At The Money: Looking Beyond Market Cap Weighted Indexes

Episode Summary

Executive Summary: Rob Arnott argues that market-cap-weighted indexes are not truly passive because they systematically buy high and sell low, create hidden trading costs, and amplify concentration risk. He makes the case for fundamental/economy-weighted indexing, which weights companies by real economic size and has historically delivered a value tilt, lower concentration, and better long-term risk-adjusted outcomes than cap-weighted value benchmarks.

Main Topics: Critique of market-cap weighting as 'active' behavior (Priority: 5/5): Arnott argues that cap-weighted indexing is not neutral passive investing; it mechanically overbuys winners and underweights or sells losers, embedding an active decision to chase price momentum. Hidden trading costs and index reconstitution effects (Priority: 5/5): He explains that index additions/deletions create predictable price pressure and front-running, adding implicit costs for index funds despite low headline turnover. The 'flip-flop' problem in index membership (Priority: 4/5): Stocks that are added and later removed—or deleted and re-added—often do so after extreme relative moves, creating severe performance drag for cap-weighted strategies. Fundamental/economy-weighted indexing methodology (Priority: 5/5): Arnott outlines weighting companies by sales, profits, net worth, and shareholder distributions, then averaging those measures to approximate economic footprint rather than market price. Value tilt and comparison with equal weight (Priority: 4/5): He notes that fundamental indexing naturally tilts toward value and that equal-weighting is a legitimate alternative with similar long-run returns but greater volatility and a small-cap bias. Live performance and rebalancing alpha (Priority: 5/5): Arnott says RAFI has been live for 20 years and has outperformed cap-weighted value indexes by about 2% to 2.5% annually, aided by systematic trimming of expensive stocks and adding beaten-down ones.

Key Arguments: Cap-weighted indexes are not fully passive because the 5% or so of annual turnover is an active trading layer that buys stocks after they have already risen and sells them after they have fallen. Index rebalancing creates legal, predictable front-running opportunities that impose an estimated drag; Arnott says avoiding trading at effective-date prices would add about 15 basis points per year. The S&P 500 and similar benchmarks are heavily concentrated, and index funds now represent a large share of the market capitalization of constituent stocks, increasing mechanical demand at reconstitution. Stocks added to indexes often have already outperformed dramatically, while deletions are frequently sold at depressed prices, producing a buy-high/sell-low structure. Fundamental indexing better mirrors the economy by using real business measures such as sales, profits, net worth, and distributions instead of market price. Averaging multiple fundamental measures reduces reliance on any single accounting input and yields a weight that can differ substantially from market cap, especially for mega-cap growth stocks like NVIDIA. Because fundamental indexing downweights expensive growth stocks and upweights cheap stocks, it has a pronounced value tilt and should be judged relative to value benchmarks. Equal weighting is a valid alternative but introduces a stronger small-cap tilt and more volatility than fundamental indexing, even if long-run returns are broadly similar. RAFI’s live track record suggests that rebalancing discipline creates alpha by trimming overheated names and adding undervalued ones, with performance exceeding cap-weighted value by roughly 2% per year. Arnott contends that the recent pain of value-oriented strategies does not invalidate the approach; rather, it reflects prolonged valuation compression that eventually can reverse.

Data Points: Number of bond funds at Vanguard: Over 80 - Promotional read describing Vanguard's active fixed-income lineup. Vanguard fixed-income team size: 200 people - Promotional read describing the global squad managing bond funds. Index fund turnover: About 3% to 5% annually - Arnott describes the relatively small but impactful active trading layer in cap-weighted indexing. Estimated trading-cost drag from index changes: 15 basis points per year - Arnott says this would be added if trades could occur at announcement prices instead of effective-date prices. S&P index fund ownership share: Roughly 25% of total market cap of each stock in the index - Arnott cites aggregated index-fund ownership in S&P constituents. Index addition outperformance before inclusion: About 75 percentage points - Stocks are often added after large relative gains. Relative valuation at addition: About 2x the market multiple - Stocks added to indexes are often expensive relative to the market. Average loss at deletion: About 7,000 basis points - Stocks may be removed after a sharp decline from their pre-addition levels. Addition flip-flop rate within a decade: About 28% - A share of additions are later dropped again. Deletion re-entry rate within a decade: Almost half - Many deleted stocks rejoin the S&P within ten years. Deletion candidate underperformance before removal: About 3,500 basis points - Deleted and re-added stocks often underperform substantially before being dropped. Re-addition outperformance: About 180 percentage points - Deletion flip-flops can then rally dramatically before re-entry. NVIDIA economy weight estimate: Around 1% to 1.5% - Arnott estimates its weight based on fundamental measures rather than market cap. RAFI assets: Over $100 billion - Assets across Schwab, Invesco, and PIMCO products tied to the fundamental index. RAFI launch timeline: About 20 years ago - Fundamental indexing was introduced roughly two decades prior to the interview. RAFI live strategy history at PIMCO: Mid-2005 - Arnott cites live implementation timing. RAFI live strategy history at Invesco: Late 2005 - Arnott cites live implementation timing. Russell value underperformance vs Russell 1000 peak-to-trough: 3,800 basis points - He highlights the severe drawdown/value headwind through the cycle. RAFI outperformance vs cap-weighted value: A little over 2% per year compounded - Arnott presents live performance versus cap-weighted value benchmarks. Tracking error versus cap-weighted value: About 2.5% - He notes variability around RAFI's relative performance. Years RAFI beats cap-weighted value: About three out of every four years - Live track record claim.

Pivotal Quotes: "What I've just described is the active side of indexing." — Rob Arnott: He reframes cap-weighted indexing as a rules-based strategy that still makes active bets through market-price weighting. "If you don't know you have a problem, how are you going to fix it?" — Rob Arnott: He explains why hidden index-reconstitution costs and flip-flops deserve more scrutiny. "Thanks for the nice high price. I'm going to trim it. If it tanks and the fundamentals don't falter, you're going to say, thanks for the bargain. I'm going to top it up." — Rob Arnott: He describes the rebalancing discipline of fundamental indexing as a systematic buy-low/sell-high process.

Implications: Listeners should view cap-weighted indexing as an active, price-chasing strategy with hidden costs and concentration risk. For investors seeking diversification and a value tilt, fundamental indexing may offer a more economically grounded alternative.