is determined by the bond’s price today and in six months’ time based on unchanged benchmark rates. In six months, the 10-year and 15-year positions will be 9.5-year and 14.5-year bonds, respectively, at a yield and yield spread point along the curve. Estimate all-in YTMs and yield spreads using interpolation to arrive at the following results:

         Debt Tenor	Date	Coupon	All-In Yield	Benchmark Yield	Yield Spread
         5 years	Today	1.00%	1.00%	0.50%	0.50%
         10 years	Today	1.35%	1.25%	0.75%	0.50%
         15 years	Today	2.00%	1.95%	1.10%	0.85%
         9.5 years	In six months	1.35%	1.225%	0.725%	0.50%
         14.5 years	In six months	2.00%	1.88%	1.065%	0.815%

   3. Incremental income due to price appreciation is therefore $342,040 (=$431,700 − $89,660), of which $215,850 is attributable to credit spread changes.

   In total, the incremental roll-down strategy generates $504,540 (=$342,040 + 162,500), of which $290,850 (= $215,850 + $75,000) is estimated to be due to credit spread curve roll down.

Derivative-based credit strategies to add credit spread duration or increase credit exposure include selling CDS single-name or index protection for longer maturities or lower credit quality or using a long–short approach to achieve a similar objective.

EXAMPLE 27

Using CDS for a Static Fixed-Income Credit Strategy

1. Returning to our earlier example of the investment-grade German media and telecommunications issuer, the investor decides instead to overweight exposure to this name by taking a long risk position in the single-name 10-year CDS market for one year. Details of today’s 5-year and 10-year CDS contracts are as follows.

   Issuer	Tenor	CDS Spread	EffSpreadDurCDS
   German Media & Telecoms Issuer	5 years	130 bps	4.669
   German Media & Telecoms Issuer	10 years	175 bps	8.680

   Describe the roll-down strategy using CDS and calculate the one-year return in euros on a €10,000,000 position assuming an annual coupon payment and a 9-year EffSpreadDurCDS of 7.91.

   Solution:

   The investor sells 10-year CDS protection on the German issuer to overweight exposure and terminates the position in one year. As with the bond example, the sold protection strategy generates coupon income if the issuer does not default and price appreciation if credit spreads decline over time.

   1. The fixed coupon received at the end of one year equals the notional multiplied by the standard 1% investment-grade coupon for the period, or €100,000, or €10,000,000 × 1.00% for one year.

   2. Estimate the CDS price change over one year by interpolating the 9-year issuer spread under a static credit curve assumption.

   Solve for the 5-year and 10-year CDS spread weights in the 9-year spread interpolation calculation.

   5-year CDS weight = w5 = 20% (= (10 − 9)/(10 − 5))

   10-year CDS weight = w10 = 80% (or (1 − w5)

   Note that (w5 × 5) + (w10 × 10) = 9

   The 9-year spread is a weighted average of 5- and 10-year CDS spreads.

   CDS Spread9yr = w5 × CDS Spread5yr + w10 × CDS Spread10yr

   1.66% (=1.30% × 0.2 + 1.75% × 0.8)

   Estimate the CDS contract price change by multiplying the change in CDS price from Equation 14 (CDS Price ≈ 1 + ((Fixed Coupon − CDS Spread) × EffSpreadDurCDS)) by the CDS notional.

   10-year CDS per €1 par: 0.934 = (1 + (−0.75% × 8.68))

   9-year CDS per €1 par: 0.947794 = (1 + (−0.66% × 7.91))

   Calculate the price appreciation by multiplying the price change by the contract notional to get €128,940 (= (0.947794 − 0.9349) × €10,000,000).

   Total return equals the sum of the coupon income and price appreciation, or €228,940 (= €100,000 + €128,940).

Dynamic Credit Spread Curve Strategies

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563613/credit-spread-curve-strategies-1

Active credit managers seek to capitalize on divergent market views using cash-based or derivative strategies related to specific issuers, sectors, or the overall credit market over the credit cycle given anticipated credit curve changes across both maturities and rating categories. The following examples demonstrate how an active manager might position a credit portfolio in anticipation of these changes to generate excess return.

EXAMPLE 28

Tactical Credit Strategies − Economic Slowdown Scenario

An active credit portfolio manager considers the following corporate bond portfolio choices familiar from an earlier example:

Rating Category	Current OAS	Expected Loss (POD × LGD)	EffSpreadDur
A	1.05%	0.06%	5.5
Baa	1.35%	0.30%	6.0
Ba	2.45%	0.60%	4.5
B	3.50%	3.00%	4.0

The investor anticipates an economic slowdown in the next year that will have a greater adverse impact on lower-rated issuers. Assume that an index portfolio is equally allocated across all four rating categories, while the investor chooses a tactical portfolio combining equal long positions in the investment-grade (A and Baa) bonds and short positions in the high-yield (Ba and B) bonds.

