Princeton Asset

The Arithmetic of Passive Investing

—

by

WP ADMIN
in

For over three decades, William F. Sharpe’s The Arithmetic of Active Management has been treated as near-scripture for passive investing. Its logic is elegant, its math is simple, and its conclusion seems inescapable: before costs, the average active manager must equal the market; after costs, they must underperform. QED.

But what if the arithmetic is right – and yet incomplete?

In this post, we compare Sharpe’s foundational 1991 paper with Lasse Heje Pedersen’s 2018 refinement, Sharpening the Arithmetic of Active Management, and connect both to our own recent findings on global equity portfolio construction. The result is a more nuanced picture of what “passive” really means, why the market portfolio is not a static object, and how the mechanism by which the portfolio structure evolves may matter as much as the assets themselves.

Sharpe’s Closed System

Sharpe’s argument rests on a single, powerful identity. If we define:

  • Passive investors as those who hold every security in proportion to its market capitalization
  • Active investors as everyone else

Then the market return is a weighted average of passive and active returns. Since passive investors are the market by construction, active investors, in aggregate, must also equal the market return before costs. After costs, they lose.

“These assertions will hold for any time period. Moreover, they depend only on the laws of addition, subtraction, multiplication and division. Nothing else is required.” – William F. Sharpe, The Arithmetic of Active Management (1991)

The brilliance of this argument is its generality. It applies to any market, any time period, any asset class. It does not depend on market efficiency, investor rationality, or economic equilibrium. It is pure accounting.

But as Pedersen would later show, it depends on something else: a static market.

Pedersen’s Open System

In his 2018 Financial Analysts Journal paper, Pedersen challenges Sharpe’s equality with a deceptively simple observation: the market portfolio changes.

Companies issue new shares. They repurchase old ones. They go public, go private, merge, spin off, and go bankrupt. Indexes add and delete constituents. Even a “passive” investor who wants to maintain market-cap weights must trade continuously to keep up.

Where Sharpe’s Equality Breaks

Pedersen identifies four mechanisms by which Sharpe’s equality fails in the real world:

1. IPOs and Seasoned Equity Offerings (SEOs) When new shares are issued, passive investors must buy them to maintain market-cap weights.

2. Share Repurchases The mirror image of issuance. Passive investors must sell into repurchases.

3. Index Reconstitution When a stock is added to an index, index funds must buy it – often pushing the price up before the effective date. When a stock is deleted, index funds must sell – often pushing the price down. Active investors can anticipate these flows. Petajisto (2018) estimated the lower bound cost to Index investors as 21-28bp annually for the S&P 500 and 38-77bp annually for the Russell 2000

4. Rebalancing and Market Timing Passive investors must periodically rebalance between risky and risk-free assets. At times when many passive investors are selling (e.g., during a panic), prices may drop below fundamental value. Active investors taking the other side can profit.

Pedersen’s conclusion is measured but profound: active management can be worth positive fees in aggregate, not because stock-picking is easy, but because passive investing is harder than it looks. The market is not a still picture it is a movie.

“Sharpe’s argument thus abstracts from a key aspect of addition and subtraction-namely, the addition of new companies and shares and the subtraction of disappearing ones.” – Lasse Heje Pedersen, Sharpening the Arithmetic of Active Management (2018)

The Passive Dimension

As Pedersen showed a market-cap index fund like VT is not a buy-and-hold portfolio. It is a continuous float adjusted engine that must absorb flows to maintain market weights. This breaks Sharpe’s closed-system assumption.

What Princeton Asset Adds

In our three-part series (Part 1, 2 and 3.) on global equity portfolio construction we showed that even within passive indexing, structural choices matter. Float-adjusted market-cap weighting is one specific rule but not the only passive implementation.

All portfolios hold essentially the same underlying securities. All are “passive”. Costs are nearly identical. But the returns differ. These are not active bets on individual securities. They are passive implementation choices with first-order consequences. The important point is not the difference in returns. The important point is the structural source of returns.

  • The weight-update engine acts as a form of factor exposure. VT embeds a continuous issuance-absorption strategy. Sliced portfolios embed momentum (Buy & Hold) or value/contrarian (rebalanced) strategies.

Conclusion:

Sharpe gave us the arithmetic of portfolio returns. He proved that within a fixed universe of securities, costs dominate active stock-picking. This remains the bedrock insight for investors evaluating high-fee managers.

Pedersen gave us the dynamics of market evolution. He showed that the market is not a static photograph but a moving picture – continuously reshaped by issuance, repurchases, and reconstitution. Even “passive” investors must trade, and that trading creates both costs and opportunities.

Princeton Asset adds a passive portfolio engineering dimension where the choice of passive implementation itself creates different exposure profiles with different historical returns and different future risks.

References

Disclaimer: This blog post is for informational purposes only and should not be considered financial advice. Past performance is not a guarantee of future results. Consult with a qualified financial advisor for personalized guidance.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *