How debt really works

Since 2010, every US credit card statement has been required to carry a small box showing how long the balance would take to clear at the minimum payment, and what it would cost. Card companies fought the requirement. Once you work through the arithmetic in this lesson, you'll see why: on a typical $5,000 balance, the honest answer is measured in decades, and the interest exceeds the debt.

This is the lesson where the compounding you'll meet in lesson 4 shows up on the wrong side of the table. Your lesson 2 budget has a debt line and a savings line; this lesson is about the debt line, and it ends by answering the question lesson 2 deferred: when money is short, does the next dollar go to the cushion, the card, or somewhere else?

What a balance actually costs

APR means annual percentage rate: the yearly price of borrowing, as a percentage of what you owe. For a quick monthly figure, divide by 12. A card at 24% APR costs about 2% a month, so a $2,000 balance costs about $40 a month just to hold, before a cent of it goes away.

The real mechanics are slightly worse than that, and worth one paragraph because they change behaviour. Card interest is charged daily, not monthly: the issuer divides your APR by 365 and applies that daily rate to your average daily balance, so interest starts accruing on interest within the month (CFPB explains the calculation).¹ And most cards have a grace period, meaning purchases charge no interest if you pay the full statement balance by the due date, but carrying any balance usually forfeits it, so new purchases start accruing interest immediately. Pay in full and the card is an interest-free convenience. Carry a balance and everything you buy costs extra from day one. There's no middle setting.

The minimum payment, worked through

Here is the worked example this lesson is really about. Say you owe $5,000 on a card at 22% APR, close to the current US average for accounts paying interest, which was 22.15% in mid 2026 (Federal Reserve G.19).² A common minimum payment formula is that month's interest plus 1% of the balance, with a floor of $25.

First month: interest is $5,000 × 22% ÷ 12, about $92. One percent of the balance is $50. So the minimum is about $142, and of that $142, only $50 touches the debt. You paid $142 and you owe $4,950.

Watch what happens next. Because the minimum is a percentage of the balance, it shrinks as the balance shrinks. The payment falls with the debt, the debt therefore falls slowly, and the schedule stretches.

Now the good news, which is just the same arithmetic pointed the other way. Every dollar above the minimum goes entirely to the balance, and a smaller balance means less interest next month, which frees more of the next payment for the balance, and so on. Fifty extra dollars a month turns 19 years into five and a half, and cuts the interest from $8,100 to $2,830. Look at the gap between the two lines below; that gap is what $50 a month buys.

This is why revolving credit card debt is an emergency in a way other debts aren't. It's not moral failure; it's arithmetic. At 22%, the debt grows faster than almost anything you could plausibly earn on savings or investments, so while it exists it silently outruns everything else you do with money. Consumer advocates argue the minimum payment structure is designed to maximise interest collected; whatever the intent, the schedule above is what it produces, and your own statement's disclosure box will confirm it (the CARD Act requires this disclosure).³

Secured and unsecured: the distinction that organises everything

Debts come in two kinds, and the difference explains most of the table below.

A secured debt is pinned to a thing the lender can take: a mortgage to the house, a car loan to the car. Because the lender has that safety net, the rate is lower. An unsecured debt (credit cards, most personal loans, medical bills, payday loans) has no collateral, so the lender prices in the risk, and the rate is higher.

Here's roughly what borrowing costs in the US as of late 2026. Rates move; the point is the pattern, not the digits.

The secured column matters twice. First, for priority when things go wrong: if you ever can't pay everything, the debts attached to your housing and your transport are the ones whose consequences arrive fastest and hardest. Second, for a trap that arrives dressed as help.

Paying off several debts: the real avalanche versus snowball question

Suppose the budget from lesson 2 frees up real money each month and you have several debts. Everyone agrees on the frame: pay every minimum, then concentrate everything extra on one target debt until it's gone. The argument is over which target.

Avalanche says target the highest APR first. The case is pure arithmetic: every dollar aimed at the most expensive debt cancels more interest than a dollar aimed anywhere else, so this order minimises total interest paid. No one disputes the math.

Snowball, the order Dave Ramsey's programme popularised, says target the smallest balance first, whatever its rate. The case is behavioural, and it isn't hand-waving: it has data. David Gal and Blakeley McShane studied thousands of clients of a debt settlement firm and found that the ones who closed individual accounts entirely, rather than spreading payments, were more likely to eliminate their whole debt (Gal and McShane, 2012).⁸ Remi Trudel and colleagues reached the same conclusion experimentally: concentrating repayments and finishing off accounts increased how hard people worked at repayment, because visible progress on a specific debt sustains the effort (Kettle, Trudel, Blanchard and Häubl, 2016).⁹

So the honest summary is: avalanche costs less if you finish, and snowball makes finishing more likely. What would settle it is a trial randomly assigning real debtors to each plan and measuring both completion rates and total interest; the evidence we have is observational and experimental proxies, which is why this stays a judgement call rather than a solved problem. The judgement is about you, not about math: if you're confident you'll grind through a multi-year plan, avalanche is cheaper; if you've started and abandoned plans before, the evidence says buy the motivation.

