Week 4: The 60/40 Portfolio — Why It Worked and Why 2022 Broke It

---

1. Why This Is Important

Sixty percent stocks, forty percent bonds. The most famous asset allocation in the history of investing, the default of every financial advisor, the benchmark every balanced fund is measured against, and — until 2022 — the closest thing to a "set it and forget it" portfolio that the textbook profession would put in print.

You need to understand 60/40 for four reasons, even if you end up not running it.

It is the baseline. Every more complex strategy — risk parity,

trend, factor tilts, the barbell shape we build in Level 5 (Weeks 47–52) — is measured against 60/40. You cannot evaluate any of those if you don't know what they are improving on.

It demonstrates what diversification actually buys you. When

stocks and bonds are negatively correlated, the 60/40 mix delivers roughly 80% of the equity return with about 60% of the equity risk. That is not arithmetic — that is a *correlation discount* on portfolio volatility, the central insight of Markowitz's Nobel-winning work.

It demonstrates the limits of diversification. In 2022, both

stocks and bonds fell roughly 18% together, the worst year for 60/40 since 1937. Understanding why matters more than the loss itself: the regime that made 60/40 work for forty years had a specific macro signature, and that signature reversed.

**It teaches you that correlation is the most important variable

in portfolio construction.** Returns shift the level of your wealth. Correlations shift the shape of the distribution. The stock-bond correlation flipped sign in 2022, and the entire industry is still digesting what that means.

This lesson covers the origin and mechanics, the historical performance by decade, the correlation history, the 2022 break, and the modern adaptations.

---

2. What You Need to Know

2.1 The Mechanics — Why Mix Stocks and Bonds at All?

The single insight: **the volatility of a portfolio is not the weighted average of the volatilities of its parts.**

For two assets with weights $w_1, w_2$, standard deviations $\sigma_1, \sigma_2$, and correlation $\rho$:

$$ \sigma_p = \sqrt{w_1^2 \sigma_1^2 + w_2^2 \sigma_2^2 + 2 w_1 w_2 \sigma_1 \sigma_2 \rho} $$

When $\rho < 1$, that square root is less than the weighted-average volatility. The closer to $\rho = -1$, the more risk reduction. Diversification is correlation arithmetic.

A worked example before the table makes the point concrete. Take US stocks at $\sigma = 16\%$ and US Treasuries at $\sigma = 6\%$, and a 60/40 mix. The naive guess — that 60/40 risk equals 60% of stock risk plus 40% of bond risk — gives $0.6 \times 16\% + 0.4 \times 6\% = 12.0\%$. That is what you would get if the two assets moved in lockstep. With a more realistic correlation of $\rho = -0.3$, the formula gives:

$$ \sigma_p = \sqrt{0.6^2 \cdot 16^2 + 0.4^2 \cdot 6^2 + 2 \cdot 0.6 \cdot 0.4 \cdot 16 \cdot 6 \cdot (-0.3)} \approx \sqrt{88.1} \approx 9.4\% $$

That gap from 12.0% down to 9.4% — about a 22% reduction in portfolio volatility for no give-up in expected return — is the entire point of mixing the two assets. It is not a budget calculation; it is the correlation discount Markowitz won the Nobel for.

For US stocks ($\sigma \approx 16\%$) and US Treasuries ($\sigma \approx 6\%$), with the long-run correlation between them:

Stock-bond ρ	60/40 σ	Diversification benefit
+1.0 (perfectly correlated)	12.0%	None — pure weighted average
0.0 (uncorrelated)	10.1%	About 16% volatility reduction
−0.3 (typical 1990s–2010s)	9.4%	About 22% reduction
−1.0 (theoretical)	7.2%	Maximum reduction

The 1990s–2010s regime delivered an average stock-bond correlation around −0.3. That is the entire reason 60/40 worked so well for so long. It was not about either asset's expected return; it was about the cross-correlation.

2.2 Historical Performance — The Forty-Year Tailwind

The chart below plots the cumulative real wealth (after CPI) of 100% stocks, 100% Treasuries, and 60/40 from 1928 forward, all rebalanced annually.

