Kristin Van Dorn

The Mental Accounting of Gift Cards

Did you give or receive a gift card during this holiday season? Since 2007, gift cards have been the most requested gift. They even beat out the ever popular request for a pony. According to the National Retail Federation, American retailers issue almost $30 billion in gift cards each year. Allowing friends and loved ones to pick out their own gifts is an attractive option for many people; gift cards are more thoughtful than cash without requiring much work from the giver. I personally love gift cards. I have several burning a hole in my wallet at this very moment.

But, according to recent research, gift cards have an interesting effect on us. They can subtly alter our preferences for certain items.

You see, we all have a behavioral bias called “mental accounting”. Imagining all of your assets in one pool of money and making choices based on that large pool can be difficult. Deciding what you want to spend on groceries would be daunting if you always had to take into consideration your larger financial picture including what you might want to give to charity this year, if you need to take a trip to visit a friend from college this summer, and if you want to buy a new shirt for an upcoming interview. So, naturally, your brain creates a few shortcuts.

We all delineate our assets into smaller mental accounts, and we closely monitor the balance of these more manageable sums so we don’t have to compare every expenditure to our bigger financial picture. You might have separate mental accounts for things like going out on dates, coffees with coworkers, clothing for work, clothing for the weekends, and different types of entertainment.

But, our mental accounts can trigger illogical responses. A good illustration of how mental accounting affects our decisions is represented in a study by Daniel Kahneman and Amos Tversky. Picture two scenarios:

  • Scenario 1: you are headed to the movie theater. You’ve already purchased your ticket for $10. When you get to the theater, you realize you’ve misplaced your ticket. You could either pay for a new ticket or skip the movie. What do you do?
  • Scenario 2: you are headed to the movie theater again. This time, you haven’t bought a ticket yet. But, on the way to the theater, you accidentally drop $10 somewhere on the street. You get to the theater and you’re $10 short. Do you still buy a ticket to see the movie?

In the two scenarios, the outcomes are exactly the same. But, when asked, most people say they are much more likely to go to the movie in Scenario 2 than in Scenario 1. And, that’s because of our mental accounting. In Scenario 1, a person has already depleted the mental account for movie going. Suddenly, it feels like spending $20 for one ticket instead of $10. In scenario 2, the person hasn’t exhausted the movie going account yet, so she doesn’t feel like she’s buying a replacement movie ticket at all.

Back to gift cards.

Researchers Nicholas Reinholtz, Daniel Bartels, and Jeffrey Parker wanted to see what kind of effect mental accounts might have on purchasing decisions. They speculated that once a person creates a mental account, they will show a preference for items or services most typical of the account’s category. To test this theory, they created a series of studies with both retail specific gift cards (like a gift card to Levi’s) and non-retail-specific gift cards like an American Express gift card.

What they found was that gift cards do in fact change our preferences. Participants who received a retailer-specific card showed a stronger desire to purchase products with high typicality for that brand.

Here’s how it works: You receive a gift card for a specific retailer. Right away, you start planning for using that gift card. You create the goal of spending your gift card along with a new mental account. During this process, you experience a subtle shift in preferences related to your completion of this goal (you want to essentially “win” on your goal, right?) Based on the brand of the retailer-specific gift card, you’ll develop a preference for products that are especially typical of that brand.

For example, even though Barnes and Noble has a selection of games, movies, and magazines, your spending goal will subtly direct you to prefer to buy a book with your gift card. Or, say you get a Gap gift card. Even though the Gap offers shoes, accessories and handbags, you’ll be ever-so-slightly more driven to buy clothing.

Why is this important, you might ask? Well, there are some practical applications. For example, retailers may want to feature their high-end core branded items during high gift card usage periods, like in January. It’s possible that companies could infer some useful cues about their brands from how their shoppers use gift cards.

On an individual level, if you want to manipulate the preferences of the people in your life, you could always do so by buying them gifts to the stores closely associated with things you most want them to enjoy. If you want to steel yourself against this kind of preference manipulation, you can think a little more deeply about how you want to spend your gift cards.

