structured and traditional managers lies in their approach to risk and benchmarks. Structured managers are highly benchmark sensitive and tend to target relatively low levels of tracking error. Further, structured managers usually attempt to hit their lower targets by relying on a relatively large number of small active deviations (i.e., overweights and underweights). By contrast, traditional active managers usually target high ex ante excess returns. Although most do not explicitly target tracking error, their quest for excess returns often results in high ex post active risk. This is because traditional managers usually restrict their active decision making to a small number of relatively large positions. The difference in the magnitude of active positions is key to understanding the risk and performance differences between traditional and structured managers. One major consequence is that traditional managers are less able to achieve symmetry between their bullish and bearish views. Why? Because of the no-short constraint that most institutional investors face. That is, managers can generally overweight a stock by as much as they'd like, but they can only underweight a stock up to its weight in the benchmark. Since traditional managers usually want to implement relatively large active deviations, this constraint is often binding. Whereas they can theoretically overweight their favorite names by as much as they'd like, they can only fully underweight their least favorite names in a few cases (i.e., those where the benchmark weight is large enough to accommodate the desired underweighting). As a result, because overweights and underweights must sum to zero, the no-short constraint effectively hinders a manager's ability to express bullish views. Consequently, the no-short constraint and related lack of symmetry will reduce a traditional manager's potential information ratio. Structured managers, in contrast, can take greater advantage of both their bullish and bearish views. They are able to more fully exploit their views because of their relatively low tracking error targets and their propensity to take a large number of relatively small active deviations. Thus, the no-short constraint is less binding because their desired underweights exceed the benchmark weights less often. A second difference between structured and traditional managers is the emphasis on risk management. With tight tracking error targets, structured managers spend a great deal of time and effort managing risk and eliminating unintended bets-just as a household on a tight budget will be more frugal. Traditional managers, in contrast, feel less constrained by tracking error concerns and spend commensurately less time on risk management. As a result, unintended and uncompensated risks can creep into their portfolios. For example, many traditional managers roughly equal-weight the names in their portfolios. This can produce large overweights in small-cap names and smaller overweights (or even underweights) in large-cap names. The resulting small-cap bias adds uncompensated risk to the portfolio. That is, the overweight in smaller names is driven by the manager's inattention to risk rather than a strong belief that small-cap stocks (as a class) will outperform large-cap stocks. By adding noise to the denominator (tracking error) without increasing the numerator (alpha), this practice reduces the information ratios of traditional managers.