Read the chart from far right to far left. By 2024, $1 invested in 1928 in real terms became roughly:

• $565 at 100% stocks
• $128 at 60/40
• $4 at 100% bonds

The bonds-only line barely keeps up with inflation across the full century — the long-run real return on Treasuries is about 1.4% per year. Stocks compound at roughly 6.8% real. 60/40 lands at roughly 5.1% real, three-quarters of the way to the equity line in compound-rate terms but with materially smaller drawdowns along the way.

By decade, the picture is more nuanced. The drawdown columns are each asset's within-decade worst peak-to-trough loss in real terms (annual data, so an event that bottoms in the next decade is attributed there).

Decade	60/40 (real ann.)	Stocks (real ann.)	Bonds (real ann.)	60/40 max DD	Stocks max DD	Bonds max DD
1930s	1.0%	-0.1%	4.7%	-23%	-38%	-1%
1940s	1.6%	3.6%	-3.5%	-23%	-25%	-32%
1950s	12.5%	16.7%	-2.6%	-6%	-13%	-11%
1960s	3.5%	5.0%	-0.7%	-12%	-14%	-17%
1970s	-0.7%	-1.4%	-1.0%	-36%	-49%	-33%
1980s	9.7%	12.0%	7.2%	-8%	-13%	-9%
1990s	8.7%	14.8%	4.6%	-5%	-1%	-11%
2000s	1.8%	-3.4%	4.4%	-15%	-37%	-13%
2010s	6.8%	11.4%	1.6%	-4%	-6%	-10%
2020-24	1.6%	6.7%	-3.5%	-23%	-23%	-34%

The 1980s, 1990s, and 2000s are the three decades that built 60/40's reputation. The 1980s–90s contributed the return story: real bond returns above 7% per year in the '80s, real stock returns above 12% in the '90s. The 2000s contributed the protection story: stocks lost 3.4% per year in real terms across the dot-com bust and GFC, and the bond sleeve carried the 60/40 portfolio to a positive real return anyway. That decade is where retail advisors learned to trust the bond hedge, not just price it. Across all three decades the stock-bond correlation ran around −0.3 to −0.4. Whatever-asset-was-falling-the-other-was-rising, and the rebalance trade paid you to mechanically buy the underperformer. The 2020–24 row is what concerns the modern industry — and it is the row the rest of this lesson exists to explain.

Why is rebalancing actually good — and why isn't 60/40 always worse than 100% stocks?

Look at the table honestly. Across the full 97-year window, 100% stocks compounds at 6.8% real and 60/40 at 5.1% — roughly $565 vs $128. **Over a long horizon, with no behaviour problem, 100% stocks wins on terminal wealth.** The course is not going to pretend otherwise. The case for 60/40 is not "higher return"; it is three related effects:

Drawdown compression. In every crisis decade — 1930s, 1970s,

2000s — 60/40 took materially less of the equity loss. -38%, -49%, -37% on stocks compressed to -23%, -36%, -15% on 60/40. That smaller hole is the behavioural alpha: an investor who does not panic-sell at -25% is a different person from one who panic-sells at -49%.

The rebalancing premium. Annual rebalancing forces a

mechanical "sell what just outperformed, buy what just underperformed." Over decades that adds roughly 0.2–0.4% per year on a 60/40 mix (Bouchey, Nemtchinov, Paulsen and others estimate it in this range). It is not a free lunch — it requires both assets to be roughly mean-reverting and negatively correlated, which is exactly what failed in 2022 — but in normal regimes it is real and persistent.

Volatility drag is non-linear. A portfolio that goes -50%

then +50% lands at -25% of starting value. A portfolio that goes -25% then +25% lands at -6%. Cutting the drawdown in half does more than halve the recovery cost. Lower-vol portfolios compound geometric returns closer to their arithmetic returns.

Honest framing for the lesson: **the correct comparison for a buy-and-hold investor with a 30-year horizon is 100% stocks vs 60/40 vs the barbell — not 60/40 vs nothing.** 100% stocks wins on terminal wealth across long enough windows; 60/40 wins on the worst two-year stretch you have to live through; the barbell tries to keep most of the equity upside while replacing the middle-of-the-curve bonds with structurally better diversifiers (cash + gold + tail hedges). We come back to that comparison at the end of this lesson.