1. Calculate excess spread on the index and tactical portfolios assuming no change in spreads over the next year (ignoring spread duration changes).

   Solution to 1:

   The following table summarizes expected excess returns E [ExcessSpreadReturn] ≈ Spread0 −(EffSpreadDur × ΔSpread) − (POD × LGD) for each of the four rating categories with no change in spreads. For example, expected excess return for rating category A is 0.99% (=1.05% − (5.5 × 0) − 0.06%).

   Rating Category	Excess Spread Return
   A	0.99%
   Baa	1.05%
   Ba	1.85%
   B	0.50%

   Solve for the equally weighted versus tactical portfolios as follows:

   Equally weighted index: 1.10% (= (0.99% + 1.05% + 1.85% + 0.50%)/4)

   Tactical portfolio: −0.16% (= (0.99% + 1.05%)/2 − (1.85% + 0.50%)/2)

2. Calculate excess spread under an economic downturn scenario for the index and tactical portfolios where both OAS and expected loss rise 50% for investment-grade bonds and double for high-yield bonds.

   Solution to 2:

   The following table summarizes expected excess returns E [ExcessSpreadReturn] ≈ Spread0 −(EffSpreadDur × ΔSpread) − (POD × LGD) for each of the four rating categories with the expected 50% increase in both OAS and expected loss under the slowdown scenario. For example, expected excess return for rating category A is −1.928% (=1.05% − (5.5 × 0.525%) − 0.09%).

   Rating Category	E(OAS)	E(Expected Loss)	E(Excess Spread Return)
   A	1.575%	0.09%	−1.928%
   Baa	2.025%	0.45%	−3.150%
   Ba	4.900%	1.20%	−9.775%
   B	7.000%	6.00%	−16.500%

   Solve for the equally weighted versus tactical portfolios as follows.

   Equally weighted index: −7.84% = (−1.928% − 3.150% −9.775% − 16.500%)/4) Tactical portfolio: +10.6% = (−1.928% − 3.150%)/2 − (−9.775% − 16.500%)/2)

This example assumes that an active manager is able to source and borrow the necessary Ba- and B rated bonds to sell short at no cost. However, in practice, the availability and cost of shorting bonds vary over the economic cycle, and shorting bonds is often far more difficult and costly during an economic slowdown. The synthetic, CDS-based strategy in the following example targets a similar objective.

EXAMPLE 29

Synthetic Credit Strategies: Economic Slowdown Scenario

As in the prior example, an active fixed-income manager anticipates an economic slowdown in the next year with a greater adverse impact on lower\u0002rated issuers. The manager chooses a tactical CDX (credit default swap index) strategy combining positions in investment-grade and high-yield CDX contracts to capitalize on this view. The current market information for investment-grade and high-yield CDX contracts is as follows:

CDX Contract	Tenor	CDS Spread	EffSpreadDurCDS
CDX IG Index	5 years	120 bps	4.67
CDX HY Index	5 years	300 bps	4.65

Assume that both CDX contracts have a $10,000,000 notional with premiums paid annually, and that the EffSpreadDurCDS for the CDX IG and CDX HY contracts in one year are 3.78 and 3.76, respectively. We ignore the time value of money for purposes of this example.

1. Describe the appropriate tactical CDX strategy and calculate the one-year return assuming no change in credit spread levels.

   Solution to 1:

   The investor should sell protection on the CDX IG Index and buy protection on the CDX HY Index. Current CDS prices are estimated by multiplying EffSpreadDurCDS by the spread difference from the standard rates of 1% and 5%, respectively:

   CDX HY: 109.3 per $100 face value, or 1.093 (= 1 + (5.00% − 3.00%) × 4.65)

   CDX IG: 99.066 per $100 face value, or 0.99066 (= 1 + (1.00% − 1.20%)× 4.67)

   Since the investor is both buying HY protection at a premium to par (that is, agreeing to pay the 5% standard coupon while the underlying CDS spread is 3.00%) and selling IG protection at a discount from par (or agreeing to receive the standard 1.00% while the underlying index spread is 1.20%), the investor will receive an upfront payment for entering both positions as follows:1,023,400 = [$10,000,000 × (1.093 – 1)] + [$10,000,000 × (1 –0.99066)]

   In one year, the return is measured by combining the net CDX coupon income or expense with the price appreciation assuming no spread change. Because the investor is long protection CDX HY and short protection CDX IG, the net annual premium paid is $400,000 (= $10,000,000 × (−5.00% + 1.00%).

