What is it?
The Style Benchmark is a combination of indexes that best defines the style of a manager. In other words, it is the combination of indexes that best replicates the historical behavior of the manager by providing the lowest tracking error to the manager.
Why is it useful and how is it built?
For many active managers, a single index benchmark may be too narrow. Choosing multiple indexes to build a blended benchmark can sometimes be more representative as a benchmark than a single index. The group of indexes used can be determined by the user. A few key factors to take note of when selecting the indexes should be:
- They should be exhaustive and include all of the investible assets in the particular asset class you are analyzing.
- They should be mutually exclusive and not have overlapping securities.
Returns-Based Style Analysis
is used to determine the combination of indexes that build out the Style Benchmark. Below is an Asset Allocation graph where the Style Benchmark built for the Alger Spectra A Fund is the composite return series comprised of 55.9% Russell 1000 Growth and 44.1% Russell 2000 Growth.
Is the Style Benchmark better than a single Market Benchmark?
In some cases yes, and we can prove this. One way to test this is using correlation. The higher the correlation the better the benchmark. Below is a picture that shows Performance Attribution pie charts of the Style Benchmark and S&P 500:
The brown section of the pie on the left shows the R-Squared (correlation squared) of the manager to the Style Benchmark as 83.6%, while the light orange section shows the residual at 16.4%. The pie on the right shows the R-Squared of the manager to the S&P 500 at 68.2%, while the residual at 31.8%. The Style Benchmark has a higher R-Squared and lower residual than the S&P 500. The residual measures the variance of the manager’s returns that is not explained by the benchmark.
However, maybe there is a single index, less broad than the S&P 500 that would be better to use. Because the Alger Spectra A Fund is considered a Large Growth Fund, let’s test the Russell 1000 Growth Index. Below shows a picture of this:
We can see that although the R-Squared to the Russell 1000 Growth is higher than it was to the S&P 500, it’s still not higher than the Style Benchmark. The unexplained variance or the residual of the Russell 1000 Growth is greater than that of the Style Benchmark.
This makes sense, because the Style Benchmark is using five independent variables to explain the manager’s performance, not just one. The Style Benchmark is flexible enough to capture most of the systematic factors that affect the returns of the manager. When using a single index, such as the S&P 500 or Russell 1000 Growth, you are unable to adjust for differences in style and market exposure.
For the calculation of Returns-Based Style analysis, please click here