PodcastThe Meb Faber ShowAndrew Ang

36% of Your Portfolio Just Disappeared In the Last 30 Years (Andrew Ang Explains How)

Andrew Ang examines how taxes affect long-term compounding, arguing that taxes reduced a taxable U.S. equity investor's portfolio by 30–36% over the prior 30 years, while after-tax comparisons also shape the appeal of private credit and income products. He discusses AI's potential to raise capabilities and reshape investment work, while emphasizing that agent-based strategies still need live performance evidence and redesigned workflows. For portfolios, he favors diversified factor exposure, multiple models, and account-specific asset placement, while noting that higher inflation and fiscal pressures leave tax and fixed-income strategies as open questions.

Preview · 17 more points in the app

Read the Full Summary

Key Takeaways

Andrew Ang argues AI could raise average capabilities and expand the right-hand tail of performance, but he cautions that replacing tasks without redesigning workflows may not deliver major productivity gains. He expects senior investment leaders and entering analysts to remain while some middle roles adapt, consolidate, or shift to other work.

Ang says AI agents could specialize across asset classes and work in parallel to produce expected returns with plain-English explanations, but whether they outperform human-based approaches remains unproven until live performance is assessed. He also calls for guardrails: agent changes and their rationales should be recorded for review during portfolio rebalancing.

Ang estimates taxes reduced a taxable U.S. equity investor’s portfolio by 30–36% over the past 30 years, or roughly 160–170 basis points annually, even though that period had lower federal tax drag than the 1926–2025 average of about 350 basis points. He says dividends drove much of the historical burden but are now a smaller share of the market and receive preferential treatment, while dividends and interest taxed as received interrupt compounding.

Ang argues investors should compare income-oriented products on after-tax returns: he says private credit yielding 10% is roughly equivalent to 7% in equity terms, and would need to yield about 14–15% to match a 10% equity return. He also says investors may be better off selling appreciated stock themselves than receiving taxable yield from income-enhanced products.

TECHNOLOGY

AI And Capability — Andrew Ang argues that AI will raise the average person's capabilities, with even larger gains possible at the right-hand tail of performance. He frames the opportunity around how much better people can become with AI agents, not just whether AI can outperform an average person.

“I think the relevant thing here is the average is going to increase. And I think the right hand tail will increase out even more…”—

Allocation Agents — Ang says agents can specialize in different asset classes, run in parallel, and quickly develop expected returns with plain-English explanations. But he says it remains unclear whether agentic strategies outperform human-based approaches; live performance is needed to assess them.

“you get an agent to specialize in each of those and they can run in parallel and formulate these expected returns very quickly and it comes with narrative with plain…”—

ECONOMY

New York Tax Rate — The host cites a 55.576% highest marginal tax rate for someone in New York City. This is an example of a marginal income-tax rate, not a measure of portfolio returns.

“The highest marginal rate for someone in New York City is it's 55.576%.”—

Equity Tax Drag — Ang estimates that taxes reduced a taxable U.S. equity investor's portfolio by about one-third over the prior 30 years—roughly 160–170 basis points per year, or a 30–36% cumulative reduction. Dividends and interest are taxed as received, interrupting compounding, while equity capital gains can compound tax-deferred.

“the bottom line of this thing is you know what's the what's Uncle Sam's cut and that number is about 1/3 right over the last 30 years right it's the…”—

Historical Tax Drag — Ang says the last 30 years were at the low end of federal tax drag across 1926–2025. He cites an average of about 350 basis points over the century, versus roughly 160–170 basis points annually in the recent period, attributing the lower modern drag partly to reduced tax rates and brackets and preferential treatment for dividends.

“if you look at the entire century, you know, 100 years from 1926 to 2025, the last 30 years has actually been on the low end, right?…”—

MARKETS

Private Credit Hurdle — Ang says private credit would need to yield about 14–15% to match a 10% equity return, accounting for the different tax treatment of the returns.

“you know 10% in private equity it's roughly about 7% in equity terms or you can do it the other way you know what what's the return in private credit…”—

Agent Guardrails — Ang says portfolio rebalancing gives investors a chance to evaluate investment decisions, including their reasoning and performance. Changes to agents should have guardrails, with modifications and rationales recorded for review by an investment committee, manager, or analyst.

“But since portfolios are rebalanced, right, and investments you buy and sell them, right? That provides an opportunity for you to evaluate the decision- making, right?”—

Factor Diversification — Ang advocates diversification at multiple levels: using multiple signals to measure each factor, holding a balanced mix of factors rather than relying on one, and using low tracking error for a core portfolio allocation. He cites value's severe 2018–2020 drawdown as a reason to hold other factors as well.

“we've applied that concept to factor investing in three levels. So the first layer of diversification is using multiple signals to measure a given factor…”—

Starts

The host introduces Andrew Ang and asks whether AI agents can match or surpass investment professionals; Ang says adoption is only beginning and AI may raise both average capabilities and the upper end of performance.

The Self-Driving Investment Portfolio with AI

Ang describes self-driving portfolios built from specialized agents that estimate returns, construct portfolios, critique proposals, and learn, while humans oversee governance, exceptions, and redesigned workflows.

Evaluating AI-Driven Investment Outputs

Ang says tightly constrained investment policy statements keep agent-generated portfolios close to conventional allocations, but live performance is needed to establish whether they outperform; he also discusses model risk, guardrails, and workforce adaptation.

Multi-Level Diversification in Factor Investing

Ang advocates diversification across multiple signals within factors, across factors, and across timing models, explaining how his approach combines momentum, quality, enhanced value, and cyclical value.

Factor Weighting, Timing, and Model Diversification

Ang explains the rationale for separate value measures and a diversified momentum factor, then describes factor weights, cautious timing, and model diversification to address changing factor returns.

The Long-Term Drag of Taxes on Equity Returns

Ang estimates taxes reduced a taxable U.S. equity investor’s portfolio by roughly 30–36% over the prior 30 years, notes that the century-long tax drag was higher, and explains dividends’ role in the cumulative cost.

Rising Taxes: Strategies and Asset Location

Ang argues that taxable investors should consider after-tax returns, use tax-advantaged accounts, and coordinate assets across taxable, retirement, trust, and business accounts over long horizons.

Tax Alpha vs. Investment Alpha & Market Impact

Ang and the host agree that tax alpha can be easier to capture than investment alpha; Ang discusses how tax uncertainty may affect valuations and reviews the wide range of estimates for optimal tax rates.

Tax-Aware Bond Strategies

Ang identifies persistent inflation as a challenge for traditional bond strategies and raises tax-loss harvesting, bond structuring, and more tax-efficient income products as possible areas to explore.

Related Episodes

Important Disclaimers

Not Investment Advice: For education only — not financial, tax or legal advice.

AI-Generated: Made by AI from source transcripts and may contain errors.

No Guarantees: Past performance does not predict future results; investing carries risk.

Our Commentary · Request Removal