The returns of bonds are influenced by a number of factors: changes in interest rates, changes in the credit ratings of the issuers, and changes in the yield curve. A bond strategy is the management of a bond portfolio either to increase returns based on anticipated changes in these bond-pricing factors or to maintain a certain return regardless of changes in those factors. Bond strategies can be classified as active, passive, hybrid.
Active strategies usually involve bond swaps, liquidating one group of bonds to purchase another group, to take advantage of expected changes in the bond market, either to seek higher returns or to maintain the value of a portfolio. Active strategies are used to take advantage of expected changes in interest rates, yield curve shifts, and changes in the credit ratings of individual issuers.
Passive strategies are used, not so much to maximize returns, but to earn a good return while matching cash flows to expected liabilities or, as in indexing, to minimize transaction and management costs. Pension funds, banks, and insurance companies use passive strategies extensively to match their income with their expected payouts, especially bond immunization strategies and cash flow matching. Generally, the bonds are purchased to achieve a specific investment objective; thereafter, the bond portfolio is monitored and adjusted as needed.
Hybrid strategies are a combination of both active and passive strategies, often employing immunization that may require rebalancing if interest rates change significantly. Hybrid strategies include contingent immunization and combination matching.
Active Bond Strategies
The primary objective of an active strategy is for greater returns, such as buying bonds with longer durations in anticipation of lower long-term interest rates; buying junk bonds in anticipation of economic growth; buying Treasuries when the Federal Reserve is expected to increase the money supply, which it usually does by buying Treasuries. Active selection strategies are based on anticipated interest rate changes, credit changes, and fundamental valuation techniques.
Interest Rate Anticipation Strategies
A rate anticipation strategy involves selecting bonds that will increase the most in value from an expected drop in interest rates. A rate anticipation swap involves selling a group of bonds so that others can be purchased based on the expected change in interest rates. A total return analysis or horizon analysis is conducted to evaluate several strategies using bond portfolios with different durations to see how they would fare under different interest rate changes, based on expected market changes during the investment horizon.
If interest rates are expected to drop, then bonds with longer durations would be purchased, since they would profit most from an interest rate decrease. If rates were expected to increase, then bonds with shorter durations would be purchased, to minimize interest-rate risk. One means of shortening duration is buying cushion bonds, which are callable bonds with a coupon rate that is significantly higher than the current market rate. Cushion bonds generally have a shorter duration because of their call feature and are cheaper to buy, since they generally have a lower market price than a similar bond without the call feature. Rate anticipation strategies generally require a forecast in the yield curve as well since the change in interest rates may not be parallel.
Yield Curve Shifts
Because the yield curve involves a continuum of interest rates, changes in the yield curve can be described as the type of shift that occurs. The types of yield curve shifts that regularly occur include parallel shifts, twisted shifts, and shifts with humpedness. A parallel shift is the simplest kind of shift in which short-, intermediate-, and long-term yields change by the same amount, either up or down. A shift with a twist involves either a flattening or an increasing curvature to the yield curve or it may involve a steepening of the curve where the yield spread becomes either wider or narrower as one progresses from shorter durations to longer durations. A yield shift with humpedness is one where the yields for intermediate durations changes by a different amount from either short- or long-term durations. When intermediate-term yields increase less than either the short- or long-term durations, then that is deemed a positive humpedness, sometimes called a positive butterfly; the obverse, where short- and long-term yields decline more (meaning that bond prices increase faster) than intermediate term yields is negative humpedness or a negative butterfly.
Several bond strategies were designed to profit or maintain value due to a specific change in the yield curve. A ladder strategy is a portfolio with equal allocations for each maturity group. A bullet strategy is a portfolio whose duration is allocated to 1 maturity group. For instance, if interest rates were expected to decline, then a profitable bullet strategy would be one with long-term bonds, which would benefit the most from a decrease in interest rates. A barbell strategy has a concentration in both short and long-term bonds if negative humpedness in the yield curve is expected, in which case, short- and long-term bonds will increase in price faster than the intermediate-term bonds.
There are 2 types of credit investment strategies: quality swaps and credit analysis strategies. Bonds of a higher quality generally have a higher price than those of lower quality of the same maturity. This is based on the creditworthiness of the bond issuer, since the chance of default increases as the creditworthiness of the issuer declines. Consequently, lower quality bonds pay a higher yield. However, the number of bond issues that default tends to increase in recessions and to decline when the economy is growing. Therefore, there tends to be a higher demand for lower quality bonds during economic prosperity so that higher yields can be earned and a higher demand for high quality bonds during recessions, which offers greater safety, sometimes called the flight to quality. Hence, as an economy goes from recession to prosperity, the credit spread between high and low quality bonds tends to narrow; from prosperity to recession, the credit spread widens. Quality swaps usually involve a sector rotation where bonds of a specific quality sector are purchased in anticipation of changes in the economy.