But, possibly most important, this information should help us to reflect on our mental accounting processes, how we earmark funds, and how our spending decisions connect with our larger financial goals. Our retail-specific gift cards might be able to support a deficit in a pre-existing mental account instead of manufacturing a brand new one.
On the other hand, gift cards are meant to be fun. Spending it on something frivolous is half the fun.


Reinholtz, N., Bartels, D. M., & Parker, J. R. (2015). On the Mental Accounting of Restricted-Use Funds: How Gift Cards Change What People Purchase. Journal of Consumer Research, ucv045.

Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. American psychologist, 39(4), 341.

The paradox of American political thought and how liberals are even more WEIRD

Normally, I love election years. I enjoy watching campaigns rise and fall, I like talking politics with friends, and I jump on new analysis with the speed and ferocity of a NASCAR pit crew. However, this political season has been a bit grueling so far. Between Trump Fatigue and an endless barrage of tragic events with politicians clamoring to express their opinions, the climate feels absolutely hostile to civil discourse and nuanced conversations. I’m beginning to dread political chatter with friends and family, and I find that I just long for it all to be settled.

I recently read an interesting finding from a Pew Research Center study:

Just 25% say that, “on the issues that matter,” their side has been winning more often than it has been losing. More than twice as many (64%) say their side loses more often than it wins. The feeling that political losses outnumber victories is widely shared across demographic groups.

A majority of Americans feel like their political party is losing ground most of the time. It seems bizarre to me that both sides would simultaneously feel that way. That means, even when we’re winning, we all somehow still feel like we’re losing. And, we’re primed for defensive positions at those holiday gatherings.

When I’m feeling out of sorts politically, I find comfort in the work of Jonathan Haidt, a researcher at NYU’s Stern School of Business and the author of The Righteous Mind: Why Good People are divided by Politics and Religion (it’s no exaggeration to say that this book was life-changing for me; it completely altered the way I interact with people. And, for about a year, I managed to shoehorn it into nearly every conversation).

Last year, he co-authored a paper with Thomas Talhelm and others that looked at the political divide between liberals and conservatives through a WEIRD lens. WEIRDs—Western, Educated, Industrialized, Rich, and Democratic—make up about 15% of the world’s population. And, over the last few years, we’ve realized most people are not WEIRD, and that WEIRDs tend to see the world quite differently. Researchers detected significant differences in WEIRDs’ sense of fairness and cooperation, visual perception, spatial reasoning, moral reasoning, reasoning styles, self-concepts and more.

Generally, researchers look at WEIRD phenomena on an intercultural level. For example, it’s often useful in looking at how international business employees work across cultures. But, in this study, the researchers looked at WEIRD characteristics on an intracultural level to see if there are differences in WEIRDs themselves and if those differences explain political identities. They predicted that liberals would be even more WEIRD than conservatives in the ways that they think and perceive the world.

I’m about to tell you what they found, but first I want to explain analytic versus holistic thinking. Analytic thinking is about drawing connections between objects and the categories they could belong to. Holistic thinking is about drawing connections between objects and the contexts they could belong to. For example, an analytic thinker might imagine the personality traits that make a person more likely to honk his or her horn while driving, and a holistic thinker might imagine the traffic conditions. This holistic versus analytic thinking style is particularly predictive of WEIRDs and non-WEIRDs. Each person has the ability to think in both ways, but WEIRDs show a bias towards the analytic thinking style, and non-WEIRDs show a bias towards the holistic thinking.

Back to the research, the study authors compared how biased towards holistic thinking a person was with how conservative a person was, and they found a strong correlation. They tested this with a group of college students, a wide range of online participants, and a group of Chinese participants.

Here comes the fun part. Once researchers discovered there was a correlation between how liberal a person was and their preference for analytical thinking, the researchers wanted to see if they could find a causality. Could they manipulate a person’s political preferences by priming them to use their alternate thinking style?

First, participants filled out a pre-screening questionnaire that asked their political identity. The researchers waited a few weeks before beginning the next part to control for prior political identity. Participants were assigned at random to one of three conditions: control, categorical, and relational. The groups took a test where each question presented three objects and asked the participant to identify which two items of the three belonged to together. The Categorical group was asked to pair items based on categories. The Relational group was asked to pair items based on which objects shared a relationship or which object uses another object. An example question might be:

  • Hand, Mitten (relational)
  • Mitten, Scarf (categorical)
  • Scarf, Hand

Participants from the categorical group and the relational group were given practice questions. If they made mistakes, their responses were excluded from analysis. The control group was not given special instructions to choose an answer based on categories or relationships.

