The Case for Custom Indexing

Episode Summary

Executive Summary: Patrick O’Shaughnessy explains why OSAM pivoted from launching an ETF to building custom indexing: ETFs are a brutal commodity battleground, while direct indexing/custom indexing can deliver tax alpha and personalized portfolio construction. The conversation covers tax harvesting mechanics, market constraints, Franklin Templeton’s acquisition of OSAM, and why customization may grow without displacing ETFs.

Main Topics: Direct indexing vs. custom indexing (Priority: 5/5): O’Shaughnessy distinguishes traditional direct indexing—owning most stocks in an index in a separate account for tax-loss harvesting—from custom indexing, which uses technology to tailor portfolios around taxes, preferences, concentrated positions, and other investor-specific constraints. Why OSAM abandoned the ETF path (Priority: 5/5): He says ETFs are attractive to investors but punishing for businesses because of unlimited competition, price pressure, and weak pricing power. That led OSAM to pivot toward a differentiated software-driven offering instead. Tax alpha and harvesting mechanics (Priority: 5/5): The discussion breaks down how separate accounts can generate tax losses while maintaining benchmark-like exposure, and why the strategy works best in low-turnover portfolios with manageable tracking error. Customization as a one-to-one portfolio model (Priority: 4/5): Custom indexing is presented as a 'one size fits one' solution that blends passive and active exposures, accommodates tax situations, and handles ESG or concentrated-stock constraints more flexibly than standard funds. Franklin Templeton acquisition and distribution scale (Priority: 4/5): OSAM’s sale/partnership with Franklin Templeton is framed as a strategic fit: Franklin brought distribution, scale, and commitment to the category, while OSAM brought technology and product innovation. Industry competition and platform shifts (Priority: 4/5): O’Shaughnessy argues major asset managers cannot ignore platform shifts like direct/custom indexing, comparing the dynamic to how firms had to adapt to ETFs or mobile transitions in tech. Philosophy, learning, and media strategy (Priority: 2/5): The conversation ends with broader reflections on learning from outside one’s industry, the influence of philosophy on decision-making, and why O’Shaughnessy prefers long-form audio over video as a communication medium.

Key Arguments: Direct indexing is fundamentally about tax-loss harvesting in a separate account while keeping exposure close to the chosen benchmark. Custom indexing goes beyond tax harvesting by allowing portfolio-level personalization for taxes, active/passive blends, ESG preferences, and concentrated positions. ETFs are efficient for many investors, but as a business model they are a near-perfect commodity market with intense competition and little pricing power. Loss harvesting benefits are real but often overstated; O’Shaughnessy says industry marketing used cherry-picked periods and aggressive assumptions. The platform’s value proposition is strongest for investors with taxable accounts and more complex financial situations, not necessarily for everyone. Customization does not need to kill ETFs; instead, it can coexist as a higher-touch, technology-enabled option for specific use cases. Franklin Templeton was attractive because it had deep distribution and a long-term orientation, complementing OSAM’s technology and product development. Successful products often show strong early demand quickly; O’Shaughnessy says Canvas got unusually high conversion because it addressed an unmet need. Learning from adjacent fields and “learning in public” helped shape Canvas’s development and OSAM’s broader strategy.

Data Points: AUM on platform: north of $3 billion - OSAM says its custom indexing platform had grown to more than $3 billion managed across nearly 2,000 open accounts. Number of accounts: nearly 2,000 - Describes current scale of the Canvas/custom indexing platform. Annualized tax benefit: 70–80 basis points - O’Shaughnessy’s estimate of average annual pickup over a 10-year holding period versus an ETF equivalent. Upper-end tax benefit in studies: 200+ basis points - He criticizes optimistic published studies that used best-case periods and assumptions. Lower bound of tax benefit: zero - He notes some 10-year periods produce effectively no net benefit from loss harvesting. Tracking error budget example: 1 vs. 5 - Explains that a standard tight budget might be 1, while a large budget of 5 can create more tax losses but greater benchmark deviation. Portfolio mix on platform: low 60s to mid 60s percent passive - He says roughly two-thirds of assets on Canvas are passive, with the rest active factor exposures. Minimum account size: $250,000 - Current minimum to use the platform, though he expects it to fall over time. Timeline of acquisition: end of last year / about 8-9 months in - He says the Franklin Templeton partnership closed at the end of the previous year and was still early-stage. Tweet follower increase: ~500 followers - Eric Beltrinas jokes that appearing on O’Shaughnessy’s podcast produced an immediate social-media bump.

Pivotal Quotes: "one size fits one" — Patrick O'Shaughnessy: He uses this phrase to define custom indexing as highly tailored, technology-enabled portfolio construction. "we thought this is just a terrible idea" — Patrick O'Shaughnessy: His blunt reaction to the economics of entering the ETF business amid intense competition and commoditization. "they bring a bazooka to the knife fight" — Patrick O'Shaughnessy: He describes Franklin Templeton’s distribution scale relative to OSAM’s smaller sales organization.

Implications: Direct/custom indexing is likely to grow as a specialized, tax-aware portfolio tool, especially for wealthy and complex accounts. ETFs remain dominant for simplicity and low cost, but incumbents must participate in this platform shift or risk falling behind.