A quality swap profits by selling short the bonds that are expected to decline in price relative to the bonds that are expected to increase in price more, which are bought. A quality swap can be profitable whether interest rates increase or decrease, because profit is made from the spread. If rates increase, the quality spread narrows: the percentage decrease in the price of lower quality bonds will be less than the percentage decrease in the price of higher quality bonds. If rates decrease, then the quality spread expands, because the price percentage increase for the lower quality bonds exceeds the price percentage increase for the higher quality bonds.
Credit Analysis: the Evaluation of Credit Risk
A credit analysis strategy evaluates corporate, municipal, or foreign bonds to anticipate potential changes in credit risk, which will usually result in changes in the issuers' bonds prices. Bonds with forecasted upgrades are bought; bonds with potential downgrades are sold or avoided. Generally, changes in credit risk should be determined before any upgrade or downgrade announcements by the credit rating agencies, such as Moody's, Standard & Poor's, or Fitch; otherwise, it may be too late to take advantage of price changes.
Fundamental Credit Analysis
The credit analysis for corporate bonds includes the following:
- fundamental analysis:
- comparing the company's financial ratios with other firms in the industry, especially the interest coverage ratio, which is earnings before interest and taxes
- leverage, which is long-term debt over total assets
- cash flow, which determines the company's ability to pay interest on debt
- working capital, which is current assets minus current liabilities
- return on equity
- asset and liability analysis, which includes assessing:
- the market values of the company's assets and liabilities
- intangible assets and liabilities, especially unfunded pension liabilities
- the age and condition of the plants
- foreign-currency exposures
- industrial analysis:
- industry and company treads
- assessing the potential growth rate for the industry
- industrial development stage
- the cyclicality of the industry
- labor and union costs and problems
- government regulations
- Indenture analysis: how protective covenants compare with industry norms.
Foreign corporate bonds are also analyzed in the same way, but additional issues include the foreign exchange rate and any risks that may occur because of political, social, and economic changes in the countries where the bonds are issued or where the company is located.
Indenture analysis and economic analysis are used to gauge the riskiness of municipal bonds.
- debt burden
- financial status
- fiscal problems:
- revenues falling below projections
- declines in debt coverage ratios
- increased use of debt reserves
- project cost overruns or delays
- frequent rate increases
- and the state of the general and local economy, such as:
- decreases in population
- loss of large employers
- increases in unemployment
- declines in property values
- the issuance of fewer building permits
Multiple Discriminate Models and the Altman Z-Score
Multiple discriminate models are statistical models that are used to generate a credit score for a bond so that its overall credit quality can be summarized, much as a credit score is used to summarize a consumer's creditworthiness. Multiple discriminate models generally assess the most important factors that determine the creditworthiness of the issuer by applying appropriate weights to each of the factors. A popular scoring model is the Altman Z-score model, developed by Edward Altman in 1968.
Altman Z-score = 1.2 × W + 1.4 × R + 3.3 × E + 0.6 × M + 1.0 × S
- W = ratio of working capital to total assets
- R = ratio of retained earnings to total assets
- E = ratio of earnings before interest and taxes to total assets
- M = market value of equity to total liabilities
- S = ratio of sales to total assets
The Altman Z-score ranges from -5.0 to +20.0. If a company has a Z score above 3.0, then bankruptcy is considered unlikely; lower values indicate an increased risk of business failure.
Altman's double prime model includes the same factors except that the book value rather than the market value of the company to total liabilities is used and the sales to total assets is excluded.
Altman Z''-score = 6.56 × W + 3.26 × R + 6.70 × E + 1.05 × M
Scores above 2.6 are considered creditworthy, while scores below 1.1 indicate an increased risk of business failure. Another score that is commonly used is the Hillegeist Z-score (HS), which is an updated estimate of the Altman Z-score
Hillegeist Z-score = 3.835 + 1.13 × W + 0.0 05 × R + 0.2 69 × E + 0.3 99 × M − 0.033 × S
The 1 year probability of default = 1 ÷ (exp (HS) +1). Like the Altman Z-score, a higher value indicates a lower probability of bankruptcy.
Passive Bond Management Strategies
In contrast to active management, passive bond management strategies usually involve setting up a bond portfolio with specific characteristics that can achieve investment goals without altering the strategy before maturity. The 3 primary passive bond strategies are index matching, cash flow matching, and immunization.