Afterwards, the participants were asked to read two news articles—a politically heated one opposing welfare programs, and a politically neutral article on opposing mainstreaming special education students—and respond to questions. Finally, they were asked for their opinions on common political issues. By instructing participants to read the articles, the researchers were asking participants to process multiple arguments and form a contextualized opinion. By asking participants to answer questions about their more general issue-based political views, the researchers were trying to evoke the participants’ stable political opinions.

The triggered thinking styles strongly predicted participant responses to the welfare question. Generally, people in the categorical/analytical group chose the liberal position, and people in the relational/holistic group chose the conservative opinion regardless of the political position they walked in with. This thinking-style prime had an impact across the political spectrum, effecting both liberals and conservatives. If there had been a vote directly after the priming exercise, when people were primed to think analytically, the liberal plan would have succeeded. When people were primed to think holistically, the welfare program would have failed.

Now this research raises a bunch of questions, many of them addressed by the authors. For one, the authors only tested the relationship between the analytic and holistic thinking styles against a person’s social politics, not a person’s economic politics. This is in part due to the effect of libertarians. For another, we tend to think of conservatives as “more American.” But, this research might suggest something quite the opposite. So, how do we tease apart what it means to think like an American today?

Probably one of the most interesting questions for me is in regards to higher education. You often hear conservatives complain that our institutions of higher education indoctrinate our students in a liberal agenda. This research could actually align with that argument. If professors are asking students to think more analytically, and then higher education is introducing students to more nuanced political contexts, over time, this would potential push students to a more liberal point-of-view.

In any case, Jonathan Haidt did not disappoint. I somehow find this research soothing. We’re more malleable in our political opinions than we realize. If we could complicate political discussions simply by evoking different thinking styles, we could hopefully develop some new tools for more civil dinner conversations, and maybe even make more nuanced policy decisions. And, maybe asking family members to play a sorting game or two before we settle down for a nice meal could ease a few tensions.

Talhelm, T., Haidt, J., Oishi, S., Zhang, X., Miao, F. F., & Chen, S. (2015). Liberals think more analytically (more “WEIRD”) than conservatives. Personality and Social Psychology Bulletin, 41(2), 250-267.

How much does self-identity inform our behavior?

“The most difficult thing in life is to know yourself.”
– Thales

A recurring question for me this year has been whether or not the process of identifying our own personality traits is always good for us. This tickle in the back of my mind likely started with a rash of Strengths Finder workshops and Myers-Briggs inventories at work. Many of my co-workers have even started listing their top 5 strengths and their Myers-Briggs types in their email signatures. In addition, after a few recommendations, I got around to reading, “Quiet: The Power of Introverts in a World That Can’t Stop Talking,” and before I knew it, a surprising number of friends and family members were talking about how they “charge their batteries”, love to watch movies on the couch in their pjs, and hate big parties. Suddenly, even extroverts were introverts.

Don’t get me wrong—I think my top 5 strengths are the coolest (competition, futuristic, ideation, input, and strategic) and I love and identify with my Myers-Briggs type (INTP). I’m easily suckered into online quizzes. You don’t even have to ask me; of course I want to know what kind of exerciser, lover, friend, thinker, co-worker, chef, dater, and communicator I am!

But, I wonder what kinds of effects all of this self-tested, self-diagnosed, self-interpreted self-knowledge has on our behaviors. Do we only see the traits we want to see? Does it give us an excuse for bad behaviors? Do we drive ourselves into over-simplified, shortcut-taking caricatures just to fit the stereotypes and preferences that entertain us?