Can rule-based / "timed" rebalancing improve on the calendar?

This is a real research literature, not folklore. The headline finding: **rule-based variants modestly improve on calendar rebalancing in backtest, but the improvement is small relative to the variance across implementations and is fragile out of sample.**

• Threshold (band) rebalancing — rebalance only when the

allocation drifts more than 5% in absolute terms (e.g., from 60% stocks to 65%) — captures most of the rebalance premium with fewer trades and less tax friction. Vanguard's working papers find it is marginally better than annual calendar rebalancing on a long-run US 60/40 backtest, ~0.05–0.15% per year.

• Trend-overlay rebalancing — only rebalance into an asset when

it is above its 200- or 10-month moving average — improves drawdowns in some windows (notably 2008) and worsens them in others (notably 2020 V-shaped recovery, where you sold equities in March and bought back at a higher price in August). Net effect is close to a wash on long horizons.

• Valuation-tilt rebalancing — over-weight equities when CAPE is

cheap, under-weight when expensive. Long-horizon backtests show ~0.3–0.6% per year of improvement on US data, but with very long drag spells (the strategy was wrong about US equities for most of 2010–2020) and with severe data-mining concerns: there is one US history, the rules are tuned on it, and the next 30 years are not the previous 30 years.

Bottom line: a calendar or band-rebalanced 60/40 captures roughly all of the robust premium. Anything more clever crosses the line into active allocation — which can work, but at that point you are no longer running a passive 60/40. We come back to discretionary rebalancing under "Misconceptions" below and again in §2.7.

2.3 The Correlation Story — The Hidden Variable

The chart below is the rolling 36-month correlation between US stocks and US Treasuries from 1928 forward. The single most important fact in the entire lesson lives on this chart.

Three regimes are visible:

• 1928–1997: stocks and bonds usually moved together. Inflation

was the dominant driver. Higher inflation meant both lower bond prices (rates rose) and lower stock prices (real earnings squeezed). Lower inflation lifted both. Correlation positive, diversification effect mediocre.

• 1998–2021: the great inverse window. Inflation faded as the

dominant driver and growth/recession became dominant. In recessions, the Fed cut rates → bonds rallied. In recessions, earnings cratered → stocks fell. Correlation deeply negative. 60/40 delivered its best forty years ever.

• 2022 onward: the inverse regime cracked. When the Fed had to

tighten aggressively to fight 9% CPI, bonds fell and the earnings multiple compressed and stocks fell. Correlation flipped positive again. 60/40 had its worst year since the Great Depression-era stretch.

The simple frame: **stocks and bonds are negatively correlated when the dominant macro driver is growth/recession; they are positively correlated when the dominant driver is inflation.** Decide which regime you think the next decade will be, and you have decided whether 60/40 is still your friend.

2.4 Bond Yields Are a Policy Variable, Not a Market Price

Before we get to 2022, one thing every investor running 60/40 needs to understand and almost no certification curriculum says out loud: the bond half of your portfolio is not priced in a free market. The level and shape of the yield curve is set, in large part, by the central bank. Short rates are literally set by the Federal Reserve at the FOMC meeting. Long rates are nominally a market price but are heavily steered by quantitative easing and tightening, by forward guidance about future policy rates, and — in some jurisdictions — by explicit yield-curve control (YCC), where the central bank commits to buying or selling unlimited size to peg the yield at a chosen level.

The cleanest worked example is the Bank of Japan. From 2016 to 2024 the BOJ ran explicit YCC, capping the 10-year Japanese government bond yield first at 0% then at 0.25%, then 0.5%, then 1.0% — every "adjustment" was a mini policy regime change that moved the entire JGB curve overnight. JGB holders for those eight years were not holding a free-market asset; they were holding a policy promise whose payoff depended on the BOJ's willingness to keep printing. When the BOJ exited YCC in 2024, the long end repriced sharply.

