What causes loss aversion?

In my first week of my first marketing job, my boss told me something I’ll never forget:

There is one thing that motivates people more than anything else: the fear of loss.

He’s right. People go to great lengths to avoid a loss. In ten years of working in marketing, I have found that appealing to someone’s fear of loss drives people to action—to click on the button you want them to click, to attend the meeting you want them to attend, to watch the product demo you want them to watch, to buy the product you want them to buy.

When you tell them the sale is ending soon, or it’s the last chance, or this product is going away forever, people behave in the same way: they take action.

You can see evidence of this everywhere:

Loss aversion last chance email
source: Sumo

In a study of how people buy cars, researchers discovered that when customers are presented with a fully-loaded model and asked to subtract features, they typically spend more than people who are presented with a base model and asked to add features. That’s because adding features feels like a gain, while removing features feels like a loss. Customers who started with a base model spent $13,651.43, while people who started with a fully loaded model spent $14,470.63.[1]

You can also see this phenomenon in the real estate market. Two economists found that among 5,773 condos listed in the downtown Boston market between 1990 and 1997, sellers who faced a loss listed their condos 25 to 35 percent between the expected selling price and their original purchase price, and then sold at a price between 3 and 18 percent of the expected selling price and their original purchase price. This was much higher than people who stood to profit by selling their condo. People who faced a loss were more motivated to list (and sell) their property at a higher price than people who faced no such loss.[2]

That’s because losses hurt more than their equivalent gains.

Suppose it’s a hot day and you’re eating an ice cream cone. Without warning, the ice cream falls off your cone, hits the sidewalk, and melts into oblivion. Suddenly, you have no more ice cream—a loss.

Now, suppose it’s that same hot day, and you’re eating that same ice cream cone. But this time the ice cream stays on the cone. You finish the cone and move on with your life. In this scenario, you’ve eaten the ice cream—a gain.

Did you know the ice cream in the first scenario is worth more than the ice cream in the second scenario? It’s true: when you don’t eat it, it’s a loss. When you do, it’s a gain.

You might be thinking: how could you possibly know this?

You’re right that you can’t exactly compare how you value the ice cream in these two scenarios, because you can only experience one of them. You can’t live through two versions of history—one version where you eat the ice cream, and the other where you don’t—and then compare the value of the ice cream in each version of history.

Fortunately, economists have figured out a clever way to compare two versions of history, just like this—where the same item is valued differently depending on whether it is experienced as a loss instead of a gain.

Here’s how: they ask people to make a bet.

Let’s say I flip a coin. If it’s heads, you win $10. If it’s tails, you lose $10. Would you take the bet? Most people wouldn’t.

What about this bet: if it’s heads, you win $20. If it’s tails, you lose $10. Would you take the bet now?

Some people would take this bet, although a few people still wouldn’t.

Two psychologists, Amos Tversky and Daniel Kahneman, that latter of whom would win the Nobel Prize in economics, found that people usually take the bet when the winnings approach $22.50.[3] In other words, if the coin lands heads, you win $22.50, but if it lands tails, then you lose $10. They discovered how much losses hurt more than gains: it takes winning $22.50 to make up for losing $10.

So we don’t need to rewind history and compare two scenarios. Instead, we can simply see what kinds of values people place on two outcomes with an equal probability of happening.

This difference in value between the two outcomes—one outcome with a loss, and the other outcome with a gain—is called the loss aversion ratio. If you value losses and gains equally, then your loss aversion ratio is 1:1. If, as Tversky and Kahneman found, you value a $22.50 gain the same as a $10 loss, then your loss aversion ratio is 2.25:1, or simply 2.25.

loss aversion

In Tversky and Kahneman’s original study, they proposed a universal loss aversion ratio of 2.25—that is, people value losses as 2.25 more than their equivalent gains. Subsequent research has found that loss aversion ratios actually vary quite a bit. In some cases, people have a loss aversion ratio as high as 5. And in other, albeit rare, cases, people’s loss aversion ratio is actually negative. Instead of being loss averse, they are loss seeking. (In a moment, you’ll see why these are such rare exceptions.)

In a meta-analysis, or a study of studies, Lukasz Walasek and his colleagues from the University of Warwick identified a mean loss aversion ratio of 1.31.[4] What this means is that, according to the best statistical analysis available—not perfect, but pretty good—if you asked everyone on the planet to place a bet on a coin toss, people would, on average, accept a loss of $10 in exchange for a win of $13.10.

But their study left one big question unanswered: why are some people more loss averse than others?

Take a look at the chart from their study:

loss aversion ratio meta-analysis
source: Walasek, L., Mullett, T., and Stewart, N. (2018). “A meta-analysis of loss aversion in risky contexts.”

Why do some people value gains and losses almost equally, while others place a very high value on losses compared to gains? And why do the same people who are loss averse in one instance on day not loss averse in another instance the next day?

In the rest of this post, we’re going to explore the complex, intertwined causes of loss aversion. We’ll start with the big picture in mind, looking at cultural, social, and environmental factors which make you more likely to be loss averse in some situations but not others. Then, we’re going to look at some of the physiological reasons behind loss aversion, and discover clues in our brain, in our hormones, and even in our genes.