Lucky for me, I am not the only one pondering these questions. Maferima Touré-Tillery of Northwestern University and Ayelet Fishbach of the University of Chicago designed a series of simple studies to test how much a person’s self-perception influenced their self-control. The question the authors attempt to solve is of the “which came first” variety. Previous studies have already demonstrated that successfully resisting temptation can help build our sense of self. We intuitively know this. We feel a sense of pride when we stick to our diets or put our credit cards away. And consequently, we may even feel as though we are good at self-control. But, this study looks at the decision-making process from the other side: are we more likely to self-regulate when the outcome reflects positively on what we already believe about ourselves? In other words, is it because we believe we possess certain qualities that we’re able to behave in a certain way in the first place?

In order to find out, first the authors had to differentiate between a person who was simply motivated to do something well versus a person who was motivated to do something well in order to validate something he or she believes about his or herself. They turned to earlier research (Touré-Tillery and Fishbach, 2012) that demonstrated people tend to more closely stick to religious traditions and ethical standards at the beginning and end of a sequence of events than in the middle of that sequence. The reason is, we intuitively feel a bias towards beginning events and end events. That’s why making a strong first impression or leaving on a high note are sort of universal hopes.

One might assume that if a goal is important to you personally (intrinsic), how you did at the beginning, middle, and end would make no difference. However, if the goal is important because of how you will be perceived by others (extrinsic), the framing events—that is the beginning and the end—would naturally be more important. To test this theory, the researchers intercepted people near a coffee shop. They asked them to imagine they were on a caffeine detox, and how they would feel about themselves if they gave into their coffee cravings at the beginning, middle, or end of the detox. Results showed that people imagined feeling worse about themselves if they gave into their cravings at the beginning or the end than if they gave into their cravings in the middle of the challenge.

In a second experiment, because how we judge others tends to be a good indicator of how we judge ourselves, the researchers asked a second group of people questions about how they would feel if a friend were on a caffeine detox. First, the researchers asked whether drinking coffee is healthy or not. Then they presented the participants with a story similar to the first scenario. If a friend gives into his or her cravings in the beginning, middle, or end of his or her caffeine detox by ordering a coffee, what does that say about his or her level of commitment? The participants who believed caffeine was bad for their friends saw the failure at the beginning or the end of the detox period as being more diagnostic of their friend’s level of commitment to the detox than a lapse in the middle.

Once the researchers determined that an event would have higher or lower influence in how a person evaluated their own behavior based on framing the event as occurring at the beginning, middle, or end of a sequence, the researchers were ready to test how strong the framing effect really was. Through four more experiments, Touré-Tillery and Fishbach tested when people were more likely to slack in their health goals and when people were more likely to spend extra money when they were trying to save. They even tested when people were more likely to select high-brow versus low-brow magazines.

What they found was that while it’s true that people will exercise self-control to achieve a goal, people will also exercise self-control when they consider what their restraint means in their own understanding of themselves. They say:

Indeed, people for whom a goal-related identity is important get more out of self-signaling that identity and are thus more responsive to self-diagnosticity cues such as framed positions. We also show people for whom a goal is important displayed better self-control than those for whom the goal is less important for choices at the beginning or end (high self-diagnosticity). However in the middle (low self-diagnosticity), people for whom the goal is important indulged and splurged as much as others.

Our identities are tangled into our goals and our abilities to meet the goals we set for ourselves. If we want to believe we’re health-conscious, financially-savvy, or even intellectual, we might exercise our restraint in order to reinforce our own beliefs about ourselves as much as we would just to meet our goal in the first place. And we can see when we’re disciplining ourselves for the sake of the goal vs the sake of our egos by how consistent we are throughout the process.

While, it is true that achieving a goal or doing well at something increases our identification with that success, if we want to be able to believe we are already competent or good at something, we’ll engineer our situation to validate that conclusion. It’s a two-way street between how we act and how we see ourselves. Just as our actions indicate to us how to think about ourselves, who we think we are indicates to us how we should act.

This study doesn’t completely prove my point. It doesn’t attempt to solve whether or not knowing our own personality traits limits how we will respond to certain situations. But, it does suggest that when we want to validate something we already believe about ourselves, we’ll possibly create the evidence we need.

Touré-Tillery, M., & Fishbach, A. (2015). It was (n’t) me: Exercising restraint when choices appear self-diagnostic. Journal of personality and social psychology, 109(6), 1117.