   CDX HY: 107.52 per $100 face value, or 1.0752 (=1 + (2.00% × 3.76))

   CDX IG: 99.244 per $100 face value, or 0.99244 (=1 + (−0.20% × 3.78))

   The investor has a $17,800 loss from the CDX IG position (= (0.99244 − 0.99066) × $10,000,000) and a $178,000 gain from the short CDX HY position (1.0752 − 1.093) × −$10,000,000). So the one-year loss is −17,800 + 178,000 − 400,000 = −$239,800.

2. Calculate the one-year return on the tactical CDX strategy under an economic downturn scenario in which investment-grade credit spreads rise by 50% and high-yield credit spreads double.

   Recall from Equation 14 that the price of a CDS contract may be approximated as follows:CDS Price ≈ 1 + ((Fixed Coupon – CDS Spread) × EffSpreadDurCDS)

   Solution to 2:

   Initial CDS prices are derived exactly as in Question 1:

   CDX HY: 109.3 per $100 face value, or 1.093 (= 1 + (2.00% × 4.65))

   CDX IG: 99.066 per $100 face value, or 0.9966 (= 1 + (−0.2% × 34.67))

   In one year, the return is measured by combining the coupon income with the price appreciation given the rise in the CDX IG spread to 1.80% and the CDX HY spread to 6.00%. In this case, the investor takes the same position to that of Question 1, namely long CDX HY protection (short risk) and short CDX IG protection (long risk), so the net annual premium paid remains $400,000 (=$10,000,000 × (5.00% − 1.00%). Respective CDS prices in one year are as follows:

   CDX HY: 96.24 per $100 face value, or 0.9624 (=1 + (−1.00% × 3.76))

   CDX IG: 96.976 per $100 face value, or 0.96976 (=1 + (−0.80% × 3.78)

   When offsetting the transaction in one year, the investor suffers a $209,000 loss from the short CDX IG position ((0.99066 – 0.96976) × –$10,000,000) and benefits from a $1,306,000 gain from offsetting the CDX HY position (1.093 – 0.9624) × $10,000,000). Subtracting the $400,000 net premium paid results in a one-year gain from the strategy of $697,000 (= $1,306,000 – $209,000 - $400,000) under the second scenario.

The early expansion phase of the credit cycle is usually characterized by rising profits and falling leverage, as shown earlier in Exhibit 8, increasing cash flow coverage available to service outstanding debt. This reduction in the likelihood of near-term financial distress leads to both lower credit spread levels and a steeper credit curve, an effect that is more pronounced for lower-rated issuers in cyclical industries. The following examples illustrate how an active manager might capitalize on this credit cycle view in cash and synthetic markets.

EXAMPLE 30

Tactical Credit Strategies: Economic Recovery Scenario

1. A long-only active credit manager faces similar corporate bond portfolio choices to those in an earlier example:

   Rating Category	OAS	EffSpreadDur	Expected Loss
   A	1.40%	5.5	0.10%
   Baa	2.00%	6.0	0.30%
   Ba	3.75%	4.5	1.00%
   B	5.50%	4.0	4.50%

   Given an expectation that an economic rebound will cause both credit spreads and expected loss rates to fall by one-third, an active manager decides to tilt her credit portfolio toward high yield. Compare the impact of this rebound scenario on an active portfolio (33.3% invested in each of the Ba and B bond categories, with the remaining 33.3% split evenly between A and Baa) versus on an equally weighted passive portfolio.

   Solution:

   The economic rebound scenario results in the following new OAS and expected losses, with expected excess returns E [ExcessSpread] ≈ Spread0 −(EffSpreadDur × ΔSpread) − (POD × LGD) in the far right column:

   Rating Category	E(OAS)	E(Expected Loss)	E(Excess Spread)
   A	0.933%	0.07%	3.898%
   Baa	1.333%	0.20%	5.80%
   Ba	2.50%	0.67%	8.705%
   B	3.667%	3.00%	9.832%

   Solve for the passive (equally weighted) portfolio returns versus tactical portfolio returns.

   Passive portfolio return: 7.095% (= (3.898% + 5.80% + 8.705% + 9.832%)/4)

   Tactical portfolio return: 7.795% (=3.898%/6 + 5.80%/6 + 8.705%/3 + 9.832%/3).