And often the price of that motivation is small. Watch it on a concrete case. Sam has a $1,000 card at 17% (minimum $25) and a $4,000 card at 26% (minimum $90), and $400 a month for both. Avalanche targets the $4,000 card and clears everything in 15 months for $781 of interest, closing the first account at month 13. Snowball targets the $1,000 card, also finishes in 15 months, costs $873, and closes the first account at month 4. The snowball premium is $92, and what it buys is a dead account eleven months sooner, one less minimum to juggle, and proof the plan works. Run your own numbers before deciding; sometimes the gap is $92, and sometimes, with big rate differences and big balances, it's thousands, at which point the motivation gets expensive.

The question lesson 2 deferred: cushion, card, or match?

Your budget frees $400 a month. You have no emergency cushion, a card balance, and maybe an employer offering a 401(k) match. Where do the dollars go first? Lesson 2 promised you the positions; here they are.

The one near-consensus answer: claim the match first. If your employer matches retirement contributions, each matched dollar is an immediate 100% return (or 50%, at fifty cents on the dollar), which outruns even payday-loan interest. Very few planners of any school argue with taking free money, and unclaimed match usually just evaporates. This is also, note, a US-specific piece of machinery; other countries have their own versions or none.

After that, it's genuinely contested. Ramsey's Baby Steps say: $1,000 cushion, then every spare dollar at the debts (snowball order) before building the full emergency fund. The argument is momentum and focus, and millions have finished the programme. The arithmetic school answers: with a card at 24%, every month a dollar sits in a savings account earning 4% instead of killing card debt, you pay a 20% spread for the privilege, so clear the card first. And the behavioural-resilience school answers back: without a cushion, the next flat tyre goes straight onto the card you just paid down, and Vanguard's research found even about $2,000 of cushion is associated with markedly higher financial well-being (lesson 2 covered the evidence and its correlational limits).

There's no settled ranking past the match, and the honest framing is the same as avalanche versus snowball: pick by failure mode. If your spending is stable and emergencies are rare in your life, minimising interest argues for the card. If surprise expenses keep landing on your card, a small cushion first breaks the loop, and its cost (some months of extra card interest) is the price of not restarting the debt every time life happens.

Credit scores: the machinery, and the myth that costs money

In the US, whether and how cheaply you can borrow runs through credit reports (your history, kept by three private bureaus) and credit scores (a number computed from a report). If you're outside the US, your country's system differs and this section is skippable; the rest of the lesson isn't.

The mechanics that matter, all verifiable at the CFPB's credit hub:¹⁰

• Payment history is the biggest input. One 30-day-late payment can sit on a report for seven years. Automating minimums (lesson 2's automation logic, applied to debt) protects this even in chaotic months.
• Utilisation is second: the share of your card limits you're using, as reported on your statements. Lower reads as better. This is where the myth lives, so let's kill it properly.
• Checking your own score or report is a "soft pull" and doesn't affect anything. Applying for new credit is a "hard pull" and dings the score modestly; many applications in a short window ding it more.
• Your income is not in your credit score. Lenders consider income separately when deciding what you can afford; the score only describes how you've handled borrowing.

You're entitled to free copies of your actual reports from all three bureaus at AnnualCreditReport.com. That exact site is the one federally mandated free source; a swarm of similarly named sites exists to sell you subscriptions. Checking your report roughly yearly is also your early-warning system for identity theft: accounts you don't recognise mean someone borrowed in your name, and IdentityTheft.gov, the FTC's recovery site, is the place to start if you find one. The same wariness applies to debt itself: if a collector contacts you about a debt you don't recognise, verify it before paying anything; the CFPB's debt collection pages explain your rights, including making collectors prove the debt is yours.¹¹

When borrowing makes sense

After a lesson about 22% card interest, it would be easy to conclude debt is simply bad. That's not the lesson, and a course that moralised at you about it would be teaching poorly.

Borrowing is a price paid to move a purchase earlier in time. Sometimes that trade is clearly worth it: almost no one can buy a house from savings, and a mortgage at 6.7%, secured, long-term, and buying a place to live plus a stake in an asset, is a different object from a card balance at 24% that bought things already gone. Student debt sits in between and resists slogans: borrowing $30,000 for a degree with strong, checkable earnings prospects has historically been one of the better trades available, and six figures for a program with weak job outcomes has trapped a lot of people; the honest advice is to run the numbers on the specific degree, not on "education" in the abstract. Whether buying beats renting is its own genuinely unsettled question with a framework this course takes up separately; a mortgage being cheap debt doesn't settle it.

