Ratios     Active Tracking   Traditional Structured Return Error Information Allocation Allocation (bps) (bps) Ratio 0% 100% 65 152 0.42 10 90 66 143 0.46 20 80 68 146 0.46 30 70 69 160 0.43 40 60 71 184 0.38 50 50 72 214 0.34 60 40 74 248 0.30 70 30 75 284 0.27 80 20 77 321 0.24 90 10 78 360 0.22 100 0 80 400 0.20 turn, tracking error, and information ratio at alternative splits between structured and traditional active managers. As with Table 14.5, we have assumed portfolios of two structured managers and four active managers. Consistent with the median values in Table 14.1, we have assumed that each structured manager has an expected information ratio of 0.30, and each traditional manager has an expected information ratio of 0.10. If we further assume that there is no correlation between manager alphas, then the portfolio of two structured managers has an expected information ratio of 0.42, while the portfolio of four traditional managers has an expected information ratio of 0.20. Notice in Table 14.8 that the maximum information ratio is achieved when the portfolio has between 80 percent and 90 percent allocated to structured equities and 10 percent to 20 percent allocated to traditional strategies. This portfolio has an expected information ratio around 0.46, and a tracking error between 143 basis points and 146 basis points. In comparison with Table 14.5, the tracking error for the optimal mix is lower, while the allocation to structured equity strategies is higher. This result should not be surprising given the relative declines in information ratios (from top-quartile to median) for the two strategies. Now, let's suppose the tracking error target for the total U.S. equity program is 200 basis points. Since this target is greater than the tracking error for the optimal portfolio, we know that risk considerations will determine the optimal split between structured and traditional strategies. That is, the allocation to structured strategies will be exactly the same as when we used first-quartile manager information ratios. As Table 14.8 suggests, we will still allocate 55 percent to structured equities and 45 percent to traditional strategies. However, the expected information ratio is now much lower at 0.36, versus 0.81 when we assumed greater skill at manager selection. For a more interesting case, suppose the tracking error target is 100 basis points. Since this target is less than the tracking error of the optimal blend portfolio, we know that we will need to dilute the active risk with passive managers. Table 14.9 contrasts the mix among passive, structured, and traditional managers