EXAMPLE 31

Synthetic Credit Strategies: Economic Recovery Scenario

1. As in the prior example, an active fixed-income manager anticipates an economic rebound that is expected to cause high-yield credit curve steepening. The manager chooses a tactical CDX strategy combining 5-year and 10-year credit positions to capitalize on this view. Current market information for these high-yield CDX contracts is as follows:

   CDX Contract	Tenor	CDS Spread	EffSpreadDurCDS
   CDX HY Index	5 years	450 bps	4.637
   CDX HY Index	10 years	375 bps	8.656

   Describe an appropriate duration-neutral portfolio positioning strategy to capitalize on this view using these CDX HY contracts. Calculate the return assuming that 5-year CDX spreads immediately fall by 175 bps and 10-year spreads decline by 25 bps for an equivalent $10,000,000 notional on the 10-year CDX index contract.

   Solution:

   The appropriate strategy is to sell protection on the 5-year CDX HY and buy protection on the 10-year CDX HY.

   1. Calculate the 5-year CDS contract notional that matches the BPV of a 10-year, $10,000,000 CDS (BPV10yr = EffSpreadDur10yrCDS × notional) using the effective spread duration ratio of 1.8667 (EffSpreadDur10yrCDS/EffSpreadDur5yrCDS = 8.656/4.637) multiplied by $10,000,000 to get $18,667,000.

      Confirm this equivalence by comparing BPV5yr and BPV10yr:BPV5yr: $8,656 = $18,667,000 × 4.637/10,000BPV10yr: $8,656 = $10,000,000 × 8.656/10,000

   2. Calculate portfolio return for a 175 bp decline in 5-year CDX HY spreads and a 25 bp decline in 10-year CDX HY spreads using Equation 15 (−∆(CDS Spread) × EffSpreadDurCDS) multiplied by the CDS notional as follows:CDX HY 5 year: $1,514,780 = (−(−1.75%) × 4.637) × $18,667,000CDX HY 10-year: −$216,400 = (−(−0.25%) × 8.656) × (−$10,000,000) Portfolio gain: $1,298,380 = $1,514,780 − $216,400.

      Note that this equals the contract BPV of $8,656 multiplied by the 150 bp credit curve steepening.

GLOBAL CREDIT STRATEGIES

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563623/global-credit-strategies-1

Learning Outcome

• discuss considerations in constructing and managing portfolios across international credit markets

While yield curve strategies across currencies were covered in an earlier lesson, we now turn to cross-border fixed-income investments in which investors face the risk that they will not receive interest and principal cash flows as expected. Investors distinguish between international credit markets in developed market countries versus emerging or frontier markets. Fixed-income markets in developed countries usually have well-established and liquid derivative and other capital markets and feature a broad range of private and public debt issuers with bonds denominated in a freely floating domestic or other major currency. Emerging or frontier fixed-income markets on the other hand are often dominated by sovereign issuers, state-owned or controlled enterprises, banks, and producers operating in a dominant domestic industry such as basic commodities. As some emerging economies face concentrated risk to a particular commodity or industry, investments across sovereign, bank, and private sector debt could offer little to no diversification. While many emerging-market bonds are denominated in a restricted domestic currency with varying degrees of liquidity, the sovereign government and a select few domestic issuers often issue global bonds in a major foreign currency such as US dollars or euros.

Credit strategies across countries must take these and other individual market differences into consideration. For example, in the case of developed markets, sector composition differences exist. A far higher percentage of the US fixed-income market (and roughly one-third of the Bloomberg Barclays US Aggregate Bond Index) comprises mortgage-backed and other asset-backed instruments versus other developed markets. Investors in developed European and Asian markets seeking commercial or residential real estate exposure might instead consider covered bonds or indirect exposure via bank bonds in markets where securitization is less prevalent. International accounting standards differences between the International Accounting Standards Board’s International Financial Reporting Standards and US GAAP in such areas as inventory recognition, restricted cash, and cash flow definitions require adjustment for financial ratio comparisons across jurisdictions. Finally, while most developed markets face common macroeconomic factors that influence the bond term premium and expected returns, such as inflation, monetary policy, and economic growth, differences in the timing and magnitude of market changes, as well as the credit cycle across countries, are often reflected in interest rate differentials, exchange rates, and credit spreads.

EXAMPLE 32

Credit Strategies across Developed Markets

1. An active United States–based credit manager is offered similar US corporate bond portfolio choices to those in an earlier example:

   Rating Category	OAS	EffSpreadDur	Expected Loss
   A	1.40%	5.5	0.10%
   Baa	2.00%	6.0	0.30%
   Ba	3.75%	4.5	1.00%
   B	5.50%	4.0	4.50%

   As in the earlier case, the manager expects an economic rebound but now believes that European economies will experience a stronger recovery than the United States. In particular, European high-yield credit spreads are expected to narrow by 25% in the near term, the euro is expected to appreciate 1% against the US dollar, and all US credit spreads and expected loss rates are expected to decline just 10% over the same period. The euro-denominated 5-year European iTraxx Crossover index (iTraxx-Xover) of liquid high-yield issuers (with a 5% fixed premium) is currently trading at 400 bps with an EffSpreadDurCDS of 4.25.