Here are the two questions we will attempt to answer:

  1. What makes some people more loss averse than other people?
  2. What makes you more loss averse in some situations but not others?

Let’s dive in.

Does your culture make you more loss averse?

Your cultural background helps determine how loss averse you are. Mei Wang and her colleagues surveyed groups from 53 countries to understand how people from different cultures value losses compared to gains.[5] They discovered people from Eastern European countries tend to be the most loss averse, with a loss aversion ratio around 2.3, while people from African countries were the least loss averse, having a loss aversion ratio around 1.4.

loss aversion culture
source: Wang, M., Rieger, M., and Hens, T. (2017). “The impact of culture on loss aversion.” Journal of Behavioral Decision Making 30(2), 270–281.

The difference in the loss aversion ratio is more apparent in the country-by-country comparison:

loss aversion culture
source: Wang, M., Rieger, M., and Hens, T. (2017). “The impact of culture on loss aversion.” Journal of Behavioral Decision Making 30(2), 270–281.

What accounts for these wide variances? The researchers identified three cultural traits that correlated with higher loss aversion ratios: individualism, power, and masculinity.

1. People from individualist cultures are more loss averse than people from collectivist cultures.

If you’re from a collectivist culture, you’re more likely closer social connections and more of them. This means if you make a bad decision, you can count on friends, family, and others to help you. Simply knowing this can lead you to take a few more risks–and not feel the losses quite as much.[6]

But if you’re from an individualist culture, you don’t have quite the same safety net. You’re expected to take care of yourself. It doesn’t mean others won’t help you–you can probably get a little help from family and friends. But such support won’t come as a result of the same ingrained cultural norms.

2. People from unequal cultures are more loss averse than people from cultures where equality is valued.

If the less powerful members of society accept that power is distributed unequally, then you are more likely to believe a loss is less preventable and comes with greater consequences. In other words: losses are scarier. As a result, you’ll do more to avoid losses.[7]

There are lots of imperfect ways to measure inequality within and between cultures, one of the best and most accepted seems to be the Power Distance Index, developed by a social psychologist from the Netherlands named Geert Hofstede. According to this index, the higher the number, the more unequal a society, and the more likely “less powerful members of organizations and institutions (like the family) accept and expect that power is distributed unequally.”[8]

Wondering where your country ranks on the Power Distance Index? The countries with the most inequality include Malaysia (104), Guatemala (95), Panama (95), and the Philippines (94). Countries with the least inequality include Austria (11), Israel (13), Denmark (18), and New Zealand (22). The United States ranks at 40, and the United Kingdom ranks at 35.

It doesn’t matter if you’re powerful or powerless. If you’re from a country where it’s accepted that some people are powerless, then you’re more likely to be loss averse.

3. People from cultures that value masculinity are more loss averse than people from cultures that do not value masculinity.

It seems both men and women are equally loss averse. Saying masculine traits lead to loss aversion does not mean men are more loss averse than women. Although a few studies do seem to indicate this, other studies indicate the opposite–that women are more loss averse than men[9]–but most don’t find any distinction.[10]

Instead, when a culture values attributes that are generally considered aggressive, prone to fear, or goal-oriented–to name just a few–people from that culture also tends to be more loss averse, too.

What does it mean for a culture to value masculinity? When a culture values goals associated with wealth or career, the researchers label it as masculine. People from cultures that value masculine traits tend to be more aggressive, suffer from depression, and a prone to fear. Cultures that value masculine traits tend to be more loss averse, too.

How cultural assumptions affect loss aversion

Loss aversion is correlated to cultural concepts of individualism, power, and masculinity. But what is it about those cultural concepts make people more loss averse?

Another group of researchers pored over similar datasets and noticed a correlation between loss aversion and a country’s gross domestic product—or the sum total of all the goods and services in its economy. These researchers found that people from wealthy countries with high GDPs are less loss averse. But in poorer countries with lower GDPs, people are more loss averse.[11]

loss aversion gdp
source: Foellmi, R., Jaeggi, A., and Rosenblatt-Wisch, R. (2018). “Loss aversion at the aggregate level across countries and its relation to economic fundamentals,” Working Papers 2018-01, Swiss National Bank.

That’s interesting. But does it mean anything?

This would be a good place to hit the brakes and make sure we’re not confusing correlation with causation. Even though you’re more likely to be loss averse if you live in the country of Estonia or Georgia, that does not mean living in Estonia or Georgia causes you to be loss averse. But why, then, are people from some cultures clearly more loss averse than people from other cultures?

It should be obvious by now that there are other factors that affect loss aversion–some of which are mediated and enhanced by someone’s cultural background, and others of which affect behavior apart from any cultural context.

Next, let’s look at some of the social factors that are likely to make you more loss averse—regardless of what culture or country you’re from.

Social causes of loss aversion

Humans are fundamentally social beings. Our spot in the social hierarchy can predict our level of loss aversion pretty well.