The Fed has done softer versions of the same thing. The 2008–2014 QE program suppressed long-Treasury yields by an estimated 100–150 basis points below where a free market would have cleared. The 2020–2021 emergency easing kept the 10-year below 1% for over a year. The 2022 tightening was the unwind of those policy positions, not a market-driven move.

What this means for your 60/40 portfolio is uncomfortable:

• The bond sleeve's expected return is the current yield, and the

current yield is partially a policy variable. When you buy a 10-year Treasury at 1% yield, you are accepting a return path that the Fed has explicitly chosen. When the Fed changes its mind — because of an inflation shock, a fiscal-dominance pivot, or a new chair with a different reaction function — the bond sleeve's price moves, sometimes brutally.

• "Tail risks" for the bond sleeve are policy events: yield-curve

control, financial repression, an explicit higher inflation target, a debt-monetisation regime, restrictions on foreign Treasury holdings. None of these are tradable risks in the Markowitz sense; they are political events.

• The classic 60/40 requires the Fed put to work. The unstated

assumption of the last forty years is: when stocks crash, the Fed cuts rates aggressively, bonds rally, and the bond sleeve carries the portfolio. Take away the Fed's willingness or ability to cut — because inflation is too high (1970s, 2022) or because rates are already at zero (2020 was the last time this constraint bit) — and the bond sleeve has nothing to give.

This is part of why bond yield level matters so much for whether 60/40 is sensible at any given moment (we revisit this directly in Q1). At a 1% 10-year yield, the bond sleeve has almost no income buffer and enormous duration risk on a rate-up move. At a 4–5% yield the math is genuinely different: even a 100 bp rate shock leaves the bond sleeve roughly flat over a year. The 60/40 portfolio is a different instrument depending on where the central bank has set the yield curve when you start.

2.5 The 2022 Debacle — What Actually Happened

In one calendar year:

Asset	2022 total return
S&P 500 (incl. dividends)	-18.1%
10-year Treasury (incl. coupons)	-17.8%
60/40 portfolio	~-18.0%
CPI (inflation)	+6.5%
60/40 in real terms (after CPI)	~-24%

That last number is the worst real-return year for the 60/40 portfolio since 1937. The mechanism was simple and terrifying. The Fed funds rate started 2022 at 0.25% and ended above 4%. Long bonds repriced because of the rate move. Stocks repriced because the discount rate on future cash flows is the long bond yield, and that doubled. Both fell. Together. With no place to hide for the unhedged investor.

The 2022 drawdown also showed something else: **a 60/40 portfolio provides no protection against an inflation shock.* Cash lost less* than long bonds because cash has near-zero duration and short T-bill rates kept reinvesting at the rapidly rising Fed funds rate; gold roughly held value; commodities rose. The classic two-asset diversifier was the worst allocation in the year of high inflation that everyone had been talking about for a decade.

What broke — a checklist

Five things had to be true for 60/40 to keep working. In 2022, all five flipped at once. Use this as your forward-looking dashboard:

Pillar	1990s–2010s setting	2022 setting
Inflation regime	Falling / anchored ≤2%	Spike to 9%, sticky
Policy rates	Low and falling, Fed put active	Rapid hiking, Fed put withdrawn
Stock-bond correlation	~−0.3 (negative)	Sharply positive
Long-bond duration risk	Hidden by 40-year bull market	Realised — 30%+ losses on TLT
Real-return capture	Stocks + bonds beating CPI	Both lost real value together

Read the right column as the warning sign. If those settings persist or recur — if inflation stays sticky and the Fed cannot credibly cut into a recession because CPI is still 4%, if the correlation stays positive — the 1990s–2010s shape of 60/40 will not return. If they reverse — disinflation back to 2%, growth scares pulling rates lower, correlation flips back negative — 60/40's "golden age" framing is appropriate again.