   Describe the position the manager would take in iTraxx-Xover to capitalize on the stronger European rebound, and calculate the expected excess return percentage assuming an equally weighted allocation to US corporate bonds and an iTraxx-Xover position that matches that of the US high-yield bond allocation.

   Solution:

   To capitalize on expected greater euro spread tightening, the manager would sell protection on the iTraxx-Xover index. To calculate expected return, first consider the US corporate bond portfolio. The economic rebound scenario results in the following new OAS and expected losses for the portfolio, with expected excess returns E [ExcessSpread] ≈ Spread0 −(EffSpreadDur × ΔSpread) − (POD × LGD) in the far right column:

   Rating Category	E(OAS)	E(Expected Loss)	E(Excess Spread)
   A	1.26%	0.09%	2.08%
   Baa	1.80%	0.27%	3.93%
   Ba	3.38%	0.90%	4.54%
   B	4.95%	4.05%	3.65%

   Return on the equally weighted portfolio is equal to 3.30% (= (2.08% + 2.93% + 4.54% + 3.65%)/4). We can estimate the initial iTraxx-Xover price by subtracting the product of EffSpreadDurCDS and the difference between the standard coupon (5%) from the market premium of 400 bps as follows:

   Original iTraxx-Xover 5-year: 95.75 per $104.25, or 1.0425 (=1 − (4.25 × 1.00%))

   If European high-yield spreads tighten by 25%, the iTraxx-Xover premium narrows by 100 bps to 300 bps, and the protection seller realizes a gain:

   New iTraxx-Xover 5-year: 91.50 per $108.5, or 1.085 (=1 − (4.25 × 2.00%))

   We can calculate the percentage return on the iTraxx-Xover investment in euro terms by dividing the price change by the initial price to get 4.077 (= (1.085 − 1.0425)/1.0425). For a United States–based investor, we must convert the euro return to US dollars as described in an earlier lesson:

   RDC = (1 + RFC) (1 + RFX) − 1

   RDC and RFC are the domestic and foreign currency returns in percent, and RFX is the percentage change of the domestic versus foreign currency.

   We solve for US dollar iTraxx-Xover returns as 5.118% (= (1 + 4.077%) × (1 + 1.00%) − 1). Given that iTraxx-Xover carries a weight equal to one-half of the US corporate bond portfolio, the strategy returns 5.86% (or 3.30% + 5.118%/2).

Emerging markets are characterized by higher, more volatile, and less balanced economic growth than developed markets, often in addition to greater geopolitical risk, currency restrictions, and capital controls. Sovereign credit risk is therefore a critical starting point in considering fixed-income investments in emerging markets, where both the ability and willingness of issuers to repay debt is of importance. An earlier lesson outlined in detail sovereign credit risk considerations such as a country’s institutional and economic profile, use of monetary and fiscal policy, the exchange rate regime, and external debt status and outlook.

Institutional considerations include political stability, institutional transparency, and adherence to property rights and contract law. Geopolitical risks include such factors as potential conflicts and trade relations, which in some instances could have a greater impact on emerging markets whose economies are highly dependent on energy or other commodity exports. As mentioned earlier, ESG factors are key elements for sustainable, balanced, long-term economic growth.

As sovereign governments tax economic activity within their borders to repay interest and principal, key financial ratios used to assess and compare sovereign creditworthiness are usually measured as a percentage of GDP. For example, government debt to GDP and the annual government budget deficit (or surplus) as a percentage of GDP are common measures of indebtedness and fiscal stability, respectively, for both developed and emerging markets.

Finally, a country’s exchange rate regime is a critical element of monetary and external flexibility. Freely floating currency regimes that allow a currency to be held in reserve outside the country enable sovereign governments to pursue an independent and flexible monetary policy. Restrictive or fixed-rate regimes limit policy effectiveness, magnifying the impact of economic crises and increasing the likelihood of financial distress. Emerging markets are usually characterized by non-reserve currency regimes with significant external debt denominated in major foreign currencies, leading analysts to incorporate external debt to GDP and currency reserves as a percentage of GDP as key leverage and liquidity measures of creditworthiness, respectively.