Having power makes you less loss averse

Ena Inesi of the London Business School found that people in power are less loss averse. Even just thinking you’re powerful makes you less loss averse. That’s because, in many cases, powerful are in a better position to accept a loss should it occur. They’re often wealthier and have the means to compensate for a loss. As a result, they give less weight to losses compared to gains than non-powerful people.

There’s another more important reason. They give more value to gains than non-powerful people.[12] Let’s go back to that coin toss example I used earlier. If I toss a coin and it lands tails, you lose $10. But if it lands heads, you gain $10. We’ve seen that most people wouldn’t take that bet. But a powerful person would. They a higher value on the gain of $10 than a non-powerful person (even though it’s, objectively, the same amount). And it’s because a powerful person isn’t going to feel the loss of $10 as acutely.

Is this a good thing? After all, isn’t loss aversion—valuing a $25 gain the same as a $10 loss—irrational? Sometimes. Sometimes being too loss averse can be costly. For example, when traders hold onto losing stocks too long because of loss aversion, but then quickly sell a stock when it makes money, it’s because they value losses more than gains–and the portfolio suffers. This is a common mistake committed by junior analysts at a firm. But senior traders have the opposite problem: they dump losing stocks before giving them a fair chance and hold onto winners longer than they should.

The phenomenon of powerful people over-valuing gains also explains why employees question their boss’s willingness to risk a loss on a new project or product. I once worked for a software company where the CEO made lots of risky decisions, some of which could have resulted in a loss. In many cases, there was real revenue on the line—and lots of it. The CEO, holding a position of power, was more likely to make the decision knowing it could result in a big loss, because he felt that loss a little less than the employees, and felt the gain a little more. (In hindsight, most of those risky decisions turned out to be brilliant, and over the years he has built an incredibly successful company. Smart CEOs take risks people under them might not be as quick to accept.)

The relationship between wealth and loss aversion

With power often comes wealth, and it seems wealthy people would have an easier time accepting losses, too. After all, losing $10 feels worse to someone with $100, compared to someone with $10,000, $100,000, or $1 million.

Unfortunately, it’s not so simple. One study of 660 people in Germany found that those with higher household incomes also tend to be more loss averse. Why would this be?[13]

A different study in Vietnam revealed a surprising connection between wealth and loss aversion. It’s not how much money you have that matters. It’s how wealthy your social environment is compared to the norm.[14] In the study, wealthier villages, as a whole, were less loss averse than poorer villages. That means if you’re wealthy and live in a poor village, you’re more likely to be more loss averse than someone who is poor and lives in a wealthy village.

The relationship between education and loss aversion

Education predicts lots of social outcomes, and loss aversion is no exception. On the face of it, this makes sense: more education usually confers some extra benefit that results in a more accurate understanding of the world. It would seem a more accurate understanding of the world would lead to less loss aversion–more knowledge translates into less risk. But it wasn’t until 2008 that social scientists were able to make a connection between education and loss aversion.[15]

This study tracked 315 cotton farmers in China from 1993 to 2004. Every year, these farmers risked losing their crop to a small pest called the bollworm. In 1993, everything changed. That year, a genetically modified variant was introduced that protected against bollworm infestation.

But most farmers didn’t adopt the new strain, at least not right away. In fact, it took eleven years for all the farmers to adopt it in the region that was studied.

loss aversion education
source: Liu, E. (2013). “Time to change what to sow: Risk preferences and technology adoption decisions of cotton farmers in China.” Review of Economics and Statistics 95(4), 1386–1403.

The first farmers planted it in 1993, but the last group of farmers to plant it didn’t do so until 2004. Why so much time?

One possible reason was that the genetically modified cotton cost more. But this reason didn’t hold up, because it’s also saved 28% in lower pesticide and labor costs–it actually made the crop more profitable.

After controlling for other variables, the researchers found that farmers with higher loss aversion ratios delayed adopting the new crop. This is no trivial matter: farmers who adopted the new crop in 1993 became wealthier than the farmers who waited until 2004. Loss aversion led to worse yields overall and a less profitable crop.

The researchers dug deeper and noticed a correlation between the number of years of education and the likelihood of adopting the new crop. For every year of education a farmer had, they were 4.3% more likely to adopt the new crop, thereby earning more than their less educated, more loss averse counterparts. Lack of education leads to greater loss aversion, and, for farmers in rural China, loss aversion was costly.

So far, we’ve been looking at external factors that increase your likelihood of being loss averse. All these factors–cultural context, wealth, power, education, and others–have one thing in common: they, like all our experiences are mediated by our brain.

That’s where we’ll turn next.

Loss aversion and the brain

How does the brain process a loss?

The brain is a complicated organ. There isn’t a one-to-one correlation in the brain between stimulus and response. It’s not like a set of switches that flip on and off depending on what you experience.

This means when we think about how our brain processes loss aversion, we can’t look for a single region. Instead, it’s a set of interconnected regions that work together. Let’s take a look at a few of these regions.

The amygdala

It’s easy to forget that only in the past couple decades have scientists been able to peer into the brain with technologies like fMRI. It’s been only recently that we’ve been able to understand how the brain reacts to losses.