2.6 Modern Adaptations — Where 60/40 Goes Now

A short history note before the adaptations. **60/40 is not a Ray Dalio invention.** The 60/40 mix is the conventional advisor and balanced-fund default that predates risk parity by decades — the "60% stocks, 40% bonds" rule of thumb appears in pension and trust literature back to the 1950s. Dalio's contribution is the **All Weather portfolio* (and its institutional sibling, risk parity*): roughly 30% stocks, 55% long bonds, 15% commodities and gold by capital weight, but balanced so each asset class contributes equal risk. The unifying idea — diversify across macro regimes rather than across asset labels — is genuinely different from 60/40. All Weather was designed precisely so that no single regime (growth shock, inflation shock, deflation shock, recession) could sink the portfolio. We come back to risk parity and All Weather explicitly in Week 15 ("Multi-Asset and Risk Parity") and again in Level 5; for the rest of this lesson we keep the focus on the plain-vanilla 60/40 baseline.

Three modifications, in increasing order of complexity.

Replace some bonds with cash. When bond yields are below

inflation, T-bills (short-duration, highest reinvestment of new yield) outperform long bonds in nearly every scenario. A 60/30/10 allocation (stocks/short-Treasuries/cash) gives up almost no long-run return and dramatically reduces 2022-style drawdowns.

Add a 5–10% gold sleeve. Gold's correlation to stocks is near

zero in normal regimes and turns negative in inflation shocks. The classic Permanent Portfolio (25/25/25/25 stocks / long-bonds / cash / gold) is the historical archetype; modern variants run smaller gold weights with a bigger equity tilt. Gold is not free — it has no yield and high real-rate sensitivity — but in the regime where 60/40 broke, gold worked.

Add a long-volatility or trend-following sleeve. A small

allocation (5–10%) to managed-futures or long-vol structures pays for itself in tail events. This is the institutional adaptation; we cover it in detail in Week 47 and Week 50.

Comparing the major archetypes side-by-side:

Allocation	Stocks	Bonds (long)	Cash / short Tsy	Gold	Other	Best regime	Worst regime
100% stocks	100	—	—	—	—	Long expansions, low CPI	Deep bear / GFC-style crash
Classic 60/40	60	40	—	—	—	Disinflation + Fed put (1990s–2010s)	Inflation shock (1970s, 2022)
60/30/10	60	30	10	—	—	Same as 60/40 but cheaper drawdown in rate shocks	Same as 60/40, milder
55/30/10/5	55	30	10	5	—	Adds inflation buffer; closer to all-weather	Sustained negative real rates with no inflation
Permanent Portfolio	25	25	25	25	—	Any single-regime shock	Long bull market in stocks (gives up upside)
Dalio All Weather (~)	30	40	—	7.5	22.5 commodities/TIPS	Diversified across all four macro regimes	Persistent stagflation with rising real rates
Barbell (Level 5 preview)	50–70 (concentrated + asymmetric)	0	20–35	5–10	5–10 tail hedges	Volatile regimes with both up- and down-tails	Long quiet grind upward (overpays for hedges)

The honest framing is the one Horace pushes throughout the course: **60/40 worked because of a specific macro regime that is unlikely to repeat with the same intensity over the next decade**. It is not broken. It is no longer optimal. The barbell shape — concentrated safety on one end, asymmetric speculation on the other, very little in the structurally-mediocre middle — is the more honest answer for investors who can stomach a different shape of returns.

Important caveat. The barbell is an advanced migration path, not a Week 4 implementation instruction. It requires the option, hedging, sizing, and tax tools we develop across Levels 2–4 (especially Weeks 25–30, 41–42, 47, and 50). Do not try to build a barbell after this lesson. Build the 60/40 (or 60/30/10) baseline first, run it through one full crisis cycle, and only then consider the barbell migration described in Week 52. The right Week 4 takeaway is "I now understand the baseline that every later allocation will compare to."

2.7 Rebalancing Strategies for 60/40

Three families of rebalancing rule, ordered by how much work they take and roughly how much they capture of the rebalancing premium.