The Bloomberg Sovereign Risk (SRSK) model shown in Exhibit 30 combines quantitative and qualitative factors such as external debt to GDP, currency reserves, GDP growth, and political risk to estimate a sovereign issuer’s one-year POD. Similar to the DRSK model discussed earlier, the SRSK model allows users to change model inputs and also derives a “model” CDS spread, which could be compared to the market CDS spread.

Exhibit 30:

Bloomberg SRSK Screen

In this example, Costa Rica has a 1.28% one-year default risk and a model CDS spread well below the market CDS spread.

EXAMPLE 33

Sovereign Risk Factors for Emerging Markets

1. A financial analyst is considering the likelihood that an emerging market sovereign issuer of US dollar–denominated bonds is able to meet its interest payments over the next 12 months. Which of the following financial ratios is most appropriate to assess the sovereign borrower’s liquidity position?

   1. Government budget deficit/GDP

   2. External debt/GDP

   3. Currency reserves/GDP

   Solution:

   The correct answer is C. The government budget deficit as a percentage of GDP is a gauge of fiscal stability for the domestic economy, while the external debt-to-GDP ratio is a measure of financial leverage to foreign lenders. Currency reserves as a percentage of GDP measure the available liquidity in foreign currency to meet external obligations.

Several additional considerations are important for investors in emerging market bonds issued by private companies. First, although some local companies might have partial private ownership and publicly traded equity, the sovereign government could exercise controlling influence on the business, including replacing management or ownership groups.

Credit quality in the emerging market credit universe exhibits a high concentration in lower investment-grade and upper high-yield ratings categories. This concentration of credit ratings is largely a reflection of the sovereign ratings of emerging markets but also reflects the fact that a “sovereign ceiling” is usually applied to corporate issuers globally. This ceiling implies that a company’s rating is typically no higher than the sovereign credit rating of its domicile.

Finally, relative liquidity conditions and currency volatility are key considerations for international credit investors. In emerging markets, liquidity is often constrained because of a relatively small number of bonds that trade regularly, resulting in investors demanding higher premiums for holding emerging market credit securities. Local bond markets might seem highly liquid and can exceed the trading volume of the local stock exchanges, but such high trading volume could also be inflated by interbank trading by local banks and retail investors. Currency volatility can be particularly significant in emerging markets as a result of restrictive currency regimes and derivative markets. Higher YTMs available in emerging market currencies versus developed markets typically suggest that these emerging currencies will depreciate over time. That said, emerging markets offer investors the opportunity to exploit divergence from interest rate parity conditions (known as the forward rate bias) by investing in higher-yielding currencies, as addressed in earlier lessons. Although temporary deviations from a fixed exchange rate are possible under such regimes, what is more common during economic crises is exchange rate regime change, central bank intervention, and/or devaluation. The following example demonstrates how such factors are considered in emerging market credit strategies.

EXAMPLE 34

Emerging Market Credit Strategy

1. An active United States–based investor is considering a portfolio allocation to the bonds of a major commodities producer headquartered in an emerging market economy. The issuer is a major exporter, and commodity exports comprise a significant proportion of the country’s economic growth. Describe how the investor would decide between purchasing a higher-yielding, local-currency-denominated bond and a lower-yielding, US-dollar-denominated bond with otherwise similar features.

   Solution:

   A United States–based investor seeking to maximize US-dollar-denominated return must consider the relationship between the higher local currency bond YTM, the lower US dollar bond YTM, and the local currency’s expected depreciation (or appreciation) versus the US dollar over the investment horizon. While uncovered interest rate parity suggests that local currency depreciation versus the US dollar would offset any benefit of a higher YTM, an investor with a bullish view of the emerging economy’s growth prospects would benefit from forward rate bias and earn a higher return in US dollar terms from an unhedged investment in the local currency bond if the local currency were to depreciate less than expected under interest rate parity conditions.

STRUCTURED CREDIT

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563633/structured-credit-1

Learning Outcome

• describe the use of structured financial instruments as an alternative to corporate bonds in credit portfolios

Active managers have access to a wide array of credit management tools beyond individual fixed-income securities that include structured financial instruments. These alternatives to direct bond investments in corporate bonds introduced in earlier lessons are summarized in Exhibit 31.