One of the first clues came from a study on fear. Researchers asked people to make the coin toss bet. One group of people made the bet after seeing a picture of a fearful face. The other group made the bet after seeing a picture of a neutral face. When people saw a picture of a fearful face, they tended to be more loss averse. The effect was enhanced when they made the bet while simultaneously looking at a fearful face.[16] This established a fairly tight connection between loss aversion and fear. But what was the connection?

Fear is processed by a region of your brain called the amygdala. But this isn’t just any fear; it’s a kind of automated, pre-conscious fear. When you see a spider, you have an instant reaction. Same goes for snakes. Or if your palms start sweating on a bumpy flight. This is the kind of fear your amygdala deals with. Even if you consciously know a spider won’t harm you, or a snake won’t bite you, or your plane won’t crash, no amount of higher-order cognitive rationalizing can override the amygdala’s automated response to fear stimulus.

Your amygdala responds to potential losses in the same way. True, your palms won’t sweat like they would on a plane, and you won’t scream like you might after seeing a spider. It doesn’t matter. When you encounter a loss, your amygdala gets to work. It processes the stimulus, and it tries to override the more rationale parts of your brain.

The connection between the amygdala and loss aversion was confirmed in a fascinating study conducted by Benedetto De Martino and his colleagues from Cambridge University. They found two people with amygdala damage—one person whose amygdala was roughly 50% damages, and the other whose amygdala was completely damaged. Next, they offered the coin toss bet to these people, along with two controls groups.

The person with partial amygdala damage had a loss aversion ratio of 1.06. In other words, they were loss neutral: they gave gains roughly the same value as losses. They accepted coin toss that resulted in $10 loss lost or $10 gain.

But the person with complete amygdala damage had a loss aversion ratio of 0.76. They were actually loss seeking—they gave losses a greater value than their equivalent gains. They would accept a coin toss bet that involved a $7.60 gain against a $10.00 loss.

Compared to the control groups, they were willing to take extra risks for decisions that involved potential losses. The reason? The amygdala.[17]

The amygdala processes fear and threats–and, it turns out, losses, too. Our brain’s immediate, preconscious reaction to a loss isn’t much different than the reaction to a spider, snake, or airplane turbulence.

But your amygdala doesn’t work alone. If your amygdala was the only part of the brain processing information, you would be paralyzed by preconscious, automated fear.

Think of it this way: on that coin toss bet that came up tails, imagine instead of losing $10, $15, or $20, you instead were forced to share a small space with a python. And now, imagine that your only means of evaluating that bet was with the preconscious, automated fear processing–the kind the amygdala handles. If it was up to your amygdala, you’d never take a bet with any loss. You’d never make a decision that had any downside. You’d be paralyzed with constant fear.

You need more than your amygdala.

The striatum

Scientists have identified other regions in your brain that are active when you process a loss. One of those regions is the striatum. Neuroscientists have found that the striatum lights up more for losses than it does for equivalent gains.[18]

But what does this mean? What does the striatum do?

The striatum handles prediction errors. When your brain predicts that something will happen but then it doesn’t, your striatum lights up. There’s a mismatch between expectation and reality: cue the striatum! That makes it an important part of the brain for learning new things. Next time, your prediction will be just a little bit better.

What does this have to do with loss aversion? This: while your amygdala is busy making you afraid of losses, your striatum is helping you avoid future losses.

The insula

And that’s not all. Your insula goes to work, too.

Your insula is the area of your brain that reacts to disgust. If you think about moldy bread or smelly garbage, your insula is responsible for that aversive reaction you’re feeling.

It’s not just food. Your insula also responds to disgusting behavior, such as a chess grandmaster beating a child in three moves. You might even react to a horrible event by saying it makes you “sick to your stomach.” That’s the connection between your insula’s reaction to disgusting behavior.[19]

It’s not hard to see how your insula can work with your amygdala to make you avoid certain kinds of behavior. The insula provokes the repulsion you feel when you see hundreds of spiders crawling over rotten eggs, the amygdala provides the gut reaction that makes you avoid it, and the striatum makes sure you won’t be in that position again.

Fear and disgust are close friends.

Neuroscientists have observed that, much like the amygdala and striatum, your insula lights up in response to a loss. The greater the prospect of a loss, the more your insula is activated compared to an equivalent gain.[20]

How your brain processes a loss

How do these regions of the brain work together? Here’s one theory. Your amygdala makes you afraid of a loss. Your striatum matches a loss against a predicted outcome. Your insula makes a loss feel revolting. And there are countless other regions of your brain that contribute, too. Remember, there’s no one-to-one stimulus between your sensory perception of a loss and your brain’s response.

How do all these regions work together to make you loss averse?

Nobody knows—at least not yet.

Fortunately, research is proceeding at a furious pace. And the good news is lots of smart neuroscientists have a pretty good idea, and in the next few years, they’ll probably find the experimental evidence to prove it.

Until that happens, one guess:

“One possible model is that the amygdala may mediate loss aversion in responses to outcomes . . . and that the weights in those outcome responses may be passed on to the striatum, leading it to represent loss aversion at the time of decision.”[21]

Or, maybe:

“The bidirectional (gain-loss) signals coded by these regions likely converge to downstream processing structures . . . where they may underpin cost-benefit analyses.”[22]

Or, to be a little more specific:

“. . . A neurostructural signature of loss aversion can be found in a network involving amygdala, thalamus, striatum, and posterior insula. All these structures play a critical role in detecting threats and prepare the organism for appropriate action.”[23]

This doesn’t answer the question of how the brain processes a loss, but it does hint at what the answer will be when it’s discovered.

