Gerard O'Reilly: Deep Dive with Dimensional's co-CEO & CIO (EP.198)

Episode Summary

Executive Summary: Gerard O’Reilly, Dimensional’s co-CEO and CIO, explained how the firm uses academic evidence plus implementation expertise to build rules-based portfolios that tilt toward expected-return drivers like size, value, profitability, investment, and momentum without becoming rigid index replicas. He defended Dimensional’s choices on profitability, goodwill, intangibles, sector control, credit, and capacity, while emphasizing that the real edge is integrated research, trading, stewardship, and client support.

Main Topics: Dimensional’s philosophy vs. index investing (Priority: 5/5): O’Reilly contrasted Dimensional’s rules-based, flexible, research-driven approach with rigid market-cap indexing, arguing that Dimensional keeps the benefits of transparency and low cost while avoiding index drawbacks like inflexibility and inability to adapt. How Dimensional identifies and uses expected-return signals (Priority: 5/5): He described how Dimensional evaluates variables before and after looking at data, focusing on sensible economic intuition, robustness across markets, meaningful premiums, volatility, and implementability in low-turnover portfolios. Portfolio construction: size, value, profitability, investment, momentum (Priority: 5/5): The discussion covered how Dimensional combines long-term signals (size, value, profitability) with shorter-term signals (momentum, asset growth, lending market information) to shape daily trading and portfolio tilts. Debates over accounting adjustments: profitability, goodwill, intangibles (Priority: 4/5): O’Reilly defended operating profitability over cash-based profitability, rejected systematic goodwill adjustment, and argued that internally developed intangible estimates are currently too noisy to improve portfolio construction reliably. Capacity, implementation, and securities lending (Priority: 4/5): He explained why Dimensional’s large asset base remains manageable through market size, small daily trading, create/redeem ETF mechanics, and high-quality execution, with securities lending and FX netting adding incremental value. Why factor premiums persist and the ICAPM question (Priority: 4/5): O’Reilly argued that factor premiums persist because investors do not share identical preferences, labor income exposures, or tolerance for tracking error, and because discount-rate differences and uncertainty remain enduring features of markets. Dimensional’s organizational edge and research process (Priority: 4/5): He emphasized that Dimensional’s difficulty to replicate lies not just in public research, but in decades of institutionalized judgment, systems, client education, and cross-functional coordination that support implementation.

Key Arguments: Dimensional is not a rigid index shop; it is a rules-based, evidence-driven manager that intentionally keeps flexibility to improve implementation and manage risk. Expected-return variables must make economic sense before being tested, be robust across datasets and geographies, and be implementable at reasonable turnover and concentration. Operating profitability is preferred over cash-based profitability because it is more stable, predicts future profitability better, and cash-based benefits disappear once other portfolio constraints are applied. Goodwill should generally remain in book value because it is an acquired asset with economic value; excluding it can make ratios noisier and less informative. Internally developed intangible estimates are conceptually interesting but too noisy in practice; adding them now would likely worsen rather than improve book-value signals. Dimensional’s integration of factors across the portfolio is more effective than blending separate market and factor funds because it allows better control of exposures, turnover, and diversification. Small-cap exposure remains important because it improves diversification and factor implementation, even if the size premium is not always isolated cleanly from other effects. Security lending, FX netting, and stewardship are meaningful second-order sources of value add that improve net investor outcomes beyond portfolio selection. The persistence of factor premiums does not require everyone to accept the same risks; differences in labor income, preferences, and willingness to tolerate tracking error keep premiums alive. Replicating Dimensional would require recreating not just the public research, but the accumulated implementation know-how, systems, and client support ecosystem built over decades.

Data Points: Dimensional assets under management: around C$850 billion - Described as the firm’s approximate scale during the introduction. Dimensional founding year: 1981 - Gerard noted this as when David Booth started the firm. Dimensional in Canada since: 2003 - Mentioned when discussing the firm’s Canadian presence. Podcast episode: 198 - This was the episode number of the Rational Reminder conversation. Interview length: about 2 hours - Hosts noted the conversation ran roughly two hours after editing. Academic data sources: 3 main sources - Market prices, income statement data, and balance sheet data. Core factor variables used: 5 out of 6 - O’Reilly said Dimensional currently uses five of the six main academic variables and is evaluating profitability growth. Value premium: 3–4 percentage points historically - He cited the historical value minus growth spread over long horizons in the US and elsewhere. Value premium t-stat: above 2 in 4 of 5 samples - Used to argue the value premium is robust across multiple independent datasets. Small-cap operating accrual effect: about 6% annually pre-1990; about 2% post-1990 - Discussed when explaining the historical decline in the accrual anomaly in small caps. Turnover for value portfolio: about 20% per year - Illustrated long-term holding periods implied by value-oriented portfolios. Momentum holding period: about 3–4 months - Used to explain momentum as a short-term signal. Daily turnover implementation: 5–10 basis points per day - Described Dimensional’s practice of small, repeated trading to shape exposures. Sector overweight limit: no more than 10% above market weight - Explained how Dimensional controls sector concentration relative to the market. Global equity market size: about US$70–80 trillion - Used to explain why the market is large enough to absorb Dimensional’s scale. ETF flows: top 10 ETF manager in about 18 months - Described rapid growth after launching active transparent ETFs in the US. SMA minimum: reduced from US$20 million to US$500,000 - Highlighted as a major expansion of access through technology. Research team size: about 100 people - Used to emphasize the depth of Dimensional’s internal research capability. ESG data providers: 8 providers - Example of the breadth of data sources integrated into Dimensional’s systems. M&A dataset: 700 mergers and acquisitions / about US$2 trillion - Used to test estimates of internally developed intangible assets. Credit bond monitoring: 15,000 bonds updated every 15 minutes - Described Dimensional’s real-time bond credit monitoring system. Client community events: about 500 events / 2,500 people - Referenced to show the scale of ongoing client education and support in 2021. Webinars: almost 100 webinars / almost 20,000 attendees - Shown as part of Dimensional’s client support infrastructure in 2021.

Pivotal Quotes: "We think that we work with mainly financial professionals, so intermediaries. We don't work with the end investor. We work with intermediaries." — Gerard O’Reilly: Explaining why Dimensional’s rules-based approach is designed to be transparent and monitorable by advisors and institutions. "If you can tolerate the tracking error, and if you work with a financial professional, then it's not a free lunch." — Gerard O’Reilly: Clarifying his view on factor tilts, risk, and why premiums may be harvestable for suitable investors. "The data didn't tell you anything." — Gerard O’Reilly: Referring to Dimensional’s internal review of goodwill adjustments and why they chose not to systematically remove goodwill from book value.

Implications: Listeners should view factor investing as an implementation problem, not just an academic one: the best portfolios depend on robust signals, prudent constraints, and disciplined execution. For the industry, Dimensional’s edge is increasingly about integrated systems and client support, not just public research.