The formula and benchmarks
IR = (Portfolio Return − Benchmark Return) / Tracking Error = Active Return / Tracking Error. If a fund returns 12% versus a benchmark’s 10%, with a tracking error (volatility of active returns) of 4%, the IR is 2% / 4% = 0.50. General benchmarks: IR below 0.0 means consistent underperformance (fire the manager). IR of 0.0–0.25 is poor. IR of 0.25–0.50 is acceptable. IR of 0.50–0.75 is good. IR above 0.75 is exceptional and extremely rare over long periods. The median active equity manager has an IR below 0.0 after fees over most rolling periods.
The fundamental law of active management
Grinold’s Fundamental Law of Active Management provides a framework for decomposing the IR: IR ≈ Information Coefficient (IC) × √Breadth. The IC is the correlation between the manager’s forecasts and actual outcomes (ranging from 0 for no skill to 1.0 for perfect foresight). Breadth is the number of independent bets made per year. This law has a profound implication: a manager with modest skill (IC = 0.05) can generate a strong IR by making many independent bets (e.g. 500 stocks vs 5 concentrated positions). Concentrated managers need much higher IC to generate the same IR. This is why systematic quantitative managers, who may trade thousands of instruments with a small IC, can generate strong IRs.
IR vs Sharpe ratio
The Sharpe ratio evaluates absolute risk-adjusted performance (against the risk-free rate). The IR evaluates relative performance (against a benchmark). A manager who always owns the same 50 stocks, regardless of market conditions, might have an IR of 0.7 relative to their equity benchmark — demonstrating genuine stock-picking skill. But their absolute Sharpe ratio might be low because equity volatility is high. Conversely, a leveraged bond strategy might show a strong Sharpe ratio but a poor IR versus its benchmark because it takes different risk dimensions. Both metrics are needed for complete evaluation.
“The information ratio is the active manager’s report card. Most students are failing. A handful deserve honours. Identifying them in advance is where the real skill lies.”
What this means for you
When evaluating an active fund manager, calculate the IR over the longest available period and across different market environments. Require at least five years of data — short-period IR estimates have enormous statistical uncertainty (a three-year strong IR can reflect luck as easily as skill). Check that tracking error is appropriate for a "active" fund — a closet index tracker with 2% tracking error cannot generate meaningful alpha even with IR of 1.0. And scrutinise whether the IR has been achieved consistently across multiple market regimes, or whether it reflects a single extended period of favourable conditions for the manager’s style.