Here it is:

There are three or four very important regions of your brain that become activated when you process a loss, plus lots of other less important regions, too. There are also parallel structures designed to help you make your way in the world, each responding to loss in its own way. These parallel structures are connected to each other, and somewhere in “downstream processing structures” your brain integrates all the data points together and prompts you to respond in a certain way. Depending on the strength of these competing inputs you’ll give some inputs more weight in your decision than others. If the input from your amygdala is especially strong, you’ll be a little more loss averse. And if it’s weak, then you’ll be less loss averse.

So there’s our answer.

Well, not quite.

We’ve established that multiple regions of your brain are involved in processing a loss. But there’s another piece to the story: the way in which these regions communicate with each other.

And one of the key communication networks involved in loss aversion is the brain’s pleasure processing circuitry.

It’s time to talk about dopamine.

Loss aversion and dopamine

The communication networks you’re familiar with rely on electrical circuitry to communicate. The brain’s network of nerves uses the same design, but with one important difference. At the point where nerves connect to each other, these electrical signals are converted to chemical signals, which are then converted back to electrical signals in the next-door nerve.

These chemicals are called neurotransmitters, because they pass signals from one neuron to another.

Your brain’s ability to communicate with itself is dependent on how the network functions. And network functionality is affected by how these chemicals pass information back and forth between nerve cells.

One of the most well-known neurotransmitters is called dopamine. Depending on how well dopamine communicates information between nerve cells has a big impact on how loss averse you are.

What, exactly, does dopamine do?

Dopamine is the chemical your brain releases to reward your behavior by making you feel good. Here are a few activities that release dopamine: drinking alcohol, taking cocaine, having sex, thinking about having sex, thinking about food (especially if you’re hungry), having an aesthetically pleasing experience (looking at art or listening to great music), punishing someone who treats you badly, winning a contest, and on and on and on. It’s the feeling of “runner’s high,” and it make us feel good when we pull down to refresh—which is why Facebook can be so addicting.

In fact, it’s just enough to think about a pleasurable experience to trigger the pleasure circuit. This is why simply being in an environment where people are drinking can cause an alcoholic to relapse. After you have a pleasurable experience, the memory of it is enough to trigger your desire to have it again.[24]

Each time you experience one of these stimuli, or even think about doing so, your brain rewards you by giving you a little shot of dopamine. Do that! It’s great! says your brain.

Here’s the problem. After a while, the pleasurable experience isn’t quite as pleasurable. You can only have the best steak of your life once. Have the same steak a second time, and it won’t feel quite as great. To get the same amount of dopamine–i.e. to experience the same level of pleasure–you now need an even better steak.

Uh oh.

Now you’re stuck on the treadmill involving an insatiable need for ever better steaks to feel just as good. You need ever greater amounts of dopamine to have the same pleasurable experience.

To borrow an illustration from Stanford neuroendocrinologist Robert Sapolsky, if a monkey presses a lever in exchange for a raisin, the monkey’s brain releases ten units of dopamine. (Pleasure! Reward! Press it again!)[25]

Next, add a second raisin.

But now, instead of releasing ten units of dopamine for one raisin, the monkey gets twenty units of dopamine for two raisins. (Twice the pleasure! Twice the reward! Press it again! Faster!)

Now the monkey runs into a problem. After a while–after lots of lever-pressing and lots of double-dose raisin entrees–the dopamine levels drop back to ten units, even though the monkey is still getting two raisins. What’s happened? The monkey has habituated to the new reality. Two raisins feels only as good as one raisin used to. To get more pleasure, the monkey now needs three raisins.

There’s another twist. If, instead of a monkey pressing a lever at random, the monkey now sees a light and learns to press a lever in response to the light–and then gets the raisin–pretty soon the light itself will actually trigger dopamine. And over more time, the light–the mere anticipation of the reward–will trigger more dopamine than the reward itself.

Researchers found that even more dopamine is released when the reward appears only some of the time. This is the scenario where the most dopamine is released, because, in the words of Sapolsky, “nothing fuels dopamine release like the ‘maybe’ of intermittent reinforcement.”[26]

loss aversion dopamine
source: Sapolsky, R. (2017). Behave: The Biology of Humans at Our Best and Worst. New York, NY: Penguin Press.

How dopamine works in the brain

Neurons are long, thin cells, lined end-to-end. When one end of a neuron is stimulated, a little message called an action potential travels down the length of the neuron to the other end.

There it stops. It’s reached the end of the neuron.

To get to the next neuron in the chain, the end of the neuron releases a little chemical to transfer the message to the next-door neuron.

Which neurotransmitter is released? It depends on the stimulus. If the stimulus is pleasurable, then the chemical that gets released is dopamine.

