Chapter 7: The classic recipes

Walk into the industry's store and the shelf looks endless: thousands of funds, hundreds of model portfolios, and a strategy with your name on it if your balance is large enough. Stand back and the shelf collapses to four recipes. All four are built from the same three ingredients, a total US stock fund, a total international stock fund, and a total bond fund, and the ingredients cost 0.03% to 0.05% a year. The prices on the shelf run from $30 a year per $100,000 to fifty times that. Nearly everything above the ingredient cost is an assembly fee, and this chapter prices the assembly.

You arrive at this shelf in an unusual position. Six plan lines are already written. Your goals have dates (chapter 1), your mix has a worst-year number you chose on purpose (chapter 2), your stock sleeve has an international share (chapter 4), your bonds have a job and a duration (chapter 5), and your cash covers dates rather than forecasts (chapter 6). Hold that plan up against the shelf and every recipe becomes legible, because each one is somebody's pre-packaged answer to the questions you have already answered.

Recipe one: three funds, or two

The three-fund portfolio is the plainest recipe sold anywhere: one total US stock fund, one total international stock fund, one total bond fund. Taylor Larimore popularized the convention on the Bogleheads forum, and its most underrated feature is what it refuses to say. The recipe is deliberately silent on the percentages. It hands you three ingredients and sends you back to your own chapter 2 and chapter 4 answers for the amounts. That silence is the design working as intended: the splits are personal, and the ingredients are not.

The price is the argument. Funds of this type charge about 0.03%, 0.05%, and 0.03% a year. On $100,000 split 50% US stocks, 30% international stocks, and 20% bonds, the yearly cost is $15 plus $15 plus $6, which comes to $36, roughly the price of a large pizza. A two-fund portfolio compresses the recipe further: one total world stock fund plus one bond fund, which removes the international decision and adds a few hundredths of a percent for the convenience. Both versions ask for one payment beyond the fee: you do the assembly, the occasional rebalancing chapter 10 schedules, and the not-fiddling in between.

Recipe two: the fund with a year in its name

The target-date fund is the recipe that assembles itself. Morningstar's research put the category at $4.8 trillion at the end of 2025, and industry coverage reports it crossing $5 trillion since. The asset-weighted average fee was 0.27% in 2025. Collective investment trusts, a 401(k)-only wrapper that does the same job at lower cost, now hold 54% of target-date assets, and Vanguard alone runs about $1.8 trillion, roughly 37% of the category. The standard glide holds about 90% stocks when retirement is decades away, eases toward 50% near the date, and settles around 30% afterward.

The Finvest ETFs & Funds Guide already opened one of these up: the four index funds under the single ticker, the glide path as the actual product, the "to" versus "through" fork, and the taxable-account caveat. None of that needs re-teaching here. What belongs in this chapter is the shelf placement: a cheap index-based target-date fund inside a 401(k) is a complete recipe at a fair assembly fee, and the 0.27% category average means many series charge several times what the cheapest do for the same job. Two newer variations deserve a flag on the way past: target-date funds with annuities inside (about $42 billion across 13 series as of March 2026) and target-date funds with private assets (announced in 2025). Both add a seller to the recipe, and both get the chapter 8 treatment.

Recipe three: the 60/40, the granddad

Before target dates and robos, balanced investing had one number pair: 60% stocks, 40% bonds. The 60/40 earned its reputation across decades in which the two sides took turns, and then 2022 arrived. Stocks fell 18.1%, the aggregate bond index fell 13.0%, and a 60/40 blend of the two worked out to a loss of about 16.1%, its worst calendar year since the 1930s by Morgan Stanley's accounting. The funeral notices were everywhere that winter. In 2023 the same mix rebounded roughly 17%, an approximate figure, and the mourners quietly filed out.

Chapter 3 already explained what actually happened: bonds hedge growth scares, not inflation shocks, and 2022 was the second kind of storm. The mix is neither dead nor sacred. Vanguard's January 2026 outlook went as far as preferring 40/60 to 60/40 on expected 10-year return per unit of risk, a published model view built from the same assumption ranges chapter 2 introduced, and worth reading as their forecast rather than anyone's command. The durable lesson is smaller: the pair of numbers is a dial. Chapter 2 taught you to set the dial by your worst-year line, not by tradition, and a 60/40 built from two index funds costs $30 a year per $100,000 either way.