Index matching is constructing a bond portfolio that reflects the returns of a bond index. Some indexes are general, such as the Bloomberg Global Benchmark Bond Indexes or the Barclays Aggregate Indexes; others are specialized, such as the Bloomberg Corporate Bond Indexes. Bond indexing is mainly used to achieve greater returns with lower expenses rather than matching cash flows with liabilities or durations of bonds with liabilities. Because there are several types of indexes, it must 1st be decided which index to replicate. Afterwards, a specific strategy must be selected that best achieves tracking an index, given the resources of the bond fund.
There are several strategies for achieving indexing. Pure bond indexing (aka full replication approach) is to simply buy all the bonds that comprise the index in the same proportions. However, since some indexes consists of thousands of bonds, it may be costly to fully replicate an index, especially for bonds that are thinly traded, which may have high bid/ask spreads. Many bond managers solve this problem by selecting a subset of the index, but the subset may not accurately track the index, thus leading to tracking error.
A cell matching strategy is sometimes used to avoid or minimize tracking error, where the index is decomposed into specific groups with a specific duration, credit rating, sector, and so on, then buying the bonds having the characteristics of each cell. The quantity of bonds selected for each cell can also mirror the proportions in the bond index.
Rather than taking a sample of a bond index to replicate, some bond managers use enhanced bond indexing, where some bonds are actively selected according to some criteria or forecast, in the hope of earning greater returns. Even if the forecast is incorrect, junk bonds tend to increase in price faster when the economy is improving, because the chance of default declines.
Cash Flow Matching
Cash flow matching involves using dedicated portfolios with cash flows that match specific liabilities. Cash flows include not only coupon payments, but also repayments of principal because the bonds matured or they were called. Cash flow matching is often used by institutions such as banks, insurance companies, and pension funds. Liabilities usually vary as to certainty. For a bank, CD liabilities are certain in both amount and in the timing. Sometimes, the liability is certain but the timing is not, such as life insurance payouts and pension distributions. Other liabilities, such as property insurance, are uncertain in both amount and time.
Because most bonds pay semiannual coupons, a cash flow matching strategy is established by 1st constructing a bond portfolio for the last liability, then for the penultimate liability, etc, working backward. The cash flows for earlier liabilities are modified according to the amount of cash flow received from coupons of the already selected bonds. However, special consideration must be given to callable bonds and lower quality bonds, since their cash flows are more uncertain, although the risk can be mitigated with options or other derivatives.
Because bonds cannot usually be selected to exactly match the cash flow of liabilities, some cash must be available to pay liabilities as they come due. Thus, a drawback to cash flow matching over immunization is that the cash is not fully invested.
Bond Immunization Strategies
Bonds have both interest-rate risk and reinvestment risk. Reinvestment is necessary to earn at least a market return. For instance, a 6% coupon bond does not earn a 6% return unless the coupons are reinvested. A 6% 10-year bond with a par value of $1000 earns $600 in interest over the term of the bond. By contrast, $1000 placed in a savings account that earns 6% compounded semiannually would earn $806.11 after 10 years. Hence, to earn a 6% return with a coupon bond, the coupons must be reinvested for at least the 6% rate.
Bond immunization strategies depend on the fact that the interest rate risk and the reinvestment risk are reciprocal: when one increases, the other declines. Increases in interest rates causes increased interest rate risk but lower reinvestment risk. When interest rates decline, bond prices increase, but the cash flows from the bonds can only be reinvested at a lower rate without taking on additional risk. A disadvantage of cash flow matching is that any reinvestment risk cannot be offset by rising bond prices if the bonds are held to maturity.
Classical immunization is the construction of a bond portfolio such that it will have a minimum return regardless of interest rate changes. The immunization strategy has several requirements: no defaults; security prices change only in response to interest rates (for instance, bonds cannot have embedded options); yield curve changes are parallel.
Immunization is more difficult to achieve with bonds with embedded options, such as call provisions, or prepayments made on mortgage-backed securities, since cash flow is more difficult to predict.
The initial value of the bonds must be equal to the present value of the liability using the yield to maturity (YTM) as a discount factor. Otherwise, the modified duration of the portfolio will not match the modified duration of the liability.
Immunization for multiple liabilities is generally achieved by rebalancing, in which bonds are sold, thus freeing up some cash for the current liability, then using the remaining cash to buy bonds that will immunize the portfolio for the later liabilities.
There are several requirements for immunizing multiple liabilities:
- present value of assets must equal the present value of liabilities
- the composite portfolio duration must equal the composite liabilities duration
- the distribution of durations of individual assets must range wider than the distribution of liabilities
Contingent immunization combines active management of a small portion of invested funds, allowing possibly higher returns, while using the remaining portion as an immunized bond portfolio to ensure a minimum return.
The target rate is the lower potential return acceptable to the investor, equal to the market rate for the immunized portion of the portfolio. The cushion spread (aka excess achievable return) is the difference in return rates of the partially immunized portfolio with the remaining part actively managed over a fully immunized portfolio.