Exhibit 31:

Structured Alternatives to Individual Bonds

Instrument	Description	Exposure	Portfolio Applications
Collateralized Debt Obligations (CDOs)	Fixed-income securities backed by a diversified pool of debt obligations	Redistribute portfolio debt cash flows across ratings spectrum	Create tailored portfolio-based debt exposure categories/profiles unavailable in the cash bond market
Collateralized Loan Obligations (CLOs)	Fixed-income securities backed by a diversified pool of floating-rate leveraged loan obligations	Redistribute portfolio loan cash flows across ratings spectrum	Create tailored portfolio-based loan and interest rate exposure profiles unavailable in the cash bond market
Mortgage-Backed Securities (MBS)	Fixed-income securities backed by a pool of commercial or residential mortgage loans	Provide portfolio-based exposure to real estate cash flows	Offer active managers exposure to real estate and to volatility (prepayment/extension risk) unavailable in the cash bond market
Asset-Backed Securities (ABS)	Fixed-income securities backed by a pool of credit card, auto, and other loans	Provide portfolio-based exposure to consumer loan cash flows	Offer active managers direct exposure to consumer loans and to volatility unavailable in the cash bond market
Covered Bonds	Senior debt obligations backed by pool of commercial/residential mortgages or public sector assets	Provide portfolio-based exposure to real estate cash flows with recourse to issuer	Offer active managers direct exposure to consumer loans and to real estate/public sector cash flows unavailable in the cash bond market

Structured financial instruments can offer active credit managers the ability to access fixed-income cash flows such as commercial or residential real estate, enhance returns by increasing portfolio exposure to interest rate volatility (via mortgage prepayment and extension risk), and add debt exposure created by the redistribution of default risk into different tranches across the credit spectrum. Exhibit 32 shows an illustrative example of the tranching that characterizes ABS and CDO transactions. In this case, the ABS issuer is a special purpose vehicle (SPV) that owns the underlying asset pool and issues debt across several tranches backed by the asset pool cash flows.

Exhibit 32:

Illustrative Tranching Example

An earlier lesson addressed the redistribution of default risk from the underlying asset pool. This is achieved by establishing higher-rated tranches via internal credit enhancement or overcollateralization, with successively lower-rated tranches absorbing a greater proportion of the associated default risk. An active investor might overweight default risk by choosing a lower-rated ABS tranche based on a tactical view. For example, such an investor might anticipate lower-than-expected defaults or believe the credit cycle is in recovery mode and that lower-rated tranches will experience greater spread tightening than higher-rated tranches. Alternatively, a portfolio manager might underweight credit exposure using a higher-rated tranche in a downturn.

While covered bonds offer real estate cash flow exposure similar to that of ABS, given the dual recourse (i.e., to both the issuing financial institution and the underlying asset pool), as well as the substitution of non-performing assets, covered bonds usually involve lower credit risk and a lower yield. The following examples demonstrate the role of structured products in active credit portfolios.

EXAMPLE 35

The Role of Structured Products in Active Credit Management

1. An active credit manager anticipates an economic slowdown led by a decline in residential housing prices. Which of the following portfolio positioning strategies involving structured products is the most appropriate to consider under this scenario?

   1. Shift exposure from an A rated tranche of a credit card ABS transaction to a BB rated tranche

   2. Increase exposure to an A rated CDO tranche and reduce exposure to a BBB rated CDO tranche

   3. Increase exposure to an A rated MBS tranche and decrease exposure to a BBB rated MBS tranche

   Solution to 1:

   The correct answer is C. As the housing sector slows and default rates rise, credit spreads of lower-rated MBS tend to widen by more than those of higher-rated MBS. The investor retains exposure to real estate cash flows while reducing exposure to spread widening. The shift to a BB rated credit card ABS tranche increases credit exposure, while the switch from BBB rated to A rated CDOs represents a reduction in overall market risk rather than a more targeted underweight, as in C.

2. An active fixed-income portfolio manager expects an economic recovery in the near term to be accompanied by rising short-term rates and a flatter benchmark yield curve. Which of the following strategies best positions an active manager to capitalize on this scenario?

   1. Increase exposure to covered bonds and decrease exposure to MBS

   2. Shift exposure from an A rated CDO tranche to a BBB rated CLO tranche

   3. Shift exposure from a BB rated tranche of an automotive ABS transaction to a A rated tranche

   Solution to 2:

   The correct answer is B. Economic recovery is typically associated with lower defaults and greater credit spread tightening among lower-rated issuers and debt tranches. CLO tranches benefit more from short-term rate rises than CDOs because CLOs comprise leveraged loans based on MRRs plus a credit spread. As for A, a shift to covered bonds from MBS reduces credit risk because of the dual recourse and substitutability of collateral characteristics of covered bonds. In C, credit exposure is reduced, limiting the benefit from credit spread reduction within the portfolio.

