Chapter 4: The equity mix
The United States is 63.5% of the world's stock market, by the count on MSCI's ACWI factsheet (May 29, 2026), and roughly 80% of the typical American portfolio's stock sleeve. One of those numbers is a measurement. The other is a choice, and most of the people holding it never remember making it.
Chapter 2 wrote your target mix from the worst-year column. This chapter steps inside the stock half of that mix, where one question towers over the rest: how much United States, how much everywhere else. A second question, whether to tilt toward small or cheap companies, gets two honest paragraphs near the end and its full hearing in Chapter 8. By the last page you will write plan line four, and you will watch Dev, who has been dodging this exact decision for two years, write his.
The world, as the index makers count it
Index providers count by market value: add up the price of every public company, sort by country, and print the weights. On that count the US is 63.5% of MSCI's all-world index and 61.9% of FTSE Russell's global index (both factsheets, May 29, 2026). Of the share outside the US, roughly three-quarters sits in other developed markets, Japan, the UK, Europe, and Canada among them, and roughly one-quarter in emerging markets.
A portfolio that owns the world at those weights is the only stock mix with no opinion in it. It accepts the market's own prices as the verdict on every country, which is the same logic that made you index in the first place. Every other mix is a tilt, and an all-US sleeve is one of the largest tilts available: a bet of more than a third of the world's stock value against everything not American. Nothing about that bet is forbidden. The trouble is how it usually gets placed: not on a date, with reasons, but gradually, out of familiar fund names and whatever the 401(k) menu led with.
The case for owning more of the world
Vanguard's published guidance frames the decision this guide will use: hold at least 20% of your stock sleeve in international funds, and about 40% captures most of the diversification benefit their researchers can measure. Notice the shape of that advice. It is a band with a floor, written by people who model this for a living and who still decline to name one right number.
Valuation is the second exhibit, and this guide states it as levels only. CAPE, the cyclically adjusted price-to-earnings ratio, compares today's price with the past ten years of inflation-adjusted earnings. In June 2026 the S&P 500's CAPE sits near 39 against a long-run mean near 17, while most international markets trade near their own historical averages. Read plainly: a dollar of US earnings currently costs more than twice its long-run average price, and a dollar of international earnings sells for close to its usual one. A level tells you what you are paying. It does not tell you what happens next, and the strongest argument on the other side comes from the people who forgot that.
The case for staying home
The counterargument deserves real weight, because it has been winning for ten years. The US premium did not appear by accident: American companies out-earned the rest of the developed world for a decade, and the US index carries more of the kinds of companies that did the out-earning. That is the earnings-dominance and sector-mix case. An investor who shifted heavily abroad in the mid-2010s, back when the valuation gap already looked stretched, spent the following decade trailing and explaining.
A gap is not a timer. Expensive markets can stay expensive for years. A gap can close through earnings catching up rather than prices falling. Cheap markets are sometimes cheap for reasons that persist. If the international discount were free money, the institutions reading these same factsheets would have collected it already.
So the honest scoreboard reads close to even: one side holds the levels, the other holds a decade of delivered earnings, and neither side holds a calendar. That is exactly why the published guidance is a band. The job this chapter asks of you is smaller than refereeing the debate: pick a number inside 20–40 that you can hold through a stretch in which one side looks foolish, because one side will.
Why is "just the US" also a concentration bet?
Because the index inside most US-only sleeves leans hard on ten names. The ETFs & Funds Guide's overlap audit attached the receipts: the S&P 500's top ten holdings are 38.4% of the index, straight from the fund's own page (April 30, 2026), while a total international fund's top ten is just 12.0% of that fund. An all-US portfolio therefore rides more than a third of every indexed dollar on ten companies clustered around one sector's neighborhood, whatever name is on the fund. Chapter 3's rule was that diversification is measured in the crash you are planning for, and the crash a US-only sleeve should plan for is the one those ten names lead downward. An international allocation is, among other things, the cheapest dilution available for that bet: different giants, different sectors, different currencies, different central banks. The home-bias question and the concentration question are the same question wearing two hats.
Factor tilts, in two honest paragraphs
A factor tilt deliberately overweights stocks that share a trait, most often small companies or cheap ones (the trade calls cheap "value"). The long-run evidence, running through the Fama-French research lineage, is real: across the full historical record, small and value stocks have earned premiums over the broad market, and the finding was sturdy enough to reshape academic finance. A tilt is a legitimate, evidence-bearing choice inside your equity mix, and the Finvest Stocks Guide covers what those traits mean company by company.
