Stocks for the Long Run Ch. 2: The Bond Surprise — Why "Safe" Means Something Different Over 30 Years

Investment Background

Chapter 1 gave the equity number. This chapter gives the bond number, and it is considerably lower than most people expect.

Over the same two centuries, US long-term government bonds returned roughly 3.5% real annually. Short-term Treasury bills returned less — about 2.7%.

The three numbers together:

Asset	Annualized real return (approx.)
Equities	6.5–7%
Long-term government bonds	3.5%
Treasury bills	2.7%

The gap looks like about three percentage points. After thirty years of compounding, it is not three percentage points.

The Wall Street Translation

Let us convert the gap into actual money, because percentages disguise the scale of compounding.

Take $10,000 invested for thirty years at real rates (so the final figure represents today's purchasing power):

Asset	Real purchasing power after 30 years (approx.)
Equities (6.5%)	about $66,000
Long-term bonds (3.5%)	about $28,000
Treasury bills (2.7%)	about $22,000

The equity outcome is more than double the long bond outcome.

A three-point annual difference becomes a more-than-2x difference over thirty years. That is the nonlinearity of compounding, and it is why the Chapter 1 number matters.

The Real Surprise: 1940 to 1981

Those figures are averages. What makes this chapter valuable is the worst stretch.

From the 1940s through 1981, US long-term government bonds suffered a forty-year disaster in real returns.

The cause was inflation. Throughout that period, inflation persistently exceeded the coupons the bonds had been issued at. By the time inflation peaked in 1981, an investor who bought long Treasuries in 1940 and held had lost more than half their purchasing power.

Note carefully what that sentence means. They did not lose money nominally — nominally they received every coupon and their principal in full.

What those dollars could buy had fallen by more than half.

Forty years. Holding the asset described as "risk-free."

Two Different Meanings of the Word "Safe"

This is the heart of the chapter, and its most useful contribution to retirement planning.

Bonds genuinely are safe over short horizons, and that is not an illusion.

If you are buying a house next year, putting the down payment in Treasury bills means you can be nearly certain the money will be there. Equities cannot promise that — they might fall 30% next year.

On a one-year horizon, bonds are safe and stocks are dangerous. That is true.

Now look at a thirty-year horizon:

Over thirty years, equities have almost never lost purchasing power, while bonds once lost half of it.

Both statements are correct. They describe different risks.

Let us name the two risks precisely, because confusing them is one of the most common errors in retirement planning:

Risk type	Definition	Most dangerous over	Who it hurts most
Volatility risk	Sharp price decline in the short term	1–5 years	Someone who needs the money next year
Purchasing-power risk	Inflation eroding the asset's real value	10–40 years	Someone who must live 30 years in retirement

Bonds protect you from the first risk while exposing you directly to the second.

Equities are terrible at the first risk and are the best historical protection against the second.

Why This Matters Especially for Retirees

Someone retiring at 65 does not have a one-year planning horizon, or a five-year one.

If you retire at 65 and live to 95, your investment horizon is thirty years.

This is routinely overlooked, because the word "retirement" feels like an endpoint. It is not an endpoint. It is the beginning of a thirty-year portfolio that must simultaneously pay your living expenses.

Over thirty years, even mild inflation erodes enormously:

Annual inflation	Purchasing power remaining after 30 years
2%	about 55%
3%	about 41%
4%	about 31%

At 3% inflation — an entirely normal figure — a dollar today buys about forty-one cents' worth of goods in thirty years.

Which means a retiree who "safely" holds everything in bonds and cash is taking an enormous risk disguised as safety.

A Necessary Balance

This chapter has so far argued for equities. The other half must be stated, or this becomes dangerous advice.

The role of bonds in a retirement portfolio is not to provide returns. It is to provide the ability not to sell equities.

This connects directly to the conclusion of Chapter 5 of When Genius Failed in this library: time can be bought with cash.

A retiree holding two to three years of expenses in bonds and cash is not trying to grow that money. The purpose is that during the three years the stock market is down 40%, they can draw living expenses from bonds instead of selling equities at the bottom.

Bonds are the buffer, not the engine.

This is precisely the core of the bucket strategy in Chapter 3 of Retirement Decumulation Mechanics. Treat bonds as the engine and you lose to inflation. Treat bonds as the buffer and they are indispensable.

Executable Trading Rules

1. Choose assets by your actual time horizon, not by your age. Retiring at 65 does not make your investment horizon zero. Ask "when do I need this money," then allocate to that answer. Money needed within five years belongs in bonds and cash; money not needed for twenty years sitting in bonds is taking purchasing-power risk.

2. Always evaluate bonds in real terms. A bond yielding 5% nominal under 4% inflation returns 1% real. The most deceptive thing about bonds is that the nominal number is always positive.

3. Size your bond holding against your spending needs, not against an age formula. Rules like "100 minus your age" ignore your actual cash flows. The right question is: how many years of spending buffer do I need in order not to sell equities at a low? The answer is usually two to three years of essential spending.

4. Understand the special position of inflation-linked bonds. US Treasury Inflation-Protected Securities are one of the few instruments offering short-term stability and long-term purchasing-power protection simultaneously. Within this chapter's framework, they are the one asset class addressing both risks at once.

5. Do not treat "I have never lost money in my life" as success. Someone who held only savings accounts and CDs their whole life never lost nominally — and may have lost more than half their real purchasing power. That is the chapter's most important psychological reminder.

Relevance to a Retirement Portfolio

This chapter's conclusion is the empirical basis for our asset allocation position across this site:

A retirement portfolio must defend against two risks at once, and no single asset can do both.

• Equities: defend purchasing-power risk, exposed to volatility risk.
• Bonds and cash: defend volatility risk, exposed to purchasing-power risk.
• The combination of the two: this is not a compromise. It is the only solution.

The operational steps appear in our Asset Allocation Optimizer and in Retirement Decumulation Mechanics.

Our standard position bears restating: this chapter is not advising you to raise equity exposure in pursuit of returns. It establishes that holding everything in nominally safe assets is an unrecognized risk decision, not a risk-free one.

Chapter 3 handles a detail skipped in Chapter 1: what that 6.5% real return is actually made of. The answer will change how you look at dividends.