When paying off debt beats investing — the specific math

You have $10,000 sitting in checking, a credit card balance at 22%, a student loan at 6.5%, and a coworker who keeps telling you the S&P returns 10% a year so you would be an idiot not to invest. The coworker is doing bad math. Here is the actual math.

The crossover rate is not 10%

The pitch you hear is simple: stocks return roughly 10% a year historically, so if your debt costs less than 10%, invest the money instead of paying it down. This is the kind of sentence that sounds like reasoning and is actually just a slogan.

The honest comparison has three adjustments.

First, inflation. That 10% nominal S&P return is closer to 7% real once you subtract long-run inflation. Your debt interest rate is already nominal — 22% on a credit card is 22% you actually pay in dollars. Comparing nominal debt to nominal stock returns is fine as long as you know the stock number includes the inflation tailwind and your debt does not shrink with inflation the way, say, a fixed mortgage does on the principal.

Second, taxes. Investment gains in a taxable brokerage account get taxed. Long-term capital gains are 15% for most people, 20% at higher incomes, plus state tax. A 10% nominal return in a taxable account is closer to 8% after federal LTCG. Meanwhile, paying off a 7% loan gives you a guaranteed 7% return, tax-free, because you are not earning income — you are eliminating an expense. There is no line on your 1040 for "money I did not pay in interest."

Third, risk. The S&P's 10% average includes years like 2008 (-37%) and 2022 (-18%). Paying off debt returns exactly the interest rate, every time, with zero variance. A guaranteed 6.5% is not the same asset class as a hoped-for 8%. In finance terms, you should demand a risk premium to take the volatile bet. Two to three percentage points is a defensible premium.

Stack those adjustments and the real crossover looks like this: if your debt rate is above roughly 6%, paying it down beats investing in a taxable brokerage on a risk-adjusted basis for most people. Above 8%, it is not close. Credit card debt at 18-25%? Investing instead of paying that down is, mathematically, choosing to lose money.

The one exception worth naming: an employer 401(k) match. If your company matches 100% up to 5% of salary, that is an immediate 100% return, which beats any debt rate short of a payday loan. Capture the match, then attack the debt.

The tax angle cuts both ways

Some debt is tax-advantaged and the effective rate is lower than the sticker.

Mortgage interest is deductible if you itemize, but since the 2017 standard deduction went up to $14,600 for singles and $29,200 for married-filing-jointly in 2024, most people no longer itemize. If you take the standard deduction, your 6.5% mortgage costs you 6.5%. Full stop. The "mortgage interest is deductible so it is really only 4.5%" line is out of date for maybe 85% of homeowners.

Student loan interest gets a deduction up to $2,500 a year, but it phases out at around $80K single / $165K joint MAGI and it is an above-the-line deduction, meaning it saves you your marginal tax rate on that interest — so a 22% bracket earner paying $2,000 in student loan interest saves $440. That drops a 6.5% loan to roughly a 5.1% effective rate for that portion of interest. Not nothing, but not the game-changer people describe.

Credit card interest, personal loan interest, car loan interest: not deductible. What you see is what you pay.

So redo the crossover with tax adjustment: a 5.1% effective student loan is a closer call against a maxed Roth IRA (tax-free growth), where the crossover argument for investing gets stronger. A 22% credit card is still a fire. Put out the fire.

The psychology bonus is real and you should count it

Economists used to hate this argument because it is not on the spreadsheet. Then they ran the numbers and found that people who pay off debt in order of smallest balance first (the "snowball") stick with the plan longer than people who pay in order of highest interest rate first (the "avalanche"), even though avalanche is mathematically optimal.

The reason banks push minimum payments is that the interest compounds against you while your motivation degrades. A $6,000 credit card balance at 22% making minimum payments of 2% takes about 26 years to pay off and costs roughly $9,000 in interest. That is not a math problem. That is a behavior trap the product is designed to exploit.

If killing a $1,200 store card first — even though your $8,000 card is at a higher rate — is what keeps you in the game for the next 18 months, do that. The behavioral return on "I finished something" is worth 1-2 percentage points of interest rate optimization for most people.

Use this script when you call the higher-rate card to ask for a lower APR — because you should:

"Hi, I have been a cardholder since [year]. I am looking at my interest rate of [X]% and comparing it to offers I am getting from other cards. Before I move my balance, I wanted to see what rate you can offer me to keep my business. What can you do?"

That call takes 8 minutes. About one in three cardholders who ask get a rate reduction, per CFPB survey data. A 4-point drop on a $6,000 balance is $240 a year. That is a $1,800-per-hour phone call.

Where the $10K should actually go

Run the ladder. If you have any credit card debt above 15%, the entire $10K goes there — no emergency fund debate, no investing debate, that is the answer. If your only debt is a mortgage under 4% (the pre-2022 crowd), the $10K goes to investing or a high-yield savings account earning 4-5%. In between — student loans in the 5-7% range, car loans in the 7-9% range — you are in the crossover zone, and the tiebreaker is whether the debt is bothering you enough to affect your decisions.

Open your highest-rate debt account right now, write down the interest rate and balance, and multiply them. That number is what the debt costs you per year. If it is more than what $10,000 would earn in a HYSA (roughly $450 at current rates), you have your answer before dinner.