Good Debt vs. Bad Debt: A Practical Test

The standard good debt vs. bad debt framing says mortgages and student loans are good, credit cards and car loans are bad. It is a reasonable first approximation and it falls apart quickly on contact with real decisions.

A mortgage on a house you cannot afford in a market you had to stretch into is not good debt. A 0% card used deliberately to spread a necessary purchase over twelve months is not bad debt. The category was never the point — the terms and the purpose are.

Here is a test that works on any borrowing decision.

Question 1: What is the real cost, in total?

Not the monthly payment. The monthly payment is the number lenders lead with precisely because it makes large borrowing feel small.

Work out the total: monthly payment × number of payments, plus any arrangement fees.

A $25,000 car loan at 9% over seven years is $402 a month. Over 84 months that is $33,768 — you are paying $8,768 to borrow, on an asset that will be worth perhaps $8,000 when you finish paying.

The same car over three years is $795 a month and $28,620 total. Better arithmetic, harder monthly. And if $795 is impossible, that is meaningful information about whether this is the right car — the seven-year term is not making it affordable, it is making it look affordable.

Question 2: Does the thing you are buying appreciate, hold value, or vanish?

This is the closest thing to the traditional good/bad split, but it is a spectrum rather than two boxes.

Appreciating or income-producing. Property in a functioning market, education that measurably raises your earnings, equipment for a business that generates revenue. Borrowing here can genuinely make sense: the asset works while you repay.

Holds meaningful value. A reliable used car that gets you to work. It depreciates, but slowly, and it enables income. Reasonable to borrow modestly for.

Depreciates fast. A new car, furniture, electronics. Loses 20% the moment you own it. Borrowing means paying interest on something already worth less than you owe.

Vanishes entirely. Restaurants, holidays, clothes, events. Consumed on the spot. The debt outlives the experience — sometimes by years.

The last category is where "bad debt" genuinely earns the name. Not because enjoying things is wrong, but because paying 22% interest for eighteen months on a holiday you took in June is a straightforwardly poor trade you would not have accepted if it had been priced that way at the time.

Question 3: What is the rate, honestly?

Rates cluster into tiers, and the tier matters more than the label on the product.

Note that the tier does not respect the traditional categories. A "good debt" student loan at 11% deserves more urgency than a "bad debt" car loan at 4%.

Two adjustments to make it honest:

Check whether the rate is fixed or variable. A variable rate that is low today is a different product from a fixed one at the same rate. It may be fine; it is not the same.

Check whether interest is tax-deductible where you live. Mortgage interest and student loan interest sometimes are, which lowers the effective rate.

Question 4: What happens if things go wrong?

The most underweighted question, and often the one that actually decides whether a debt was a mistake.

Can you pause or reduce payments? Federal-style student loans commonly have deferment, forbearance, or income-driven repayment. Credit cards have essentially nothing. A private student loan may look identical to a public one on paper while offering none of the protections — this is a genuinely important distinction that is easy to miss at signing.

Is it secured, and against what? A secured debt can take the asset. A mortgage risks the house; a car loan risks the car. Unsecured debt is more expensive precisely because the lender has fewer options — which means it is less immediately dangerous to your home, though it can still end in court.

Does it survive bankruptcy? In many jurisdictions student loans are extremely difficult to discharge. That makes them structurally stickier than almost any other debt, regardless of the rate.

How would you cope with a 30% income drop? Run the number. If the payment only works at full current income, the debt is riskier than its rate suggests.

Applying it: three cases

A mortgage at 4.5%, payment 28% of take-home, house you plan to stay in for a decade. Low rate, asset that plausibly holds value, payment with slack, secured against something you also need. Good debt by every measure.

The same mortgage at 42% of take-home in a market you stretched into. Same rate, same asset, entirely different debt. There is no slack for a bad year, and the security means the downside is losing your home. The rate did not change; the risk did. This is why the four questions beat the categories.

A $6,000 credit card balance at 23% funding ordinary living costs. High rate, nothing to show for it, no protections. This is not a moral failing — it is usually a symptom of income not covering costs — but it is the highest-priority debt in almost any portfolio. See avalanche versus snowball.

Two rules that survive most situations

Never borrow for something consumed faster than you repay it. If the holiday, the dinner, or the outfit will be finished long before the debt is, the trade was bad at the moment you made it.

Borrow at a rate below what the money can reliably earn or save you. A 4% mortgage while investing at an expected 7% is a defensible use of leverage. A 23% card while investing anything at all is not — you are borrowing expensively to invest cheaply, which is the wrong side of the saving vs. investing trade in every case.

A better test than good debt vs. bad debt

Stop asking whether a debt is good or bad. Ask: is this the cheapest way to get something genuinely worth having, on terms I can survive a bad year on?

That question handles a 0% purchase plan (often yes), a stretched mortgage (often no), a modest used-car loan (usually yes), and a holiday on a credit card (essentially never) — without needing any categories at all.