The Confidence Tax

Math, Economics, and Personal Finance

The most common question I get from students who find me on social media isn’t about inflation, or what the Fed is doing, or what my favorite economic model is. It’s this:

Do I have to take math if I want to study economics? I’m not good at math.

I used to take so much time answering that question and adding a lot of nuance. Now, it just makes me feel frustrated. It is not a bad question, but it tells me about the person asking it and the environment they were exposed to before they’ve ever set foot into my classroom. By the time someone sends me that question, they’ve usually already made the decision that economics isn’t for them.

Here’s what I learned through my research, even though I’d been watching it happen in my classroom for years. For a huge share of students, especially women, whether they’re “good at math” was never actually what mattered.

Whether they believe they’re good at math is the determinant. The gap between perception and reality is costing people money, careers, and confidence they never needed to lose in the first place.

What the data actually says

My co-authors Whitney Buser, Darshak Patel, and I set out to test something that sounds almost too simple: does believing you’re good at math predict your financial literacy separately from actually being good at math?

We surveyed 529 college students across three very different schools. We sampled students from a large state university, a regional university, and a small liberal arts college. We gave them two things:

• an objective 10-question math assessment built from SAT/ACT-style content,

• five-question financial literacy test

Then we also asked them to self-rate their math confidence on a simple scale, using strongly agree to strongly disagree.

Then we examined the differences by gender, because the research had already established that men tend to outscore women on financial literacy tests, and our goal was to understand why.

If you want the full methodology and regression tables, the study is:

Al-Bahrani, A., Buser, W., & Patel, D. (2020). Early causes of financial disquiet and the gender gap in financial literacy: Evidence from college students in the Southeastern United States. Journal of Family and Economic Issues, 41(3), 558–571.)

For men, the answer was as expected: actual math ability predicted financial literacy. Their confidence didn’t change the outcome once we controlled for real ability. A man who thought he was great at math but tested below average scored no better than a man who correctly rated himself as below average.

For women, the entire relationship flipped.

Actual math ability, the objective test score, dropped out as a predictor entirely. What predicted a woman’s financial literacy score was whether she believed she was good at math, regardless of whether the test backed that belief up.

We split students into four groups:

• high ability/high confidence

• high ability/low confidence

• low ability/high confidence

• low ability/low confidence

A woman with high ability but low confidence performed statistically the same as a woman with low ability and low confidence. Her actual math skills did not help if she didn’t believe in them. Meanwhile, a woman with low ability but high confidence outperformed her low-confidence but equally-skilled peers.

We left thinking ability, without belief, was worth almost nothing for the women in our sample. But why?

The mechanism: you’re not bad at math, you’re mispricing it

The economic model is built on investment theory: people invest in financial knowledge up to the point where the expected benefit equals the expected cost. But nobody actually knows their true cost of learning something before they try. They have to estimate it using whatever reference point is closest at hand.

For most people studying economics and finance, math is that reference point. It’s the thing they were graded on for twelve straight years. So when you ask yourself, unconsciously, “how hard would it be for me to learn about compound interest, or bond prices, or how mortgages actually work,” you’re referencing your math ability as a proxy of your ability.

You’re pricing that cost using your stored-up sense of “am I a math person.” If that internal price tag is set too high, because you were told at 14 that you weren’t a numbers kid, you will rationally underinvest in learning the financial concepts that price tag is attached to. Sadly, even though the true cost of learning them was never actually that high. You’re not bad at math; you were convinced that you aren't good at math.

Math is for Boys

For girls, that belief has been set before they even reach middle school.

Cvencek, D., Meltzoff, A. N., & Greenwald, A. G. (2011). Math–gender stereotypes in elementary school children. Child Development, 82(3), 766–779

Meltzoff and Greenwald (2011) tested American children as young as six and found that by second grade, kids had already absorbed the cultural stereotype that math is “for boys” and that boys identified with math more strongly than girls did.

Once again, this wasn’t a finding about actual ability. They found no accompanying performance gap to justify it. It was a belief, installed early. By the time a 14-year-old girl is told, or simply absorbs, that she “isn’t a numbers kid,” the groundwork for that message was laid in elementary school, in classrooms, dinner tables, and offhand comments that had nothing to do with her actual aptitude. The 14- year-old isn’t pricing her cost of learning finance off her own track record. She’s pricing it off a stereotype she inherited before she was old enough to test it.

