Effective Number of Bets
The Effective Number of Bets (ENB) quantifies the true diversification of a portfolio by measuring how many equally-weighted independent positions the portfolio's concentration is equivalent to. A portfolio with 100 stocks but 90% of the weight in 5 names has an effective number of bets far below 100, revealing the hidden concentration risk.
Overview
Counting the number of positions in a portfolio is a poor measure of diversification. A portfolio of 500 stocks with 50% of its weight in a single name is far less diversified than a portfolio of 20 equally-weighted stocks. The Effective Number of Bets, also known as the Herfindahl effective number, provides a single number that captures the degree of weight concentration.
The metric is derived from the Herfindahl-Hirschman Index (HHI), which is widely used in economics to measure market concentration. In the portfolio context, it answers the question: "If I replaced my portfolio with an equally-weighted portfolio, how many positions would I need to match my current level of concentration?" The answer is the effective number of bets. Meucci (2009) extended this concept to account for the correlation structure of assets, defining the effective number of independent bets based on principal portfolio analysis.
Mathematical Formulation
Core Formula
The Effective Number of Bets is the reciprocal of the Herfindahl-Hirschman Index of portfolio weights:
where is the weight of asset in the portfolio and is the total number of positions. The denominator is the HHI, which ranges from (equal weights) to 1 (single position).
Diversification Ratio
The effective number of bets can be expressed as a fraction of the total number of positions to obtain a diversification ratio:
This ratio ranges from (single-stock concentration) to (perfect equal weighting). A diversification ratio of 50% means the portfolio's concentration is equivalent to an equally-weighted portfolio with half as many positions.
Properties
- Range: . The minimum of 1 occurs when a single asset has all the weight. The maximum of occurs when all assets are equally weighted.
- Equal weights: When for all , then , so .
- Monotonicity: Transferring weight from a larger position to a smaller position always increases .
Worked Example: Equal Weights
Consider a portfolio of assets, each with weight :
As expected, an equally-weighted portfolio of 10 stocks has an effective number of 10 bets, and the diversification ratio is 100%.
Worked Example: Concentrated Portfolio
Now consider a portfolio of assets with one dominant position: and the remaining 9 positions each have :
Despite holding 10 positions, the portfolio's concentration is equivalent to only 3.6 equally-weighted positions. The diversification ratio is , revealing significant concentration risk.
Advantages & Limitations
Advantages
- Intuitive:Directly answers "how many effective positions do I have?" in a single number.
- Concentration detection: Reveals hidden concentration risk that simple position counts miss.
- Simplicity: Trivial to compute -- requires only the portfolio weight vector.
- Monitoring tool: Useful for ongoing portfolio monitoring and compliance with concentration limits.
- Well-established: Based on the Herfindahl-Hirschman Index, a thoroughly studied measure with known statistical properties.
Limitations
- Ignores correlations: The basic formula considers only weights, not the correlation structure of returns. Highly correlated positions provide less diversification than the weight-based measure suggests.
- Weight-only view: Does not account for differences in asset volatility or risk contribution.
- Short positions: The interpretation becomes ambiguous for portfolios with significant short positions.
- No return information: Tells nothing about expected performance -- a highly concentrated portfolio may be concentrated in the best opportunities.
References
- Grinold, R. C., & Kahn, R. N. (2000). Active Portfolio Management (2nd ed.). McGraw-Hill.
- Meucci, A. (2009). "Managing Diversification." Risk, 22(5), 74-79.
- Choueifaty, Y., & Coignard, Y. (2008). "Toward Maximum Diversification." Journal of Portfolio Management, 35(1), 40-51.