The latte factor is the most-cited and most-criticized idea in personal finance. Both reactions are correct at the same time, which is a sign that the idea is doing something interesting and also something wrong. Worth breaking down the two parts.
The original pitch, and why it stuck
David Bach's version, published in a series of books starting in the late 1990s, is clean and quotable. Skip a $5 daily coffee. Invest the savings. Compound at historical market returns. Retire a millionaire.
The math holds up. $5 a day is $1,825 a year. Compounding at the S&P 500's roughly 10% average since 1957[1]Based on historical returns. Past performance doesn't predict future results. over a 40-year working career produces a result around $900,000 — not quite a million, but close enough that the pitch rounds up honestly. Over 30 years, closer to $330,000.
The reason the framing stuck: it gives the reader an actionable, small, repeatable decision with a large asymptotic payoff. "Save 15% of your income" is abstract. "Skip this one specific purchase" is concrete. Concrete beats abstract for getting people to take an action, which is why the latte has outlived every piece of financial advice that out-competes it on rigor.
Why the critics are right
The critique, which is correct as far as it goes:
It confuses savings with restriction. The math the latte factor uses doesn't depend on quitting coffee. It depends on $1,825 a year going into an invested account. If you have that $1,825 in your budget already — in the form of underspending, side-income, or any other source — the math works without any lifestyle change. Bach's framing reads the dependency the wrong way. Most people who could plausibly save $1,825 a year could also not save $1,825 a year for reasons that have nothing to do with coffee.
It focuses on the wrong line item. $5 a day is $1,825 a year. A household that saves an extra 3% on housing (smaller apartment, refinanced mortgage, longer commute) clears $3,000-$5,000 a year without a single lifestyle change to food. A household that starts contributing to a 401(k) with a match clears even more — and adds a tax benefit the latte version doesn't capture. The coffee question is optimized for relatability; it's suboptimized for impact.
It assumes behavioral compliance that empirically doesn't happen. The chain of steps required for the latte factor to produce the claimed result — (1) actually skip the coffee, (2) transfer the saved dollars to a brokerage account on a regular schedule, (3) invest them in a diversified equity product, (4) not touch them for decades — has low compliance rates at every step. The fraction of would-be latte-factor adherents who produce the $330,000 outcome is a rounding error.
The nominal-vs-real problem. $330,000 in 2056 dollars, assuming 3% inflation, has the purchasing power of about $135,000 today. The latte-factor claims typically run in nominal dollars, which makes the outcome sound bigger than it is in real terms.
The Helaine Olen book Pound Foolish makes the fuller case. The critique is load-bearing and should chasten anyone inclined to sell the latte factor as universal advice.
Why the critics are also missing something
Despite all that, the latte factor gets at a real thing. Small, regular contributions compound. Daily purchasing decisions do have long-run financial consequences. The scale of the impact is smaller than the usual framing claims, but the direction is right.
The specific defensible claim, stripped of the pitch: consistent small contributions beat inconsistent large ones over long horizons, and everyday spending is a reliable source of consistency.
This is DCA with everyday spending — the claim that a small-dollar, high-frequency contribution pattern captures time-in-market more reliably than a larger-dollar, lower-frequency one. The compounding is identical; the empirical savings rate is higher, because the rhythm replaces the willpower.
What the latte factor correctly identified: the daily purchase is a behavioral trigger. People who have consistent spending patterns could have consistent saving patterns if they piggyback on the same triggers.
What it got wrong: the way to piggyback isn't to refuse the purchase. It's to pair the purchase with an ownership event.
The spend-to-own version
Rewrite the latte factor without the restriction:
You buy the coffee. The coffee is worth $5 to you, or you wouldn't buy it. On top of the coffee, 3% of the transaction buys a fractional share of the coffee company. The $5 becomes $5 of coffee and 15 cents of Starbucks stock, on every cup.
This version drops the "skip the coffee" step entirely. It also drops the "remember to transfer to a brokerage" step. The transaction itself becomes both the consumption event and the investment event.
The 30-year math is smaller, because the contribution per purchase is smaller (15 cents vs. $5). But it compounds at roughly the same rate, and the empirical compliance rate is much higher, because nothing about existing behavior had to change.
Run the two scenarios side-by-side in the calculator. Scenario A: skip $5/day of coffee spending, invest that $5/day in an index fund. Scenario B: keep buying the coffee, route 3% of each purchase into SBUX. Scenario A has the bigger endpoint number if you actually do it. Scenario B is the one you'll actually do.
The 10-year treatment of Scenario B is worked out at the SBUX what-if — small, steady contributions into Starbucks compounded at roughly 7% over the past decade, below the S&P. The MCD version of the same drip compounded closer to the index. The takeaway isn't that either is the "right" ticker; it's that the latte-factor math is always contingent on the company you're latte-factoring into.
The third scenario — and the one spend-to-own points at — is running that 3% across every purchase, not just the coffee. See the spend-to-own guide for the full mechanic, and the coffee money calculator for the Starbucks-specific version of the math.