Rule	What it does	Trades / year	Premium captured	Tax cost (taxable account)
Buy & hold (no rebalance)	Let weights drift forever	0	None — equity allocation grows toward 100%	Zero
Calendar — annual	Reset to 60/40 every January	1	Most of it (~0.2–0.4% / yr)	Moderate
Calendar — quarterly / monthly	Reset on schedule	4–12	Marginal additional gain; transaction & tax costs eat it	High
Threshold / band — 5% absolute	Rebalance only when stock weight drifts ≥5% from target	~0.3 in calm markets, several in crises	Comparable to annual, with fewer trades	Lower than annual
Threshold / band — relative (e.g., ±25% of target)	Wider bands; even fewer trades	~0.1–0.5	Slight underperformance vs annual	Lowest
Trend / momentum overlay	Only rebalance into an asset when it is above its trend (e.g., 200-day or 10-month MA)	Variable	Helps in trending bears (2008); hurts in V-recoveries (2020)	Variable
Valuation tilt	Over-weight the cheaper asset (CAPE, real-yield rules)	1–4	~0.3–0.6% / yr in long backtests; severe data-mining risk	Moderate–high

Two practical recommendations:

**For most retail 60/40 investors, threshold rebalancing on a 5%

absolute band, checked quarterly, is the right answer.** It captures essentially all of the calendar-rebalance premium with fewer trades, lower taxes, and one psychological benefit: you only act when something has actually moved enough to matter.

**Direct new contributions toward the under-weight sleeve before

ever selling.** Selling-to-rebalance is the last resort — contribution-rebalancing has zero tax cost and zero spread cost. Many investors going through their accumulation years never need to sell-rebalance at all if they are still adding new money.

The deeper trade-off — when "rule-based" rebalancing crosses the line into discretionary active management — is covered under Misconception 5 below.

---

3. Common Misconceptions

Misconception 1: "Bonds are the safe part of 60/40."

Bonds are less volatile than stocks. They are not safe. In 2022, US Treasuries lost more than they had in any year of the prior seven decades. Long bonds lost roughly 30%, which is a larger loss than the median equity bear market. The "bond sleeve = safety" framing is a 1980–2020 artefact of a 40-year bull market in bonds caused by falling rates.

Misconception 2: "60/40 has always worked."

In real terms, 60/40 lost money over the entire 1965–1981 stretch. For 16 years a balanced investor saw real wealth shrink because inflation outran the combined nominal returns of stocks and bonds. It was not until the 1980s that the long-run real trend returned.

**Misconception 3: "Adding bonds to a stock portfolio always reduces return."**

Not necessarily. When you rebalance annually, the rebalancing trade — selling whichever asset just outperformed and buying the underperformer — is a small but persistent source of additional return on top of the buy-and-hold blend, because it is a mechanical "buy low, sell high." Some 60/40 backtests over long horizons show higher compound returns than 100% stocks because of this rebalancing premium plus the smaller drawdowns. The free-lunch component is small, but real.

Misconception 4: "International diversification fixes the problem."

International stocks have a high correlation to US stocks in the events that actually matter (recessions, financial crises, the COVID crash). They reduce some country-specific risk but do little to solve the stock-bond correlation problem in 60/40. Asset-class diversification, not geographic diversification, is what shifts the portfolio's response to inflation shocks.

Misconception 5: "60/40 is actively managed if you rebalance."

Calendar or band rebalancing back to a fixed allocation is not active management — it is a mechanical rule with no view on the future. Strict 60/40 with annual or 5%-band rebalancing is fully passive at the strategy level. The key word is fixed.

Where the line moves is when the target itself changes based on a view: reducing the bond weight when yields are low, raising stocks when CAPE is cheap, switching to cash when the 200-day moving average breaks. Those are active decisions. They may each be coded as a mechanical rule, but the choice of rule is a discretionary view on the future of returns and correlations. Once you do that, you are no longer running passive 60/40 — you are running a tactical allocation overlay, and you should evaluate it as one (alpha, edge, out-of-sample robustness, drawdown vs benchmark). The honest test is: would a different family of rules — selected today, with the same logic — give a different allocation? If yes, it is active.

**Misconception 6: "I should rebalance frequently to capture the rebalance bonus."**

The rebalancing premium is real but small (~0.2–0.4% per year on typical 60/40 backtests). Quarterly or annual rebalancing captures most of it; weekly or daily rebalancing has three problems that together usually destroy the premium:

Selling against momentum. Most market moves run in trends

over horizons of weeks-to-months before mean-reverting. Rebalancing weekly forces you to sell whatever just rallied while it is still rallying — selling a winner that is about to keep winning is the worst version of "buy low, sell high."

