Expected Returns Ch. 1: Returns Are Stacked, Not Given

Investment Background

Our library already contains a book that decomposes returns, so we must be clear about what this one decomposes and along which axis.

Stocks for the Long Run Chapter 3 splits long-run equity return into three pieces: dividends, earnings growth, valuation change. That is a decomposition of one asset class, running backward through two centuries.

This book cuts along a different axis.

It asks: today, for each asset I hold, what risk am I being paid to bear?

One is a vertical decomposition through history; the other is a horizontal decomposition of the present. They are complementary, and this book supplies the second.

The Wall Street Translation

The Core Reframing

Most people think about returns like this: "equities return about 7% over the long run."

Ilmanen asks you to think differently:

Your return is not "what equities gave you." It is a stack of risk premia, each one compensation the market pays you for bearing a specific discomfort.

Take it apart:

Suppose you hold a corporate bond. Your expected return is not one number but a stack:

Layer	Why you are compensated
Risk-free rate	Forgoing consumption now and lending the money
Term premium	Lending for longer, bearing interest rate movement
Credit premium	The borrower may default
Liquidity premium	You may be unable to sell at fair value when you need to

Four layers. Each carries its own price, and that price is observable today.

The power of this view: it converts an unknowable question (what will returns be) into a partly observable one (what am I being paid at each layer right now).

Why This Beats Historical Averages

This is the chapter's practical core.

Planning with a historical average has a fundamental problem: it assumes your starting point today matches the average starting point of history.

Usually it does not.

A concrete contrast:

Suppose the historical average long-run real equity return is 6.5% (the Chapter 1 figure in Stocks for the Long Run).

• Start when dividend yields are high and valuations low → you will very likely receive more than 6.5%.
• Start when yields are low and valuations high → very likely less.

The historical average does not know which starting point you occupy. The current yield does.

Which is Ilmanen's core methodological claim:

The best estimate of expected return comes from currently observable yields, not from past realized returns.

Bonds make this easiest to see: a Treasury yielding 4% to maturity will return very nearly 4% nominally — requiring no forecast, only reading a number.

Equities are noisier, but the logic holds: dividend yield plus expected real earnings growth is a better starting point than "the historical average."

Division of Labor With the Rest of the Library

Boundaries are needed, because this sits close to three existing books.

Book	Owns
Stocks for the Long Run ch03	The historical decomposition of equity return — dividends, earnings growth, valuation change, running back two centuries
Stocks for the Long Run ch05	CAPE as a planning input — starting valuation shapes the following decade
Mastering the Market Cycle (Marks)	Cycle position — which end of the pendulum, adjust posture accordingly
What Works on Wall Street	Stock-level factors — price-to-sales, momentum, cross-sectional selection
This book	Stacked risk premia across asset classes, priced off today's observable yields

The distinction from the CAPE chapter most needs stating, because it is nearest.

Stocks for the Long Run ch05 says: starting valuation affects future equity returns.

This book says: that is one special case of a more general phenomenon. Every asset class has an observable starting yield, and equity CAPE is merely the noisiest member of that family.

A bond's yield to maturity is the same thing, and far cleaner. A credit spread is the same thing.

One is a special case; the other is the framework.

The distinction from Marks is one of purpose: Marks uses cycle position to adjust positions. This book uses current yields to adjust the assumption in your spreadsheet. The first is tactical, the second is planning.

An Immediate Qualification

This book carries a genuine danger, and Chapter 1 must handle it.

Ilmanen is an institutional investor, and much of the book concerns harvesting these premia through leverage, shorting, and derivatives.

That part is not for you, and we explicitly do not recommend attempting it.

This book's value to an individual retirement investor lies entirely in its diagnostic half, not its implementation half:

Ilmanen's use	Applies to you
Understanding what your return is made of	✅ Fully
Setting planning assumptions from current yields	✅ Fully — this is the book's main value to you
Spotting the same risk you are unknowingly holding twice	✅ Applies (Chapter 4)
Equalizing premia contributions with leverage (risk parity)	❌ Requires leverage; not applicable
Shorting low-premium assets	❌ Not applicable
Rotating tactically between premia	❌ Requires forecasting; the evidence is thin

The first three rows are this book; the last three belong to institutions. Chapter 6 handles that boundary in full.

Executable Trading Rules

1. Stop asking "what will markets do" and start asking "what am I being paid now." The chapter's most important reframing. The first is unknowable; the second is partly observable. For the bond portion, almost fully observable.

2. For each part of your portfolio, write down its source of compensation. Concretely: list your holdings and beside each write "I receive this return for bearing what?" If you cannot write it, you may be holding something whose risk source you do not understand.

3. Plan with current yields rather than historical averages. For bonds this is direct (use yield to maturity). For equities, use the current dividend yield plus a conservative real earnings growth assumption, rather than reaching for 6.5%.

4. Understand that "high expected return" always means "bearing some discomfort." There are no free premia. If something offers a high yield and you cannot see what you are bearing, that does not mean there is no risk — it means you have not found it yet. That is Chapter 3's subject.

5. Do not read this chapter as permission to time. Using current yields to set planning assumptions and using them to decide when to buy and sell are entirely different acts. The first has evidence behind it; the second does not.

Relevance to a Retirement Portfolio

This chapter supplies a tool for one very specific retirement decision: the expected return number you type into a calculator.

Every withdrawal simulator on this site asks you for that number. And most people supply a historical average.

This chapter says: adjust that number for today's starting conditions.

Concretely:

Your holding	A better expected return estimate
Bond portion	Use the current yield to maturity directly — this is nearly certain
Equity portion	Current dividend yield plus one to two points of real earnings growth
Cash	The current short rate

Then weight and sum.

That estimate usually differs from the historical average, and the direction of the difference changes your savings-rate or withdrawal-rate conclusion directly.

Our standard position bears restating: this is not advice to change allocation based on yields. It is advice to plan with more honest numbers. The allocation remains a low-cost, globally diversified core plus a cash buffer.

This chapter changes the number in your spreadsheet, not your holdings.

Chapter 2 handles the most basic and most overlooked layer of this framework: the risk-free rate, and why it determines everything else.