What Is Compound Interest and How It Works

Why does my money crawl, then suddenly race – or my debt refuse to die?

Early compounding feels broken. A small balance earns a small dollar amount; the “interest on interest” slice is almost invisible. Later, the same rate looks unfair – in your favor on savings, against you on revolving debt. That lived whiplash is the real question behind what is compound interest and how does it work, not a textbook definition.

Same engine, two directions. On a card, interest is typically added daily on the average daily balance (as of common U.S. issuer practice described by Experian and consumer-finance explainers – this may have changed for a specific product, so read your agreement). Minimum payments often fail to shrink principal fast enough, so new interest keeps capitalizing and the balance stalls or climbs.

Where the usual explanations fall short

Most writeups recite “interest on interest,” paste A = P(1+r/n)^(nt), run a pretty year-by-year table, name-drop the Rule of 72, and say “start early.” Fine as far as it goes. Weak on the parts that change decisions: frequency → effective annual rate, tax and inflation drag, daily credit-card mechanics, and how to stress-test your own inputs.

Think of it like a snowball rolling downhill. Near the top the ball is tiny and picks up little mass each turn. Mid-slope, every rotation grabs more snow because the surface is already large. Tutorials show the finished snowball. They skip the slope angle – and the sticky patches (fees, rate resets, taxes) that slow or reverse it.

The actual mechanics: formula, frequency, and the effective rate gap

Frequency is the gap most glossaries bury. Two products both labeled “10%” are not equal once you annualize compounding. Per Investopedia’s effective-rate examples, a 10% nominal becomes roughly 10.250% EAR semiannual, 10.381% quarterly, 10.471% monthly, and 10.516% daily. APY folds that effect in for deposits; APR is still the usual borrowing sticker (fees are treated differently – compare definitions, not vibes).

The mechanism itself is simple: each period the base resets so prior interest also earns (or costs) interest. Standard form:

A = P * (1 + r/n) ** (n * t)

# A = final amount
# P = starting principal
# r = annual nominal rate (decimal, so 5% = 0.05)
# n = compounding periods per year (1=annual, 12=monthly, 365=daily)
# t = time in years

Interest earned = A – P. Clean check from Investopedia’s worked numbers: $10,000 at 5% compounded annually for 3 years → $1,576.25 interest. Simple interest on the same terms stops at $1,500. For a longer official classroom arc, SEC Investor.gov walks $100 at 5% to $105 after year one, $110.25 after year two, roughly $162+ by year ten, and almost $340 by year twenty-five with no extra deposits.

Continuous compounding is the ceiling: A = P × e^(r t), e ≈ 2.71828 – the limit as n → ∞. Jacob Bernoulli hit e in 1683 by asking what infinite compounding does to interest. Banks never quite run pure continuous; daily sits close. Formula background also tracks on Wikipedia’s compound interest page.

Pro tip: Convert to EAR or APY before you compare offers. A slightly lower nominal with daily compounding can beat a higher nominal that compounds once a year.

Rule of 72: years to double ≈ 72 ÷ rate(%). At 8%, call it ~9 years. Approximation – drifts at extreme rates, assumes steady compounding, and ignores contributions, taxes, and fees. A version shows up in Luca Pacioli’s 1494 Summa de arithmetica (via St. Louis Fed / history notes).

When the same math works against you

Credit cards: daily compound on average daily balance is the norm called out by Experian’s explainer (verify your card – terms differ and can change). Unpaid interest joins principal; tomorrow’s charge runs on a bigger number. Minimum-only payoff schedules often crawl because interest keeps refilling what you chipped off.

Savings-side friction gets skipped in glossy charts too. Outside tax-sheltered accounts (IRA, 401(k), and similar – rules depend on jurisdiction and may have changed), interest and realized gains are generally taxable. Inflation eats purchasing power; longer-run Fed communications have often centered a 2% inflation target (as of recent policy frameworks – this can shift). A nominal 5% after tax and after ~2% inflation is a thin real residual, not the poster-chart curve.

Model your own numbers instead of trusting charts

Use the free SEC Investor.gov compound interest calculator. Principal, monthly add/withdraw, rate, years, frequency. Then break it on purpose: cut the rate 1-2 points for fees or weaker returns, apply a tax haircut, subtract an inflation assumption.

Or three lines of Python (or have an LLM write the loop and explain each term):

principal = 10000
rate = 0.05
n = 12 # monthly
years = 10
A = principal * (1 + rate/n) ** (n * years)
print(round(A, 2), round(A - principal, 2))

Flip n between 1 and 365 and watch the gap. Add a monthly contribution loop when you want something closer to real life. Sensitivity shows up faster in a script than in a static table.

One open question worth sitting with: if your expected real after-tax return is only a couple of percent above inflation, how many years of contributions does it take before compounding actually overtakes new deposits? Personal math. Almost never printed in intro pieces.

Next action: open the Investor.gov calculator, enter a starting balance and monthly amount you could actually save, set monthly compounding, and run best-case vs 2 points lower. Screenshot the gap. That pair of numbers beats another abstract lecture.

FAQ

Is compound interest the same as APY?

No. Compound interest is the mechanism. APY is the yearly yield figure that already includes compounding frequency so deposit products can be compared directly.

Does the Rule of 72 work for debt too?

Roughly. 72 ÷ rate ≈ years for an unpaid balance to double with no payments and a fixed rate. On a 24% card that’s about three years – enough to reorder priorities. Still an approximation; fees and rate changes are outside the shortcut.

Why do early years of compounding feel so weak?

Growth is multiplicative on a base that starts near pure principal. The interest-on-interest slice stays small until several cycles stack. Regular contributions change the story: each deposit starts its own compounding clock immediately. A late lump sum never gets those missed early multiplications back – even if the headline rate looks identical.