Transaction costs and bid-ask. Each unnecessary trade pays

the spread plus, in some accounts, commissions. On a 60/40 portfolio with weekly rebalancing the round-trip cost can match the entire rebalance premium.

Tax friction in taxable accounts. Every sell-side rebalance

is a realisation event. Short-term gains are taxed as ordinary income (often >35% federal + state). A weekly rebalancer in a taxable account can give all of the rebalance premium back to the IRS — and then some. This is exactly why §2.7 recommends contribution-rebalancing first and threshold-band rebalancing second; both minimise realisations.

Annual is the conventional answer; semi-annual or 5%-band quarterly is fine. More frequent than that is over-engineering that pays momentum, spread, and the tax man for the privilege of looking busy.

---

4. Q&A

Q1: Should I run 60/40 in 2026?

A: Honestly, "it depends on the bond yield." 60/40 is most defensible at higher nominal bond yields and least defensible at low ones, because the entire bond sleeve is asymmetric to the starting yield. At a 1% 10-year yield (the 2020–2021 setup), the bond half delivers 1% of carry and takes a -10% hit on every 100 bp of rate-up move — that is a structurally bad trade. At a 4–5% 10-year yield (closer to where the curve has reset post-2022), the bond half delivers a meaningful coupon, the duration risk is partially compensated, and the math of 60/40 is genuinely much better. As of writing, US 10-year yields in the 4-handle make 60/40 more defensible than it was in 2020 — but still inferior to the cash-tilted variants.

The deeper issue is the Fed-put assumption. The classic 60/40 implicitly assumes that when stocks crash, the Fed will cut rates aggressively, bonds will rally, and the bond sleeve will carry the portfolio. That assumption holds whenever inflation is anchored near 2% (1990s–2010s). It breaks whenever inflation is sticky and the Fed cannot credibly cut without re-igniting CPI (1970s, 2022). Before committing to 60/40, ask yourself: *do I believe the next crisis will be deflationary (Fed cuts, bonds save me) or inflationary (Fed cannot cut, bonds fall with stocks)?*

For a long-horizon investor with no edge and no time to manage a portfolio actively, 60/40 is still a defensible answer — but the better defensible answer in the current regime is 60/30/10 (stocks / short-duration Treasuries / cash) or 55/30/10/5 (adding gold). Pure long-bond 40% is the part most exposed to a reversal of the 2022-style positive correlation regime, and it is also the part most exposed to a Fed that is no longer free to cut.

Q2: Why not 70/30 or 50/50? What's special about 60/40?

A: Nothing is special — it's a convention, not a derivation. The Sharpe ratio is fairly flat from about 40/60 to 70/30. The "right" allocation is the one you can hold through a 35% drawdown without selling. 70/30 is for higher-tolerance / longer-horizon investors; 50/50 is for shorter-horizon / lower-tolerance ones. 60/40 sits in the middle and got institutionalised because of that.

Q3: How do I implement 60/40 with ETFs in practice?

A: Two ETFs is enough. VTI (or VOO/SPY for S&P 500) for the equity sleeve; AGG or BND for the bond sleeve. Set monthly automatic contributions in 60/40 proportions. Once a year, rebalance back to 60/40. Total annual time commitment: about thirty minutes. Total expense ratio: roughly 0.04%.

Q4: What about "all-in-one" balanced funds?

A: Vanguard's VBAIX (60/40), iShares' AOR (60/40), and similar products do the rebalancing internally. Slightly higher expense ratio (0.07–0.30% vs 0.04% for the two-ETF version), no rebalancing chore for you. Reasonable for accounts you don't want to touch. For taxable accounts the two-ETF version is better because you can control the tax timing on rebalances.

Q5: Why did 2008 not break 60/40 the way 2022 did?