And when a neuron releases a little dopamine, it stimulates the next-door neuron, triggering the process all over again in that neuron. After it’s done that, the dopamine returns back to its neuron-of-origin. In other words, dopamine gets passed back and forth.

Here’s a little wrinkle: it’s not a simple as a neuron releasing a little dopamine, the dopamine sauntering over to the neuron next door, saying hi, and walking back.

That’s because the entry and exit points for the dopamine aren’t all the same.

There’s not one single dopamine door on your neurons. There are five. And depending which of the five doors your dopamine is knocking on, your neighbor neuron will react differently.

Each of these entry and exit points is a called a receptor. Since there are five, neuroscientists call them D1, D2, D3, D4, and D5 receptors, respectively.

The problem is that sometimes these receptors get blocked. Let’s say a neuron releases some dopamine to transmit an urgent message to the next cell. The dopamine exits the cell and heads next door, but—crap—the entry door is locked. That’s the D2 receptor. But there’s more and dopamine molecules exiting the original cell, getting stuck in the intermediate space between the two cells. This buildup of dopamine causes a pleasurable experience. There are lots of ways to break a D2 receptor. One of them is to take cocaine.

To summarize: your nervous system communicates using a mix of electrical and chemical signals. The chemical signals are called neurotransmitters, and the neurotransmitter that regulates feelings of pleasure—thereby reinforcing the behavior that caused it—is called dopamine. There are five specific entry and exit points at the ends of each neuron, and these are called receptors. If these are blocked, this increases dopamine production, making a pleasurable experience feel even more pleasurable.

Let’s return to our coin-toss scenario.

Remember, you get $10 if it’s heads, but you lose $10 if it’s tails—and most people don’t take the bet until the gains roughly double the losses. That’s the loss aversion ratio.

The question we’re trying to answer is this: why do some people value losses far more than gains, and why do others value them more equally? And how does this affect their behavior?

In part, someone’s ability to calculate a loss in a risky choice is affected by how much dopamine their brain is releasing.

That’s right: your behavior is controlled, in part, by how much of a little chemical is present in your brain.

More dopamine equals less loss aversion.

Lots of dopamine reduces your ability to predict losses, which, in turn, reduces your aversion to loss. That’s because when your brain evaluates a decision, the pleasure circuit in your brain seems to get more votes–it helps your brain see the upside of taking a risk and drowns out other parts of your neural circuitry that represent the downside.

When researchers gave healthy people a dose of a chemical that mimics dopamine,[27] they found they had a harder time learning in response to gains.

Then, when they gave people a dose of a chemical that’s the opposite of dopamine,[28] they found they could increase their ability to learn from gains.

In other words, lots of dopamine “releases the brakes” on how the brain reinforces experiences through feelings of pleasure, and dopamine “has an invigorating effect on reward seeking behavior.”[29]

This also explains why older people are less loss averse than younger people. As people age, their dopamine receptors begin to atrophy, especially in neurons connected to the striatum, where magnitude processing happens.[30]

This doesn’t mean you’ll take the bet where you could either win $10 or lose $10. It just means you’re more likely to take bets where the gains get closer to the losses in value. Your losses still loom larger than their equivalent gains, just not quite as much. You’re also more at risk of pathological behavior, such as compulsive gambling.

On the plus side, if you’ve got lots of extra dopamine, then you’ll be less likely to fall for marketing tricks that appeal to your sense of loss. While everyone else is scrambling to get that new couch before the sale ends, you’ll walk right on by the furniture store. Being rationally superior has its perks, but only to a point since you’re also marginally more at risk for heading to the casino instead and gambling away your life savings. Loss aversion isn’t always rational, and that’s a good thing, because it prevents some pretty risky behavior.

So, more dopamine leads to less loss aversion.

What causes more dopamine?

There are lots of reasons some people have more dopamine than others, and why your own dopamine levels may vary. But there are two in particular: the first is your baseline level of dopamine production, and the second is the ability of your dopamine receptors to function properly. Both factors influence how much dopamine you’ve got floating around your brain—which, in turn, influences how loss averse you’ll be.

What, then, regulates dopamine productions? And what regulates the effectiveness of your D2 receptors?

Your genes.

Let’s ask our question a different way: is loss aversion genetic?

The genetic origins of loss aversion

Before we get too deep into our discussion of how genes explain the differences in loss aversion between individuals, I need to clarify the extent to which genes dictate behavior.

This is an incredibly complicated topic, and I’m not going to try to explain it in full. For now, I’ll simply note that with any trait—physical, behavioral, or otherwise—you can attribute some amount to variation and some amount to the environment. One of the best way to figure out how much genes are responsible and how much the environment is responsible is by comparing identical twins who have been raised in different environments. Identical twins share the same genes. If two twins are separated at birth, and one twin grows up to become taller than the other, then there’s likely an environmental cause for the variation. Two molecular biologists from Tufts University studied this phenomenon. If someone is extra tall, is it because of their genes, or because of their environment? It turns out it’s a bit of both. They found that genes cause between 60 and 80 percent of the variation in height, while the environment—such as proper nutrition—is responsible for the rest.[31]

As you’ll see, there are genes that make some people less loss averse than others. But it would be a mistake to say a gene is solely responsible for this variance. In some cases, environmental changes can negate variance from a gene. In other cases, they can exacerbate. So, while it’s true certain forms of genetic variation can make some people more loss averse than others, it’s usually the result of, in the words of Robert Sapolsky, a “magnificent fascinating nuanced interaction between nature and nurture.”

