This article outlines the annual performance differentials between six major index makers: Dow Jones (and DJ Wilshire), Lipper, MSCI, Morningstar, Russell and Standard & Poor's. Mo2re to the point, the article explores "index dependency."
As will be demonstrated, there can be substantial performance differences between indexes that measure the performance of U.S. equities within the same style box. Understandably, there are differences in both methodology and in the definitions of style boxes between different providers. As a result of such differences, performance comparisons between individual funds and benchmark indexes (and comparisons between index funds) can be significantly affected by the choice of index(es) in the comparison. In fact, the impact can be huge: As the data show, the annual differential in performance between two different indexes within the same style box can exceed 4,000 basis points.
My primary raw data source for this study was Morningstar Principia.
We start by examining six different large-cap value indexes: Dow Jones Large Value, Lipper Large-cap Value, Morningstar Large Value, MSCI U.S. Large-cap Value, Russell 1000 Value and the S&P 500/Citigroup Value. In 1998, the difference between the best (maximum) and worst (minimum) performing indexes was 750 basis points (see Figure 5). In 2000, the variance between maximum and minimum performance was 1,086 basis points. Over the 10-year period from 1997-2006, the average annual difference between best- and worst-performing index was 627 basis points. Nine-year data was also included because the Morningstar value and growth indexes were not available until 1998. Over the most recent five years (ending December 2006), the average annual difference between the best- and worst-performing large-cap value index was 481 basis points.
Next, U.S. large-cap blend indexes. The most commonly used benchmark for this space is the S&P 500 Index. Its 10-year average annualized return from 1997-2006 was 8.42 percent. But there are other indexes that measure that particular component (large-cap blend) of the U.S. equity market, as well. For instance, the DJ Wilshire U.S. Large-cap Index had an 8.53 percent annualized return over the same 10 years, a difference of just 11 basis points. The Russell 1000 Index had a return of 8.65 percent, different by 23 basis points. The 10-year annualized return of the MSCI U.S. Large-cap 300 Index was 7.96 percent, or just 46 basis points different than the S&P 500. The 10-year return of the Morningstar Large-cap Index was 7.59 percent, a difference of 83 basis points. Finally, the Lipper Large-cap Core Index had a return of 7.00 percent, or 142 basis points lower than the S&P 500 Index. Clearly, the performance differential between Lipper and S&P represents two different methodologies at work; neither is right or wrong, they are simply different.
In the year 2000 (still looking at the large-cap blend category), there was a 950 basis points difference between the best-performing index (Lipper Large-cap Core) and the worst-performing index (MSCI U.S. Large-cap 300). By contrast, in 2005, there was "only" a 163 basis point difference between the best- and worst-performing index. Over this 10-year period (1997-2006), the average annual difference between the best-performing large-blend index and worst-performing large-blend index was 482 basis points. Over the most recent five years, it was 268 basis points.
Among large-cap growth indexes, the variance among the six index providers expands. In 2000, for instance, there was a 2,134 basis point difference between the Lipper Large-cap Growth Index (-12.2 percent) and the Morningstar Large Growth Index (-33.5 percent). Over the last 10 years, the average annual one-year spread between best-performing large-cap growth index and worst-performing large-cap growth index was 990 basis points - not exactly a small difference. Over the past five years, the average spread between the best- and worst performing index was 557 basis points. The 10-year, nine-year and five-year annualized performance of these 18 large-cap indexes (six per style box) is summarized in Figure 5.
Now, we turn to mid-cap U.S. equity indexes (Figure 6). Among the six value indexes, there was an average annual differential of 900 basis points between best- and worst-performing indexes over the past 10 years, and an annual average of 612 basis points over the past five years. In 1999, there was a 2,048 basis points differential between the Dow Jones Midcap Value Index (-11.0 percent) and the Lipper Mid-cap Value Index (9.5 percent).
Among mid-cap blend indexes, the average annual differential between best- and worst-performing index was 898 basis points over the past 10 years, 902 basis points over the past nine years and 434 basis points over the past five years. From 1998-2000, the average annual differential between the best- and worst-performing mid-cap blend index was nearly 1,800 basis points. Interestingly, the "laggard" index in one year is not necessarily the laggard in a subsequent or prior year (note: this is true for the indexes in all nine style boxes).