A: In 2008, stocks fell ~37% and long Treasuries rallied roughly +20%. The 60/40 drawdown was about −22% — bad, but materially less than the equity drawdown. The bond rally had two compounding causes: (1) **the Fed cut the funds rate from 5.25% in mid-2007 to essentially 0% by December 2008** — a roughly 500 bp easing cycle that mechanically lifted long-bond prices through duration; and (2) a flight-to-safety bid, where investors fleeing equities and credit parked cash in Treasuries, adding to the price move on top of the policy-rate cut. The Fed-cut leg is by far the bigger driver of the 20% bond rally — pure flight-to-safety in a quiet rate environment usually delivers a few percent at most. In 2022 the Fed was raising rates instead of cutting, so neither force was available: there was no monetary tailwind for bonds and no durable safe-haven bid (bonds were the asset people were fleeing from). Mechanism difference: 2008 was a deflationary credit shock with the Fed put fully active; 2022 was an inflationary monetary shock with the Fed put withdrawn. The 60/40 portfolio is hedged against the former and exposed to the latter.

Q6: Should I rebalance with new contributions or by selling?

A: With new contributions if you can. Direct your monthly inflow to whichever sleeve is below target — that rebalances without realising taxable gains and without paying spread. Sell-to-rebalance only when contributions are no longer enough to fix the drift, or once a year as a clean-up.

Q7: What about international bonds?

A: For US-domiciled investors, hedged international developed-market bonds (BNDX) provide marginal additional diversification. In practice the diversification benefit is small (the world's investment-grade sovereign debt is heavily correlated through the global rate cycle), and the FX hedge eats some of the yield. Most practitioners skip them and just hold US Treasuries.

**Q8: Where does a "barbell" allocation fit relative to 60/40?**

A: The barbell rejects 60/40 at the philosophical level. The middle of the risk spectrum (investment-grade bonds, dividend-stock defensive allocations) is exactly the part the barbell strips out because it is the part that gets crushed in inflation shocks. The barbell holds more concentrated safety than 60/40 (cash, short T-bills, gold) and more asymmetric speculation than 60/40 (long calls, momentum equity, Bitcoin or specific asymmetric trades sized as speculation — not a generic "crypto" allocation), with very little in between.

Q9: Is the rebalancing premium taxable?

A: In a taxable account, every rebalance trade is a potential realisation. To preserve the premium net of taxes, rebalance with contributions where possible (no realisation), and use tax-advantaged accounts (IRA, 401k, 529) for the sleeve that accumulates fastest-growing gains. The two-ETF 60/40 is genuinely cheap on expense; on taxes, your account structure matters more than the ETF choice.

Q10: How does this connect to the rest of the course?

A: 60/40 is the baseline. Week 5 goes deep on the bond half (coupons, prices, yields, duration) — the mechanics behind why the bond sleeve behaved the way it did in 2008 and 2022. Week 6 covers gold and commodities, which are the natural inflation-shock complements to 60/40. Week 11 merges rebalancing into the behavioural-discipline lesson and covers the mechanics in detail. Week 15 introduces multi-asset and risk parity — the Dalio "All Weather" alternative to 60/40. Week 23 covers factor investing, which slices the equity sleeve into return premia. Week 47 covers the long-volatility / tail-hedge sleeve that you can graft onto 60/40 to address the inflation-shock vulnerability. Level 5 (Weeks 47–52) is where the barbell portfolio shape is actually built. Every subsequent allocation discussion will compare to the 60/40 baseline you have just learned.

Interactive panel

The interactive panel below is a generalisation of the cumulative wealth chart in §2.2: it starts from the same Damodaran 1928–2024 dataset and the same 60/40 line, but lets you sweep the stock weight in 10% steps from 0% to 100% (so the static 60/40 line in the §2.2 chart is what you see when the slider sits at 60%), toggle between annual rebalancing and pure buy-and-hold, and switch between real (CPI-adjusted) and nominal terms. Underneath the wealth curve it draws the rolling drawdown from each running peak, and reports the geometric annual return, annualised volatility, worst drawdown, and Sharpe ratio (using the 3-month T-bill as the risk-free rate). Slide it. Watch where the Sharpe-ratio peak sits, and watch how the 1973–74 and 2008 drawdowns change shape as you move the stock weight.