With that out of the way, let’s dive in to the genetics of loss aversion.

There are two genes that we know of that affect loss aversion, both of which are tied to how the brain handles dopamine:

  1. A gene that controls BDNF, a chemical in the brain involved in dopamine production, and
  2. A gene that controls D2 receptors—remember, that’s a specific entry- and exit-point for dopamine passing chemical messages between neurons.

It’s not too surprising that genes related to dopamine can affect loss aversion, especially since we’ve seen how dopamine regulates the communication networks in the brain that deal with evaluating gains and losses.

What is strange is that when there’s an error on only one of the genes, nothing happens. It’s only when there’s an error on both of these genes do people become less loss averse.

Let’s take a closer look.

How genes work

Your body is comprised of around 100 trillion cells. In each of those cells is a nucleus. And inside every nucleus of every cell in your body are two sets of the genome—one from mom, the other from dad. In your genome are roughly 20,000 genes, which sit on twenty-three chromosomes. Each of your twenty-three chromosomes is a long molecule called DNA.

Next, there are four chemicals, called bases, that attach themselves to the sides of DNA molecules. These four bases are called adenine, cytosine, guanine, and thymine, or A, C, G, and T for short. The catch is each base can only be paired with one other base: A with T, G with C. This pairing produces a ladder, or the image of the famous double helix.

Three sets of letters are called codons. Or, if you’re looking at a real ladder in your garage, a codon would be three steps. Then another codon would start.

And the codons are interrupted by sequences called exons and introns, which denote the beginnings and ends of key sections of DNA.

Another way of thinking about this would be to imagine a book.[32] That book is your genome. In the book are twenty-three chapters called chromosomes. There are around 20,000 stores in the book called genes. In each story, there are paragraphs called exons, interrupted by introns, which indicate the beginnings and endings to each story. The words are made up of codons. Each word is written with three letters called bases.

Cool. You’ve got DNA. But how does your DNA get translated into a trait? If you’ve got a gene for blue eyes, how does that gene make you have blue eyes?

First—and I’m skipping a few in-between steps here—your DNA turns itself to RNA, which is kind of like half DNA–instead of two strands, it’s just one. But just like DNA, RNA uses the letters A, C, and G, but instead of T, it uses U. And just like DNA, these letters are grouped into words of three letters each.

Second, a ribosome, which is like a little machine in your cell, comes along and works down the length of the RNA from top to bottom.

Third, as it goes from top to bottom, this little ribosome looks at each codon—a three letter sequence of A, C, G, or U—and then turns that three-letter sequence into a different letter—a single letter. But it’s not just a random letter. Each specific set of three letters gets turned into one of twenty-three other letters. For example, AUG gets turned into M, and GUC gets turned into V. These twenty-three letters are called amino acids. (Remember when you were in school and you passed coded notes to evade discovery by the teacher? This is basically the same thing. The RNA writes a bunch of indecipherable letters, the ribosome decodes it, and the result is a bunch of amino acids your body can use.)

Fourth, these amino acids are strung together in the same order as on the RNA (and on the DNA). And the introns—the breaks between the paragraphs—tell the ribosome where to start and stop the amino acid code.

Fifth, after the ribosome has finished, the whole chain of amino acids forms a unique shape. Now it’s collectively called a protein. Proteins do the work of keeping you alive. They create antibodies to help you fend of foreign intrusion from bacteria and viruses, they create enzymes that dictate the chemical reactions in your cells, they create hormones that help regulate other functions of your body, and they literally create the structure of your body. And all of this happens because of the instructions your genes give.

dna rna amino acid protein
source: https://epicgenetic.wordpress.com/2014/06/08/gene-therapy-finally-a-viable-option/

The BDNF gene

One of those proteins is called BDNF, which stands for brain-derived neurotrophic factor. BDNF does lots of things. It helps the nervous system function at the best possible level. It promotes cell growth. It helps connect nerves to each other. It also opens up new pathways in your nervous system.[33] It’s really active when you’re trying to learn something or trying to remember something. And when things are going well, the BDNF protein acts as a control switch when you need a few more dopamine molecules.[34]

bdnf
source: Wikipedia

Remember that there are twenty-three chromosomes—or chapters—in the book that is your genome. On chromosome 11, there’s a gene that regulates the amount of BDNF in your brain.[35] This BDNF gene contains 11 exons. On exon 5, in the 66th codon, at position 196, the letter G is supposed to appear. But every once in awhile, the letter A appears instead of the letter G. Because of the letter swap, the codon reads ATG instead of GTG.

When there’s a single wrong letter, it’s called a single nucleotide polymorphism, or SNP for short.

Why would there be a wrong letter in the first place? Because each time your DNA is passed to your offspring, it needs to copy all 3 billion letters of itself, and in doing so, it usually makes about 100 mistakes each generation. Sometimes these errors are pretty bad—so bad the offspring won’t survive or won’t reproduce and then pass on the genes again. In those cases, these errors die out. But other times, the errors stick around, either because they’re beneficial, or because they’re neutral.

