$70 Billion. 18 Straight Outperforming Years | David Giroux on the Index Trap and AI Hype

Episode Summary

Executive Summary: T. Rowe Price CIO David Giroux argues that markets are highly mispriced by sector and by valuation lens, and that disciplined bottom-up work can uncover pockets of strong returns in utilities, healthcare, software, and select AI beneficiaries. He sees AI as powerful but still early, favors market drawdowns as buying opportunities, and thinks fixed income offers less opportunity than stocks today.

Main Topics: Structural market inefficiencies and GARP (Priority: 5/5): Giroux explains how stocks can be mispriced because they don't fit neatly into value or growth buckets, creating persistent gaps that patient investors can exploit. Buying equities on market drawdowns (Priority: 5/5): He reiterates that sharp market declines historically improve forward returns and justify adding risk when fear is highest. Today’s market valuation and index composition (Priority: 5/5): He argues the S&P 500 is very different from past cycles and must be judged by current company mix, not old index history. AI as an early but transformative theme (Priority: 5/5): Giroux believes AI could be as impactful as railroads or the internet, but says real enterprise adoption and ROI are still emerging and will create winners and losers. Sector opportunities: utilities and healthcare (Priority: 4/5): He is overweight utilities and healthcare because AI-driven power demand, aging demographics, and biotech acquisition opportunities create attractive expected returns. Fixed income, treasury curve, and fiscal risk (Priority: 4/5): He sees limited opportunity in credit, thinks the belly of the Treasury curve is most attractive, and worries about long-term U.S. fiscal sustainability. Portfolio management philosophy (Priority: 4/5): He stresses independent thinking, making active bets, and avoiding benchmark hugging as the path to long-term outperformance.

Key Arguments: GARP names often have no natural owners, leaving them mispriced versus both value and growth managers. Historical drawdowns are often the best time to add risk because 12-month forward return expectations improve when markets fall 15%-30%. The S&P 500 must be analyzed by current business mix, since today’s index is much more growth-oriented and less tied to old low-multiple sectors. Headline valuation measures can be misleading because changing mix and profit structure alter what a fair multiple should be. AI is real and useful, but many enterprise use cases are still too early to justify massive spending without broader adoption. NVIDIA’s moat and margins may narrow as more companies use orchestration layers, AMD, TPUs, and multiple model providers. Utilities may benefit from AI-driven power demand and data center buildouts, while healthcare offers acquisition-driven upside and demographic tailwinds. Treasuries are fairly valued overall, but the 4-7 year part of the curve offers the best risk/reward because long-end debt carries fiscal-risk skew. Successful portfolio managers must think independently, make bets, and avoid simply chasing what has recently worked.

Data Points: Market decline and forward risk: 15%-30% declines often lead to lower next-12-month loss risk - He said drawdowns historically improve forward return profiles. Expected return after declines: Rises from about 10% to 15%-20% per year - His historical framework for market selloffs. April equity buying: $4 billion - Equities purchased over three days during the April swoon. COVID downturn equity buying: $7 billion - Equities bought during the last days of the COVID selloff. GARP valuation: 18-19x earnings - Typical valuation where value managers dismiss GARP names as expensive. Double-B high yield advantage: 100 basis points higher yield - Giroux says double-Bs can offer investment-grade-like default risk with better yield. S&P 500 sector mix in 2006: 45% from financials, materials, and oil - Used to show how different the index was in the past. S&P 500 growth mix today: 53% of market cap in businesses growing organically in the high single digits - Used to argue that the index is structurally different now. 2031 market multiple estimate: 19x to 19.5x earnings - His bottom-up estimate of fair value for the index. Expected market return next five years: About 6% - His aggregate market return estimate from company-level analysis. AI utility company earnings growth: 9%-12% - Projected growth for select utility holdings benefiting from data centers and AI demand. Nysource example: 11% earnings growth; 20x earnings; 200% dividend yield - Illustrative utility name he cited as a low-risk opportunity. Treasury inflation assumption: 2%-2.5% - Used in his rough fair-value framework for yields. Treasury fair value: 4%-4.5% for the 10-year - Derived from inflation plus fed funds assumptions. Five-year vs 10-year Treasury spread: About 40-50 bps historically; around 45 bps currently - He says the skew is toward wider spreads over time. Meta organic growth: Double-digit, around 20% - He cited Meta as still growing strongly despite heavy CapEx. NVIDIA market share: 93%-94% dollar share of GPUs today - He expects this to fall as competition grows. AI share of S&P 500 earnings: 5%-6% - He noted NVIDIA remains a relatively small part of index earnings. T. Rowe Price firm alpha vs equity market: 350 bps per year over 19 years - Performance claim linked to disciplined active management. T. Rowe Price firm alpha vs fixed income market: 300 bps per year over 19 years - Performance claim linked to fixed income strategy. Outperformance versus 60/40 index: 2500 bps per year over 19 years - He cited this as evidence of active, independent portfolio construction.

Pivotal Quotes: "When the market falls 15% or the market falls 20% or 30%, the risk of loss over the next 12 months actually are lower, not higher." — David Giroux: Explaining why his team adds equity risk during selloffs. "Could this be as impactful as the railroads or internet? Absolutely. Maybe even more, maybe even more powerful. But I think it's still to be determined how powerful that is." — David Giroux: His view on AI’s potential versus its still-uncertain economic impact. "The S&P 500 of today is vastly different from the S&P 500 of 2006 and very, very different from the market of 2011." — David Giroux: Arguing that market valuation must be judged by current composition, not old benchmarks.

Implications: Listeners should expect active managers to favor cheapened risk during selloffs, prioritize bottom-up analysis over index narratives, and seek AI winners beyond mega-cap tech. Sector dispersion and fiscal risk may matter more than broad market calls.