High Active Share Portfolio Construction
Diverge deliberately from the benchmark and manage asymmetric tracking error
- Difficulty
- Expert
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 96%
The framework begins with a mathematical constraint: a portfolio that closely replicates its benchmark but charges more must underperform after costs. Genuine active management therefore requires high active share—meaningful differences in holdings and weights. As divergence rises, the probability of substantial outperformance can increase, but so can downside tracking error and business risk from impatient clients. Portfolio construction must consequently preserve differentiated upside while controlling catastrophic downside through security analysis, position sizing, diversification of risks, and opportunistic reorientation during dislocations. The manager must also select clients who understand that temporary relative underperformance is unavoidable. Low costs help, but they cannot substitute for a portfolio capable of producing returns different from the benchmark.
Origin
Bill Miller described this construction logic while explaining Miller Value Partners' unusually high active share on Masters in Business.
Core principles
- 01A benchmark-like portfolio cannot beat its benchmark after higher costs
- 02Greater benchmark divergence increases both opportunity and risk
- 03Tracking error is necessary for meaningful active management
- 04Downside deviation must be mitigated without eliminating upside deviation
- 05Clients must understand that underperformance comes with the territory
How to run it
- 1
Define the Benchmark Honestly
Choose the benchmark that best represents the portfolio's opportunity set and measure current overlap.
Pro tip Use holdings-based active share rather than relying only on return correlations.
Watch out An inappropriate benchmark can make active share appear more meaningful than it is.
- 2
Remove Closet-Index Positions
Challenge holdings whose primary justification is reducing benchmark deviation. Retain securities only when the investment thesis supports them.
Pro tip Ask whether each position would still be owned if it were absent from the index.
Watch out Removing benchmark overlap indiscriminately can create unintended factor bets.
- 3
Build Around Differentiated Conviction
Allocate meaningful weight to the best researched opportunities while diversifying sources of permanent-loss risk.
Pro tip Concentrate ideas, not correlated failure modes.
Watch out Conviction does not excuse weak downside analysis.
- 4
Model Tracking-Error Asymmetry
Estimate how the portfolio could diverge in both rising and falling markets. Seek structures where upside deviation exceeds tolerable downside deviation over time.
Pro tip Stress test security-specific, factor, liquidity, and macroeconomic shocks separately.
Watch out Historical tracking error may understate future regime changes.
- 5
Use Volatility Opportunistically
During large declines, reorient toward assets with stronger recovery potential; after large gains, trim them for more resilient holdings.
Pro tip Set valuation ranges in advance so volatility triggers analysis rather than panic.
Watch out Frequent reactive trading can turn reorientation into market chasing.
- 6
Align the Client Base
Explain expected drawdowns and benchmark-relative droughts before accepting capital. Prefer clients whose horizons match the strategy.
Pro tip Use historical underperformance episodes to make the trade-off concrete.
Watch out Misaligned clients can force selling at the worst time.
In the wild
Miller described being substantially behind the market in August 2019 before a change in the Federal Reserve's reaction function reduced recession fears. Because the portfolio differed sharply from the benchmark, it recovered the deficit rapidly and finished ahead rather than remaining locked into index-like performance.
→ High active share enabled a fast benchmark-relative recovery, while also having caused the earlier shortfall.
Common mistakes
Charging Active Fees for an Index
A benchmark-like portfolio has little opportunity to overcome its higher costs.
Ignoring Downside Tracking Error
Benchmark divergence creates losses as well as gains. Position sizing and risk analysis must address both directions.
Accepting Misaligned Capital
Clients who cannot tolerate relative drawdowns may redeem before the differentiated thesis can work.
Is it for you?
Best for
It is best for skilled active managers and clients willing to tolerate meaningful benchmark-relative volatility.
Not ideal for
It is not ideal for investors who require index-like quarterly returns or cannot endure extended relative underperformance.
From the transcript
“your portfolio can't look exactly like your benchmark.”
“If your portfolio looks like your benchmark exactly, then you're gonna underform your benchmark.”
“your probabilities of increasing about performing grow as you diverge from your benchmark.”
From the episode
Bill Miller on the Classical Value Portfolio (Podcast)