The useful habit isn't sorting debt into good and bad. It's asking three questions before borrowing: what's the APR, is it secured and by what, and will what I'm buying still matter when I'm still paying for it?

What people get wrong

"Minimum payments keep me in good shape." They keep you in good standing, which is a different thing: no late fees, no credit damage, and, as you computed above, 19 years of interest on a debt you could clear in five. The minimum is the floor that protects the lender's relationship with you, not a recommendation.

"Carrying a balance builds credit." Covered above, and worth repeating because surveys keep finding most cardholders believe some version of it. Paying in full shows the bureaus the same behaviour and costs nothing.

"A lower interest rate always means a better loan." Rate is one dimension. A home equity consolidation at 9% can be a worse position than cards at 24% if there's real risk you can't pay, because of what's now securing it. Compare worst cases, not just rates.

"Debt payoff order is a solved problem." Neither the avalanche purists nor the snowball loyalists are giving you the full picture. The math favours avalanche; the completion evidence favours snowball; which matters more depends on which way you tend to fail. Distrust anyone who tells you one order is right for everyone.

Practice

Connections

This lesson is lesson 2's debt line, opened up. The budget tells you how many dollars you have to aim; this lesson told you where to aim them, and it answered the cushion-versus-debt question lesson 2 deferred (match first; after that, it's a judgement call this lesson equips you to make). Lesson 4 takes the same compounding arithmetic you just watched work against you on the $5,000 card and turns it around: the identical mechanism, running on savings instead of debt, is where the money your payoff plan frees up eventually goes. Lesson 5 is what to do with it once it compounds.

Go deeper

• CFPB, credit reports and scores hub: the official, readable reference for everything in the credit section, including dispute procedures when a report is wrong.
• Khan Academy, Loans and Debt: the debt unit of the free course; a second pass over this ground with videos, plus material on loan types we didn't cover.
• Gal and McShane, "Can Small Victories Help Win the War?" (2012): the study behind the snowball evidence; readable methods, and a good example of behavioural research on real financial data.
• Kapoor, Dlabay and Hughes, Personal Finance (McGraw Hill), chapters 6 and 7: the standard college-textbook treatment of consumer credit and its cost, with the full legal detail on credit protections.

Sources

1. Consumer Financial Protection Bureau, "How do credit card companies calculate interest charges?". Daily periodic rate applied to average daily balance; grace period forfeited when a balance is carried.
2. Federal Reserve Board, Consumer Credit G.19 release. Average APR on credit card accounts assessed interest: 22.15% (Q2 2026).
3. Consumer Financial Protection Bureau, "What is a minimum payment on a credit card?". The CARD Act of 2009 requires statements to disclose minimum-payment payoff time and cost.
4. Consumer Financial Protection Bureau, "What is a payday loan?". A typical $15-per-$100 two-week payday loan works out to 391% APR.
5. Experian, State of the Automotive Finance Market, Q1 2026. Average auto loan rates: 6.39% new, 11.43% used.
6. Federal Student Aid, Interest rates for federal student loans. 2026-27 rates: 6.52% undergraduate, 8.07% graduate, 9.07% PLUS.
7. Freddie Mac, Primary Mortgage Market Survey. 30-year fixed averaging about 6.7% as of September 2026.
8. Gal, D. and McShane, B. B., "Can Small Victories Help Win the War? Evidence from Consumer Debt Management", Journal of Marketing Research 49(4) (2012). Debt-settlement clients who closed individual accounts entirely were more likely to eliminate their whole debt.
9. Kettle, K. L., Trudel, R., Blanchard, S. J. and Häubl, G., "Repayment Concentration and Consumer Motivation to Get Out of Debt", Journal of Consumer Research 43(3), 460-477 (2016). Field study and experiments: concentrating repayments on one account increases motivation and repayment effort.
10. Consumer Financial Protection Bureau, Credit reports and scores. Score factors, soft versus hard inquiries, and the carrying-a-balance myth; free reports via AnnualCreditReport.com.
11. Consumer Financial Protection Bureau, Debt collection, and FTC, IdentityTheft.gov. Collector-verification rights and identity-theft recovery.

Minimum-payment schedules, the two-line chart, and Sam's avalanche/snowball comparison were computed for this lesson month by month at the stated assumptions (minimum = interest plus 1% of balance with a $25 floor; interest = APR ÷ 12 on the running balance; no new charges).