Recipe four: the robo-advisor

The newest recipe is a subscription. Robo-advisors manage roughly $1.2 trillion as of 2026, typically for 0.25% a year, and the package is consistent across brands: an ETF portfolio matched to a questionnaire, automatic rebalancing, tax-loss harvesting in taxable accounts, and an interface designed to discourage tinkering. On $100,000 the fee is $250 a year on top of about $50 in ETF ingredients.

Be precise about what the $250 buys, because the ingredients and the rebalancing can be had for $36 and a calendar reminder. What the fee mostly buys is a wrapper around your own behavior: the autopilot, the friction before bad decisions, the absence of a sell button at the center of the screen. Whether that is expensive depends entirely on what you would do unattended. Chapter 6's evidence cuts both ways here: investors earned 7.0% a year in funds that returned 8.2% over the ten years to 2024, and that 1.2-point behavior gap dwarfs a 0.25% fee for anyone who would otherwise tinker.

What are you actually paying the assembler for?

Three separate services hide inside every assembly fee, and pricing them separately is the whole skill of this aisle. The first is assembly itself: choosing the splits. For you that service is now worth almost nothing, because chapters 1 through 6 already did it. The second is automation: rebalancing on schedule, gliding the mix as the date approaches. That is worth something real, and the target-date wrapper proves it can be bought for a handful of basis points in a cheap index series. The third is behavior: something or someone standing between you and the sell button in a year like 2022. That is the only service on this shelf that can honestly be worth hundreds of dollars a year, and only to the investor who genuinely lacks it. When a recipe costs more than $36, ask which of the three services the extra money buys. Sellers who answer "returns" have answered wrong on the evidence of every cost chapter in this library.

Here is the full stack on $100,000, with each row adding across: ingredients plus assembly equals the total.

Where people go wrong

Buying two recipes at once. A target-date fund already owns every market, so the S&P 500 fund parked beside it quietly doubles the same mega-cap bet and undoes the glide. The ETFs & Funds Guide's overlap audit exists for exactly this. One recipe per account.

Grading a recipe on one year. The 60/40 was declared dead after its 16.1% loss in 2022 and quietly rebounded roughly 17% in 2023. Anyone who graded it in December 2022 sold the dip in their own plan. You bought a decades-long mechanism, so judge it on the storm types from chapter 3, not the latest receipt.

Paying assembly prices for finished work. Once your six plan lines exist, a 1.00% fee to choose splits is a fee for work already done. Automation is worth basis points and behavior coaching may be worth more, but assembly itself is now your cheapest line item.

Confusing the wrapper with advice. A good human planner who handles taxes, insurance, and estate questions sells something real and separate. Price the planning as planning. The error is paying planning prices for a pie chart of index funds wearing a strategy name.

Key takeaways

• The shelf is four recipes built from the same 0.03–0.05% ingredients: the two- or three-fund (about $30–36 a year per $100,000, you assemble), the target-date fund ($4.8 trillion at end-2025, 0.27% average fee, CITs at 54% of assets), the 60/40, and the robo (roughly $1.2 trillion at a typical 0.25%).
• The three-fund convention is deliberately silent on splits; your chapter 2 and chapter 4 lines fill the silence at ingredient prices.
• The 60/40 lost about 16.1% in 2022 (stocks at 18.1% down, bonds at 13.0% down), its worst calendar year since the 1930s per Morgan Stanley, then rebounded roughly 17% in 2023; Vanguard's January 2026 outlook prefers 40/60 on forward model assumptions, which is their view to publish and your dial to set.
• Every assembly fee bundles three services: choosing splits (yours is done), automation (buyable for basis points), and behavior (the only one worth real money, and only if you lack it).
• The fee stack on $100,000 runs $30 to $300 across the four recipes, and a 1.60% advisor proposal prices the same ingredients at $1,600.

Sources: Bogleheads three-fund convention · Morningstar target-date research · Morningstar 150-year 60/40 stress test · Vanguard return forecasts (VCMM) · Morningstar Mind the Gap · Finvest ETFs & Funds Guide · Morgan Stanley research on the 2022 60/40, cited by name.