The safety margin is the cushion, the actively managed part. While positive, management can continue to actively manage that part of the portfolio. However, if the safety margin declines to 0, then active management ends and only the immunized bond portfolio is maintained. If long-term rates decline, then the safety margin is increased, but any increases in the interest rate will decrease the safety margin, possibly to 0.
The trigger point is the yield level where immunization mode becomes necessary to achieve the target rate or the target return. At this point, active management ceases.
Classical immunization may not work if the shifting yield curve is not parallel or if the duration of assets and liabilities diverge, since durations change as interest rates change and as time passes.
Because duration is the average time to receive ½ of the present value of a bond's cash flow, duration changes with time, even if there are no changes in interest rates. If interest rates do change, then duration will shorten if interest rates increase or lengthen if interest rates decrease.
Hence, to maintain immunization, the portfolio must be rebalanced, which involves bond swaps: adding or subtracting bonds to appropriately modify the bond portfolio duration. Risk can also be mitigated with futures, options, or swaps. The primary drawback to rebalancing is that transaction costs are incurred in buying or selling assets. Hence, transaction costs must be weighed against market risk when bond and liability durations diverge.
Rebalancing can be done with a focus strategy, buying bonds with a duration that matches the liability. Another strategy is the bullet strategy, where bonds are selected that cluster around the duration of the liabilities. A dumbbell strategy can also be pursued, in which bonds of both shorter and longer durations than the liabilities are selected, so that any changes in duration will be covered.
To immunize multiple period liabilities, specific bonds can be purchased that match the specific liabilities or a bond portfolio with the duration equal to the average duration of the liabilities can be selected. Although a bond portfolio with an average duration is easier to manage, buying bonds for each individual liability generally works better. If a portfolio requires frequent rebalancing, a better strategy may be matching cash flows to liabilities instead of durations. Some bond managers use combination matching, or horizon matching, matching early liabilities with cash flows and later liabilities with immunization strategies.
Fundamental Valuation Strategies
Another common strategy to benefit from the bond market is to find and buy underpriced bonds and sell overpriced bonds, based on a fundamental analysis of what prices should be. The intrinsic value of the bond is calculated by discounting the cash flows by a required rate of return that generally depends on market interest rates plus a risk premium for taking on the debt. Naturally, the calculations must account for any embedded options and the credit quality of the issuer and the credit ratings of the bond issue. The required rate of return equals:
Required Rate of Return = Risk-Free Interest Rate + Default Risk Premium + Liquidity Premium + Option Adjusted Spread
Bond managers incorporate the basic characteristics of a bond and its market into models, such as multiple discriminate analysis or CDS analysis, to forecast changes in the credit spread. Option pricing models may also be used to determine the value of embedded options or to forecast changes in the option adjusted spread. Note that forecasts are not nearly as important when using fundamental valuation strategies, since buying bonds that are cheaper because of a temporary mispricing by the market and selling the same type of bonds when the market starts pricing the bonds at their intrinsic value will yield better results regardless of how the bond market changes.
A pure yield pickup strategy (aka substitution swap) is based on the yield pickup swap, which takes advantage of temporary mispricing of bonds, buying bonds that are underpriced relative to the same types of bonds held in the portfolio, thus paying a higher yield, and selling those In the portfolio that are overpriced, which, consequently, pay a lower yield. A pure yield pickup strategy can profit from either a higher coupon income or a higher yield to maturity, or both. However, the bonds must be identical in regard to durations, call features, and default ratings and any other feature that may affect its market value; otherwise, the different prices will probably reflect the differences in credit quality or features rather than a mispricing by the market. Nowadays, it is more difficult to profit from yield pickup strategies, since quant firms are constantly scouring the investment universe for mispriced securities.
A tax swap allows an investor to sell bonds at a loss to offset taxes on capital gains and then repurchase the bonds later with similar but not identical characteristics. The bonds cannot be identical because of wash sale rules that apply to bonds as well stocks. However, the wash sale criteria that apply to bonds are less defined, allowing the purchase of similar bonds with only minor differences.
An intermarket-spread swap is undertaken when the current yield spread between 2 groups of bonds is out of line with their historical yield spread and that the spread is expected to narrow within the investment horizon of the bond portfolio. Spreads exist between bonds of different credit quality, and between differences in features, such as being callable or non-callable, or putable or non-putable. For instance, callable/non-callable bond swaps may be profitable if the spread between the 2 is expected to narrow as interest rates decline. As interest rates decline, callable bonds are limited to their call price, since, if the bonds are called, that is what the bondholder will receive. Hence, the price spread between callable and noncallable bonds is narrower during high interest rates and wider during low interest rates. Thus, as interest rates decline, the callable/noncallable spread increases.