The drought is just as real. The 2010s were the worst stretch for value investing since the Depression, and 2022, a historic year for value, arrived only after that decade had shaken out most of the people who tilted. That sequence is the entire risk: the premium pays on no schedule, the drought can outlast your patience, and a tilt abandoned mid-drought converts a paper disappointment into a permanent loss. Treat any tilt as a decade-scale commitment you sign in writing, or take market weights and skip it with a clear conscience. Chapter 8 runs the full stall, prices included.
Dev writes his number down
Dev's $140,000 sits in a three-fund portfolio at 80/20, which makes his stock sleeve $112,000. The international share inside it has been 30% since the day he copied it off a forum, and for two years he has quietly wondered whether to raise it for the valuations or cut it for the US run. This chapter handed him a cleaner exit: both arguments are serious, neither comes with a date, and his actual job is to pick a defensible number inside the band and stop relitigating it. He kept 30%, near the middle, and for the first time wrote down why. In dollars, the decision looks like this.
| Line | Share of portfolio | Dollars |
|---|---|---|
| US stocks | 56% | $78,400 |
| International stocks | 24% | $33,600 |
| Bonds | 20% | $28,000 |
| Total | 100% | $140,000 |
Inside the stock sleeve, $78,400 against $33,600 is exactly 70/30. The bond line waits for Chapter 5.
Dev's note, dated and saved beside his plan, runs four sentences. "30% of stocks international. Reasons: the US is 63.5% of the world but my US fund rides 38.4% of its money on ten names; Vanguard's band says 20 to 40 and most of the measurable benefit arrives by 40; I cannot referee CAPE 39 against a decade of US earnings, so I will stop trying. This number changes when these reasons change, never because one side had a hot three years." He says the last sentence is the whole point of the file: somewhere in year two of the next US winning streak, a future Dev will open it looking for permission, and the note is built to say no.
Pick an international share inside Vanguard's 20–40% band, write one sentence of reasons beside it, and allow changes only when the reasons change. Scoreboard envy after a US streak does not qualify, and neither does a valuation gap dressed up as a timer.
Where people go wrong
Letting the menu decide. Most all-US portfolios were never chosen; they accumulated from a 401(k) default and a famous index name. An unplaced bet is still a bet, and this one is more than a third of the world's stock value, declined by accident.
Trading the band like a dial. Sliding from 20 to 40 and back as US and international take turns winning collects the worst of both sides: you arrive at each region just after its run. The band is a one-time decision with a written reason, not a thermostat.
Reading the gap as a timer. CAPE 39 against international's ordinary levels is information about price, not a schedule. Going all-in abroad because of the gap is the same error as ignoring it, pointed the other way.
Buying funds instead of deciding the split. An international fund held at 5% beside two overlapping US funds is decoration. Run the overlap audit from the ETFs & Funds Guide, then set the split deliberately; the percentage is the decision, the fund is just the container.
1. Every goal, with its date and its bucket (Chapter 1).
2. Target mix: __% stocks / __% bonds, chosen from the worst-year column (Chapter 2).
3. The storm list: what protects me in each kind of crash (Chapter 3).
4. Inside stocks: __% US / __% international. New this chapter. Dev's line reads 70/30, with one sentence of reasons beside it.
Key takeaways
- The US is 63.5% of MSCI ACWI and 61.9% of FTSE's global index (factsheets, May 29, 2026); of the rest, roughly three-quarters is developed and one-quarter emerging. Market weight is the only no-opinion mix, and an all-US sleeve is a large tilt, usually placed by accident.
- Vanguard's published guidance: at least 20% of the stock sleeve international, with about 40% capturing most of the measurable diversification benefit. The shape of the advice, a band rather than a number, is itself the lesson.
- The valuations are levels, not forecasts: S&P 500 CAPE near 39 versus a long-run mean near 17, international markets near their own averages (June 2026). The counterargument is a delivered decade of US earnings dominance and sector mix, and a gap is not a timer.
- "Just the US" is also a concentration bet: the S&P 500's top ten is 38.4% of the index (fund page, April 30, 2026), against 12.0% for a total international fund's top ten.
- Factor tilts carry real long-run premia (Fama-French lineage) and real droughts: the 2010s were value's worst stretch since the Depression before 2022's historic year. Sign on for a decade or take market weights; Chapter 8 prices the aisle.
Sources: MSCI ACWI factsheet · FTSE Russell Global All Cap factsheet (May 29, 2026) · Vanguard on international allocations · AQR on value investing · Finvest ETFs & Funds Guide · Finvest Stocks Guide