able to react as quickly to changes in realized tracking error as we are able to do with daily estimation. Potentially, two years would have to elapse before we had a reasonable estimate of the portfolio's realized tracking error. Needless to say, a lot can happen in two years. Let's say for argument's sake that monthly data was all we had access to. If we ran a rolling analysis after two years and found the realized tracking error to be well in excess of our expectation of 550 to 1,000 basis points, chances are it would be too late to react to the signals. Potential unidentified flaws in the investment process would be caught too late and could have the unfortunate ability to detract from plan value. The framework of setting tracking error bands is a combination of both art and science. In areas of the market where the valuation transparency is low and market liquidity constrains a manager's ability to react, there needs to be an even greater emphasis on judgment. Emerging equity and high-yield debt markets are two examples that readily come to mind. High-yield markets are typically illiquid, which at times makes the costs of trading prohibitively expensive. If market conditions were such that managers could not trade their portfolios efficiently, then we would expect larger, uncontrolled deviations from the benchmarks. In these cases, we would suggest using wider bands to accommodate the need for a smooth portfolio transition when market volatilities are changing. GREEN SHEET One of the tools risk managers can deploy when managing a large portfolio of investment managers is what we call the "Green Sheet." The Green Sheet is a diagnostic tool developed to help risk managers better understand the active performance and risk drivers at the total plan level. In doing so the Green Sheet allows risk managers to focus their attention on managers and asset classes that are exhibiting performance or risk out of line with expectations. In looking at the sample Green Sheet shown in Table 15.2, we notice that ABC pension fund's tracking error over the past 60 days is 128 basis points, which is far in excess of its 65 to 110 basis point target. As expected, this puts the plan's tracking error in the upper yellow zone. On a stand-alone basis, knowing that ABC's plan has exceeded risk expectations doesn't shed much light for the risk manager about what the potential risk drivers may be. However, the Green Sheet is quick to highlight for the risk manager that most active large and small cap managers are experiencing tracking errors that exceed expectations. For example, small cap growth manager G has a 60-day tracking error two times expectations. In fact, at the asset class level both large caps and small caps are exhibiting large deviations from their targets. Further investigation through the use of a second tool, the risk budget (Table 15.3), shows that our large cap managers are exhibiting higher correlations to one another than expected. We will be talking more about the practical applications of the risk budget in the next section. It is the risk manager's responsibility to spend time understanding the market and portfolio dynamics before initiating a conversation with a portfolio manager. Many times the risk management team will be able to attribute the deviations in active risk away from targets without manager discussions. Factors such as changes