We ran our study on college students specifically because we wanted to catch people before real financial life started. Before mortgages, before retirement accounts, before anyone had specialized into “the one who handles the money” in a relationship. The gap was already fully formed. That means we can exclude the learning-by-doing hypothesis of why men usually manage household finance. This gap is coming from something that happened earlier — in school, at home, in culture — long before anyone opened a checking account.

Why math confidence matters for the study of economics and personal finance

Here’s why I get frustrated. If confidence only predicted how someone scores on a quiz, I’d file it under “interesting” and get back to teaching supply and demand. But a separate, much larger study suggests the effect doesn’t stop at the test; it follows people into real financial decisions they make for the rest of their lives.

Allgood, S., & Walstad, W. B. (2016). The effects of perceived and actual financial literacy on financial behaviors. Economic Inquiry, 54(1), 675–697

Allgood and Walstad (2016) built a combined measure of financial literacy that included an actual test score plus a self-rating of overall financial literacy, and tested it against a national sample of 28,146 U.S. adults and households. They tracked 22 real financial behaviors across five topics: credit cards, investments, loans, insurance, and financial advice. Do you carry a revolving balance? Do you shop around before buying insurance? Do you seek out financial advice, or avoid the conversation entirely?

They found that perceived financial literacy predicts real-world behaviors about as strongly as actual financial literacy did. Confidence in your own financial knowledge shaped what you actually did with your money almost as much as your knowledge itself did.

Put that next to what my co-authors and I found in college students, and you get a full confidence-and-behavior pipeline. The confidence gap forms early. It impacts what you major in and what you invest in learning. You carry that into adulthood, and it shapes whether you negotiate your loan terms, whether you diversify your investments, whether you ask a professional for help instead of guessing. It turns out that how you feel about your own competence matters.

This is also, I’d argue, the real reason the “do I have to be good at math to study economics” question matters so much to me. It’s the first data point in a pattern that, if the research holds, will keep repeating for that student’s entire economic and financial life. Keeping someone out of an economics classroom over a mispriced belief about their own math ability will cost society access to that student and the impact they will have on policy and research.

The student in my inbox

Every time I get the “do I have to be good at math” message, I think about the student. She wants to study economics because she likes the news, likes understanding why things cost what they cost, likes arguing about policy at the dinner table. And she is about to talk herself out of the field before her first class, because somewhere along the way she became convinced that she isn’t good at math.

This is why I spend as much time on confidence-building in my classroom as I do on actual content. We call it the “figeroutable” muscle and the audacity to believe you can do it.

Dr. A’s take

I think we have the entire financial literacy conversation backward. We treat it as an information problem. We think that not enough people know what a bond is, so let’s teach more personal finance classes. That’s not wrong, but it’s not the bottleneck for a huge share of people, especially women. The bottleneck is a mispriced belief about their own ability, usually formed years before they make it to an economics or personal finance course.

You cannot fix a belief problem with more information. And based on our data, college might already be too late to fix it for a meaningful share of students. This needs to start in middle and high school, in how we talk to kids, especially girls, about math.

Next time you walk into an economics or finance class and see that it is mostly young men, remember that this trend was developed culturally many years ago when women were told that math is for boys.

What to do with this

If you’ve ever avoided a financial decision, a class, or a career path because you decided years ago you’re “not a numbers person,” I want you to treat that sentence as a hypothesis, not a fact. The reality is that an offhand comment from a teacher or parents calling a sibling “ the mathy” and the other “the artsy” one is shaping your decision today.

If you’re a parent or an educator, the actionable takeaway from this research is that the confidence conversation needs to happen years before the financial literacy conversation does. By the time a student is filling out a college application, the gap I study is usually already locked in.

And if a version of “do I have to be good at math to study economics” is running through your own head about your own money right now, the honest answer is the same one I wish I could give every student who messages me: you don’t need to already be good at it. You need to stop deciding, on someone else’s evidence, that you can’t be.

Math can be learned, and you can be good at it.

Have you seen this happen in your life? We want to hear from you about your experiences.

As usual, if you want a copy of the paper, reach out via email.