|
Page 5 of 7
Risk Factor Attribution
The question arises as to where the outperformance of the S&P 500 EWI is derived from. Exhibit 10 shows a style map of the S&P 500 and S&P 500 EWI. The map exhibits the influence of both style and size on the relative performance of the two indices. The performance of the S&P 500 EWI, while consistently in the large cap half of the chart, tends to fall somewhere between the S&P 500 and S&P 400 style indices, thus showing an influence on its return by movements in mid-size stocks. Also, the S&P 500 EWI varies between being more heavily influenced by growth or value factors. However, in the majority of periods it falls on the value side of the chart. Thus, over the last five years the S&P 500 EWI has been influenced both by the size factor and the value factor relative to the S&P 500. This suggests that equal weighting results in unique exposure to a complex and dynamic combination of size and style risk factors. It may be difficult to replicate S&P 500 EWI return outcomes through a simple combination of style and sector indices.
Looking at attribution by sectors also yields results mostly in line with expectations. Exhibits 11 and 12 show attribution by sector for two periods - 1995 through 1999, the major period of significant underperformance for the S&P 500 EWI, and 2000 through 2007, a period of significant outperformance for the S&P 500 EWI.
As would be expected, the majority of the underperformance of the S&P 500 EWI during the late 1990s can be attributed to the Information Technology sector which contributed 42.34% of total underperformance, close to half of total underperformance of 91.39%. During this time Information Technology had the largest return of any sector in both the S&P 500 and S&P 500 EWI. However, the S&P 500 EWI had both a lower weight in this sector and a lower sector return. Health care and Telecommunication Services were also large contributors to underperformance. For the time period from 2000 through 2007 the largest contributors to outperformance were Information Technology and Consumer Discretionary sectors. During this period the attribution was much more spread out between sectors. In fact, all of the sectors attributed positively to the outperformance of the S&P 500 EWI during this time.
Interestingly, looking at the sector attribution also further clarifies the importance of the differences in constituent weights due to equal weighting. For the 1995 through 1999 period the S&P 500 EWI not only underperformed as a whole but also in eight out of 10 sectors. Conversely, for the period of 2000 through 2007, the S&P 500 EWI outperformed in every sector and every sector had a positive attribution. This implies that most of the outperformance is due not to differences in sector weightings but to the sector index returns of the S&P 500 and S&P 500 EWI. However, since the stocks in the indices, and thus in each of the sectors, are the same, the differences are caused solely by the different weighting and rebalancing schemes of the two indices.
|