FIXED-INCOME ANALYTICS

https://study.cfainstitute.org/app/cfa-institute-program-level-iii-for-august-2024#read/study_task/2563643/fixed-income-analytics-1

Learning Outcome

• describe key inputs, outputs, and considerations in using analytical tools to manage fixed-income portfolios

Fixed-income analytical tools continue to adapt not only to technological change but also to the market and regulatory environment within which active fixed-income practitioners operate. The inputs and outputs of these models have become more complex as market participants integrate tasks across operational duties and portfolio decision making and execution. These tasks include portfolio construction, risk analytics, trading and settlement, cash and collateral management, daily valuation, portfolio accounting, and regulatory reporting.

Primary inputs for fixed-income models include all long and short cash bond and derivative positions, repurchase agreements, and cash across currencies. Fixed-income security inputs use CUSIP or ISIN identifiers to capture all relevant features such as interest and principal payment dates, day count conventions, and put–call features. Portfolio derivative and repo position inputs also include details of such agreements, such as settlement dates, option strike prices, and collateral terms necessary to satisfy derivative counterparty or clearing requirements based on market changes.

Real-time market data feeds usually sourced from vendors via application programming interfaces include spot and forward rates, credit curves, implied volatilities, and exchange rates that are used to value historical, existing, and potential future new portfolio positions. These tools value inactively traded fixed-income instruments using matrix pricing (or evaluated pricing) based on observable liquid benchmark YTMs of similar maturity and duration and credit spreads of actively traded bonds with comparable times to maturity, credit quality, and sector. Additional model inputs include index subscriptions, ESG and credit ratings, and issuer balance sheet data. In contrast to more static equity indexes, fixed-income indexes are subject to constant change as a result of both new debt issuance and bond maturities as well as ratings changes, bond callability, and prepayment.

Model assumptions include user-defined parameters such as term structure models, investment time horizon, VaR methodology, historical and/or specific market scenarios, and portfolio filters that could involve inclusion or exclusion of specific sectors or a minimum ESG rating threshold for consideration.

Fixed-income analytical model outputs support each stage of the active portfolio management process, namely portfolio selection and construction, risk analysis of existing and prospective portfolio positions, and trading and position management. A portfolio summary or landing page typically aggregates current portfolio risk and return across sectors, ratings, and currencies versus the benchmark index. Model applications supporting research and portfolio construction allow managers to assess the expected change in portfolio performance by including incremental long or short cash bond, derivative, or structured product positions. Portfolio risk dashboards embedded in these tools provide detailed insight into portfolio duration and convexity as well as tail risk. These statistics are often further disaggregated into key rate duration measures for benchmark rates and credit spreads by maturity. VaR and expected shortfall (or CVaR) are calculated based on user threshold and methodology settings. Finally, trading, cash, and position management outputs quantify existing cash positions, anticipated cash inflows and outflows from existing positions, and liquidity risk. Exhibit 33 summarizes the key elements of a fixed-income portfolio analytics tool.

Exhibit 33: Illustrative Portfolio Analytics Tool

CFA1824-R-EXH33.png

CFA1824-R-EXH33b.png

Key considerations for fixed-income analytical tools include both the accuracy of model inputs and assumptions and the degree of alignment between model outputs and specific fixed-income manager objectives.

Bond price and YTM calculations are affected by assumptions related to the term structure of benchmark rates and volatilities and how they change over time based on term structure models. Model outputs are often tailored to match an active manager’s objectives. For example, an index fund manager might seek to minimize the tracking error defined earlier as the deviation of portfolio returns from an index. An active fixed-income manager with fewer constraints might maximize risk-adjusted returns, while estimating and categorizing how each position contributes to active risk taking. For example, performance attribution measures returns from credit, duration, sector, and currency tilts, among other factors. Finally, an active manager facing liability constraints usually models the fixed-income characteristics of obligations to maximize the expected surplus of assets over liabilities. Practitioners applying these tools must both recognize their limitations and anticipate and interpret model results, as in the following example.

EXAMPLE 36

Applying Fixed-Income Analytical Tools

1. An active fixed-income manager is conducting scenario analysis for the MBS component of a portfolio. Which of the following analytical model input changes is most likely to reduce the future value of the MBS subportfolio versus similar option-free bond holdings?

   1. An increase in benchmark yield curve volatility

   2. A decrease in benchmark yield curve volatility

   3. Upward parallel shift in the benchmark yield curve

   Solution:

   The correct answer is A. The value of a bond with an embedded option is equal to the sum of the value of an option-free bond plus the value to the embedded option. The value of the embedded call option owned by the issuer will increase as volatility rises, reducing the value of the MBS versus a similar option-free bond. Answers B and C are more likely to result in an increase in the value of MBS versus an option-free bond.