Fixed-income Return Attribution Methods
There are three approaches that can be used to perform fixed-income return attribution. These approaches are the exposure decomposition approach (duration-based), the yield curve decomposition (duration-based) and the yield curve composition approach that is full-repricing based.
On this page we discuss all three approach. In practice, special software is used or spreadsheet in Excel are created to perform this kind of analysis.
Exposure decomposition (duration-based)
Exposure decomposition is a top-down approach that utilizes duration to quantify active portfolio manager decisions regarding interest rate decisions relative to the benchmark. Exposure decomposition is best thought of as a process that segments risk by some specific characteristics. In this case, duration is used to quantify the impact on the portfolio resulting from interest rate risk.
In other words, exposure decomposition using duration segments portfolios by their market value weights and assigns securities to duration buckets based on the security’s maturity
Yield curve decomposition (duration-based)
The second approach is yield curve decomposition. This approach can be either top-down or bottom-up and uses both the duration and the yield to maturity (YTM) in computing price return.
total return = % income return + % price return
where the % price return is approximately equal to (-duration – change in YTM). Yield curve decomposition based on duration look at what factors drive returns when YTM changes. When used on both the portfolio and the benchmark, a comparison of the return drivers allows one to determine the impact of active management. Compared to the exposure decomposition we discussed above, yield curve decomposition requires more data and is more complex.
Yield curve decomposition (full repricing)
The third method is full repricing. In this case, securities are repriced based on zero-coupon curves or spot rates. The use of spot rates is also referred to as the full-repricing method. It is the most accurate method of the three. At the same time, it is even more data intensive and more difficult to implement than the two other methods we discussed.
We discussed the three most commonly used methods used to perform fixed-income return attribution.