The Psychology of Money: Why We’re Bad at Predicting Expenses and Income

Ever wondered why your budget always seems to fall short? Or why that freelance consulting gig didn't pay off as much as you expected?

Associate professor Ray Charles “Chuck” Howard's research on financial decision-making sheds light on the psychological pitfalls we face when it comes to money. From expense prediction bias to the surprising benefits of unrealistic budgets, his insights offer practical tools for anyone looking to improve their financial health.

Ideas to Action spoke with Howard to explore why we're so bad at predicting our expenses, how optimistic budgets can actually help us spend less and why gig workers tend to overestimate their income.

Howard recently joined the faculty in the marketing area at the University of Virginia Darden School of Business, where he will teach marketing analytics. These are edited excerpts from the conversation.

Q: One of your research areas is financial decision-making. Why?

A: Financial decision-making is getting harder and more of it is falling to the consumer.  For example, 50 years ago you would go work for a company. That company would have a pension plan, you’d spend your whole career there, then you’d retire. And that was that. Now you might work in the gig economy, which means your income is volatile and hard to predict. Even if you do work in a more traditional industry, you almost certainly don't have a pension, so — you have to make your own retirement-savings decisions. All of that involves making predictions about how much money you're going to earn, spend and save. Many people find that challenging, so I think it's important to identify when and why our financial decision-making goes awry, and how we can help people make better financial decisions.

Q: In 2021, you published a paper on expense prediction bias. What did you uncover?

A:  Expense prediction bias refers to the fact that most people dramatically under-predict their future spending. This happens because we base our spending predictions on really common or typical outcomes, and we do a very poor job of accounting for atypical, anomalous things. So, if we're trying to predict our spending for the next week or month, it's really easy to think of things like groceries, gas, rent or the mortgage payment, but it's a lot harder to think of things like a car repair that will cause us to spend more money than we typically would. And so those atypical events don't end up being built into people’s predictions, and they up spending a lot more money than they thought they would, or than they wanted to.

Q: Why are we so poor at predicting our expenses?

A: There are several reasons, but it’s mostly about cognitive accessibility: The things that come to mind first and most easily when you predict your expenses are the things that happen most frequently. It's things like your groceries, which most people buy every week; not things like healthcare expenses, which most people encounter irregularly. It’s not that people are trying to make optimistically low spending predictions. It's just how our brains work — it is easy for us to think of highly typical expenses that are reinforced by our own behavior, and hard for us to think of contingencies.

Q: What’s the solution?

A: If you want to make more accurate spending predictions, simply take some time to think of reasons why your expenses might be different than usual. That will almost always lead you to think of reasons why your expenses will be somewhat higher than you think. As simple as it sounds, that leads to much more accurate predictions.

Q: You have published another paper on when and why underpredicting one’s expenses may be beneficial. What did you discover?

A: There is a very important difference between when you want to accurately predict your future spending and what you want to do if you're setting a budget that you're going to track your spending against on a regular basis.

Let's say you're sitting down to decide how much you can afford to spend on a major purchase like a house. In situations like this you have to have a very clear prediction of what your monthly spending is going to be in the future, so that you can determine how much of a mortgage you can afford to pay every month.

In that case, you want to make the most accurate expense prediction that you can and, if anything, you want to err toward over-predicting your expenses so that you don't end up taking on higher mortgage payments than you can comfortably pay. That's where you want to have accurate spending forecasts.

Here’s the twist: If you are trying to manage or reduce your spending, and you decide to set a budget to accomplish that, our research shows that what you actually want is an optimistic budget — in other words, a budget that is optimistically low.

Let's say you usually spend $500 a month going out to restaurants and you see that as an area where you can spend less. You might say, “OK, well, a realistic budget, given my social obligations, would be something like $400 a month.”

If you really want to decrease your spending, you should just be really optimistic and say, “No, next month it's going to be $250. That’s my total budget for eating out.” And if you're like most people, you're going to end up spending more than $250 because something will come up that you just can’t say no to. But let's say you end up spending $350 — you spent substantially more than you budgeted, but you also spent way less than you used to.

Now, research also shows that if you had set your budget at $400 because you wanted to be realistic, you would have spent at least $400, so you also spent less than you would have if you were trying to budget realistically. This is a nuanced point and it’s really important: You actually don't want people to make accurate budgets. You want them to set optimistically low budgets, because the lower the budget, the stricter that reference point becomes, and the more likely they are to reduce their spending, even though they're almost certainly going to spend more money than they budgeted.

Q: You’re also working on a paper that explores whether people who work in the gig economy tend to mispredict their future income. What did your research uncover?

A:  The gig workers in our studies display an income prediction bias in which they substantially over-predict their gig income. Interestingly, the reason they over-predict is because they over-predict the number of hours they will work. They are actually really good at predicting how much money they will earn per hour. This stands in contrast to the allegation that gig economy firms mislead workers about what their hourly wage will be.

Q: Are people simply too optimistic?

A: Being inherently optimistic does nothing to predict whether you know how much money you're going to make from the gig economy. The reason why predictions end up being optimistically high is because people fail to account for atypical events that will interfere with their ability to work, like getting sick, or their car breaking down.

Q: Why are we such poor planners when it comes to the unexpected?

A: We tend to focus on what is typical in a normal week, but atypical events happen often enough that they are bound to influence our prediction accuracy over time. In the context of working in the gig economy, those atypical outcomes are almost always bound to lead to less income than more.

Q: How does your gig research apply to the real world?

A: Let’s say you’re a gig worker who needs an accurate sense of how much you'll earn next week. Our research shows that the best way to make an accurate prediction is to simply look back at your weekly earnings over the past month and take the average. And ultimately, if your income prediction is more accurate, you make better spending decisions.

Q: Presumably that means taking on less debt?

A: Yes. Imagine you’re a gig worker and you’re at Best Buy trying to decide which TV to get for your new apartment. The $2,000 TV is much better than the $1,200 TV, and you’re really confident that you’re going to earn $2,000 driving for Uber next month, so you use your credit card to buy the $2,000 TV, believing you’ll have it paid off by the end of the month.

Now, what happens when the end of the month comes, and you’ve only made $1,200? The remaining $800 sits on your credit card. If you work in the gig economy, you might even be taking out a payday loan to pay for some of your purchases — and you're paying an astronomical APR that makes it very hard to get rid of that debt.

Then you’re stuck paying off the debt rather than, for example, buying a house to build equity or saving for retirement. You’re in financial turmoil that could have been avoided if you had a clearer forecast of what was going to happen in the future.