At The Money Fan Favorite: Looking Past Market Cap Weighted Indexes

Episode Summary

Executive Summary: The discussion critiques market-cap weighted indexing for concentrating risk in expensive mega-cap stocks and highlights hidden trading costs from index additions/deletions. Rob Arnott argues that fundamental indexing—weighting companies by economic measures like sales, profits, dividends, and net worth—reduces concentration risk, creates a value tilt, and has historically outperformed cap-weighted value benchmarks with lower tracking error than many expect.

Main Topics: Critique of cap-weighted indexing: Arnott argues that cap weighting is a form of momentum-based buying: investors end up owning more of what has recently risen, which can inflate concentration risk and valuation exposure. Index reconstitution costs and market impact: Additions and deletions create trading friction, front-running incentives, and a measurable performance drag for index funds forced to transact at close prices. Flip-flops in index membership: Stocks that are added, dropped, and later re-added often exhibit extreme price swings, which Arnott says are costly for passive investors and evidence that index mechanics are not truly passive. Fundamental indexing / RAFI methodology: Fundamental indexes weight stocks by economic footprint—sales, profits, net worth, and shareholder distributions—then average those measures to create a broader, more valuation-sensitive index. Comparison with equal weighting: Equal weighting is presented as a legitimate alternative, but Arnott notes it tends to create small-cap and value tilts and can be more volatile than fundamental indexing. Live performance and value cycle: Arnott emphasizes that RAFI has been investable for about 20 years and argues that its long-run outperformance versus cap-weighted value comes with less volatility than many investors assume.

Key Arguments: Cap-weighted indexes are not fully passive; they mechanically buy winners and sell losers based on market capitalization changes. Index reconstitution creates avoidable trading costs because index funds must transact at forced prices, often after the market has already moved. The aggregate holdings of S&P index funds are so large that index additions/deletions can materially affect prices and invite front-running. Stocks that are added after large outperformance and later deleted after large losses produce a costly buy-high/sell-low pattern. Fundamental indexing better reflects the economy by weighting firms according to real business measures rather than market enthusiasm. RAFI naturally creates a value tilt because it downweights frothy growth stocks and upweights cheaper, economically larger businesses. Compared with cap-weighted value indexes, RAFI has historically added about 2% to 2.5% per year compounded, with relatively tight tracking error. Equal weighting is a valid alternative but tends to overemphasize small, cheap, and more volatile stocks, making it less precise than fundamental indexing for many investors.

Data Points: Annual index turnover: 3% to 5% - Arnott says index funds still incur trading costs despite being labeled passive. Potential annual benefit from better execution timing: 15 basis points per annum - Estimated gain if trades occurred at announcement price rather than effective date price. SP index fund market share in constituents: ~25% of total market cap - Aggregated S&P index-fund ownership of stocks in the index, increasing price impact at reconstitution. Typical relative trading cost per stock per trade: 300 to 500 basis points - Arnott’s estimate of the cost embedded in reconstitution trading when turnover is 3% to 5% annually. Stocks added that later get dropped: ~28% within a decade - Frequency of add-then-delete flip-flops in index membership. Deleted stocks that rejoin within a decade: Almost 50% - Frequency of delete-then-readd flip-flops. Average outperformance before addition: 75 percentage points - Stocks are typically added after massive relative gains. Average loss before deletion: About 7,000 basis points - Arnott describes stocks being removed after sharp relative declines. Relative underperformance before deletion in flip-flop stocks: 3,500 basis points - Stocks later re-added tend to be badly weak before being removed. Relative outperformance before re-addition: 180 percentage points - Deleted stocks often rebound strongly before re-entering the index. Fundamental weight for NVIDIA: About 1% to 1.5% - Illustrative economic footprint versus much larger cap weight. RAFI assets: Over $100 billion - Assets across Schwab, Invesco, and PIMCO products tied to RAFI. RAFI outperformance vs cap-weighted value: 2% to 2.5% per year compounded - Arnott’s stated long-run live performance advantage. Value underperformance peak-to-trough: 3,800 basis points - Russell value underperformed Russell 1000 by this amount at one point by end of 2025 in the discussion. Variable tracking error / variability: About 2.5% - Arnott says RAFI’s value add has relatively tight dispersion around the average. Years invested / live history: 20 years - Fundamental index strategies cited as live since mid/late 2005.

Pivotal Quotes: "What I've just described is the active side of indexing." — Rob Arnott: He argues that indexing is not entirely passive because reconstitution forces active buy/sell behavior. "It's a herd of elephants trying to go through a single revolving door." — Rob Arnott: He uses this metaphor to describe crowded index rebalancing and the friction it creates. "If you want to weight the economy instead of weighting the market, fundamental indexing is the way to do it." — Rob Arnott: Core explanation of why economic weighting is presented as a superior alternative to cap weighting.

Implications: For investors, the key takeaway is that cap-weighted indexing can hide concentration, valuation, and rebalancing costs. Fundamental indexing offers a more economically grounded alternative that may better manage mega-cap risk and deliver long-run value with less dependence on market froth.