Where the honest tradeoffs live
Spend-to-own fixes the behavioral failure mode of the latte factor. It does not fix the other critiques.
It's still small-dollar optimized. 3% of coffee purchases isn't the highest-impact financial move available to most households. Filling up a 401(k) match, paying down high-interest debt, building an emergency fund — all higher priority. Spend-to-own is a complement to those, not a substitute.
It's still equity-exposed. The contributions are in stock, not cash. When equities drop, the contributions drop with them. The 30-year numbers look great on paper; year 3 of a bear market is where people bail. The behavioral advantage over the latte factor is real, but not unlimited.
It's still subject to concentration risk in the underlying tickers. If you happen to shop at a handful of companies that underperform, your spend-to-own portfolio inherits that underperformance. Spread across a broader wallet reduces but doesn't eliminate this. The comparison to cashback (cash today vs. equity later) is worked through in cashback vs. stock rewards.
The time horizon is still long. The impressive numbers in latte-factor math come from 20-to-40-year horizons. Over 5 years, spend-to-own produces a small-to-modest balance; the compounding only starts to feel large after a decade. People who bail at year 3 don't see any of it.
What to take away
The latte factor isn't wrong. It's misfiled. The version that works isn't "quit coffee"; it's "when you buy things, also own a piece of what you're buying." The difference is that the second version doesn't require giving anything up. It runs on the spending you were already doing and quietly turns it into a savings pattern.
That's the version spend-to-own implements. It's not a millionaire-in-a-decade pitch; it's a real-compliance, small-dollar, long-horizon compounding mechanic. The math is more modest than the original latte claim. The empirical outcomes are better, because the mechanic actually runs.
Next steps
Model the two versions for your actual spending in the Slyce calculator. Compare "skip and invest the difference" to "buy and slyce a percentage." The gap between the idealized math and the empirical math is the whole case for automated spend-to-own as the successor framework.
More calculator scenarios
Spoke
529 plan calculator: project college costs in 2026
Project a 529 plan balance to age 18, see what state-deduction adds back, and compare a 529 to a taxable UTMA at the same return assumption.
Explainer
Cashback vs. stock rewards: which is actually worth more
A 2% cashback card gives you cash today. A 3% stock-rewards card gives you equity that moves with the company. Here's the full math.
Explainer
The coffee money investing calculator
If you spend $5 a day on coffee, you're not throwing it away — you're funding Starbucks's earnings. The coffee money calculator shows what it compounds to if you slyce it too.
Explainer
Dollar-cost averaging with your everyday spending
Dollar-cost averaging is easy to explain and hard to do. Routing it through your everyday spending makes the habit self-executing.
Frequently asked
- Who invented the latte factor?
- David Bach, in a series of personal-finance books starting in the late 1990s. The original version used a hypothetical $5 daily coffee to illustrate compound interest over a career-long horizon. The idea predates Bach's framing — the general 'small habits compound' argument shows up in financial writing going back decades — but Bach's latte example is the one that stuck.
- Is the latte factor real?
- The math is real. $5 a day invested for 30 years at 10% compounds to around $330,000. The behavioral claim is the weaker part — that quitting coffee is a reliable path to building that habit. In practice, most people who cut $5-a-day expenses don't end up with an extra $1,825 a year in their brokerage account. They end up with the same savings rate and a slightly less pleasant morning.
- Why does the latte factor get criticized?
- Two main reasons. First, it conflates 'savings math' with 'spending restriction,' which implies the former requires the latter — untrue for most people who could afford to save more without cutting coffee. Second, it focuses attention on small-dollar items while ignoring larger drivers of savings rate (housing, transportation, retirement-account contributions). The Helaine Olen critique in Pound Foolish makes both points at length.
- What's a better version of the latte factor idea?
- Automated, transactional, and agnostic to what you buy. Any framework that relies on 'spend less to save more' runs into willpower. A framework that says 'when you spend, a slyce of it goes to ownership' removes the willpower step. The $5 coffee purchase still happens; 15 cents of it becomes SBUX stock on the back-end. The math lands in a similar place without the lifestyle-change requirement.
- How much is $5 a day invested over 30 years?
- About $330,000 at a 10% annual return. $1,825 per year contributed × 30 years, compounding annually. The actual result depends heavily on the return you earn and the sequence of annual returns. At 7% annual (closer to the S&P's real return after inflation), the same contribution compounds to about $184,000. Nominal numbers are bigger; real numbers are smaller; both are bigger than most people would guess.
- Does the latte factor still work in today's economy?
- The compounding math works regardless of the economy. What changes is the plausibility of the savings rate — $5 a day was a meaningful fraction of discretionary budget when Bach wrote the book; it's a smaller fraction today for most middle-class households. The same idea scaled appropriately ($15 a day, or 3% of everything you spend) is structurally identical and matches modern spending better.
Keep reading
Slyce Editorial
Published Apr 14, 2026 · Updated Apr 14, 2026