Remember that DNA encodes RNA, so when there’s a mistake that switched GTG to ATG, the mistake gets translated to the RNA, too: the RNA is now ATG instead of GTG as well.

And why does that matter? Because, as you’ll remember, each three-letter RNA codon is coded into a one-letter amino acid. So at this specific spot, because of this error, this RNA codon ATG gets encoded as a V amino acid instead of an M like it’s supposed to. (This error is called the val66met SNP, because it exchanges V, which stands for valine, for M, which stands for methionine, on codon 66.)

The string of amino acids starts and stops at each intron. When the RNA hits a “stop” intron, all the amino acids since the last intron fold in on themselves, which becomes the base of the protein. But because of the val66met SNP, one of those letters is wrong.

True, it’s a really tiny mistake. Let’s compare. When everything happens as it’s supposed to, the protein sequence looks like this. Each of these letters stands for a chemical, and I’ve highlighted the V at the 66th codon:

MTILFLTMVI SYFGCMKAAP MKEANIRGQG GLAYPGVRTH GTLESVNGPK AGSRGLTSLA
DTFEHVIEEL LDEDQKVRPN EENNKDADLY TSRVMLSSQV PLEPPLLFLL EEYKNYLDAA
NMSMRVRRHS DPARRGELSV CDSISEWVTA ADKKTAVDMS GGTVTVLEKV PVSKGQLKQY
FYETKCNPMG YTKEGCRGID KRHWNSQCRT TQSYVRALTM DSKKRIGWRF IRIDTSCVCT
LTIKRGR

But with this error—with people who have the val66met SNP, the protein sequence looks like this. I’ve now highlighted the M at the 66th codon:

MTILFLTMVI SYFGCMKAAP MKEANIRGQG GLAYPGVRTH GTLESVNGPK AGSRGLTSLA
DTFEHMIEEL LDEDQKVRPN EENNKDADLY TSRVMLSSQV PLEPPLLFLL EEYKNYLDAA
NMSMRVRRHS DPARRGELSV CDSISEWVTA ADKKTAVDMS GGTVTVLEKV PVSKGQLKQY
FYETKCNPMG YTKEGCRGID KRHWNSQCRT TQSYVRALTM DSKKRIGWRF IRIDTSCVCT
LTIKRGR

Bummer.

This (now incorrect) string of amino acids functions like the architectural plan for the BDNF protein. We’ve made one tweak to this sequence by swapping an M for a V at codon 66, so now we’re going to run into some trouble when the protein gets put together.

Remember that BDNF helps your nervous system perform at an optimal level: it promotes the growth of nerve endings, nerve connections, and new neural pathways. But with the protein coding mistake, we now have less-than-optimal BDNF functionality. Your neurons are producing less BDNF. And less BDNF has been linked with an increased risk for depression, schizophrenia, eating disorders, and numerous other problems.

People with the val66met BDNF gene are also less loss averse. That’s because, as I mentioned earlier, the BDNF gene regulates dopamine pathways in the brain. And dopamine levels are strongly tied with loss aversion. Lots of dopamine, lower loss aversion.

The BDNF gene also regulates dopamine pathways in the brain. Remember from our earlier discussion that if you have too much dopamine, you have a harder time predicting losses, which reduces loss aversion. In theory, people with a mutated BDNF gene should be less loss averse. A fascinating study by a team of scientists found this to be true.[36]

Of all the participants in their study, they found an average loss aversion ratio of 2.08. Or, for a coin toss gamble, a potential loss of $10 would have been valued the same as a potential gain of $20.80. But for people with the val66met SNP, however, the loss aversion ratio dropped to 1.88.

It might seem that a difference between 2.08 and 1.88 isn’t much. But it does indicate that something is going on with people who have the val66met SNP.

What was most surprising, however, was that when the val66met SNP combines with another error on a different gene, the change become much more pronounced. Let’s take a look.

The D2 receptor gene

We’ve seen that dopamine production is regulated by BDNF, but, just like there are lots of genes associated with how tall someone is, there are lots of genes associated with how much dopamine they have, too.

In our discussion of how dopamine works in the brain, I mentioned that there are five specific receptors at the ends of each neuron that allow dopamine to enter and exit neighboring cells. The brain’s ability to communicate is affected, in part, by how efficiently dopamine can enter and exit through these entry and exit points.

Let’s go back to chromosome 11, but go just a little further down from the BDNF gene. There, you’ll find a gene that regulates how efficiently one of these dopamine receptors transmits messages to neighboring cells—the second one, or the D2 receptor, to be more exact. This gene is called the DRD2/ANKK1.

Before we proceed, let’s decode this name:

  • DRD2 is the name of the gene where geneticists thought dopamine regulation happened. But after scientists mapped the human genome and then developed more sophisticated tools for finding such things, it turns out it actually wasn’t there.
  • ANKK1 is the name of the gene where dopamine regulation actually happens. It’s an easy mistake to make, because it’s just next door to the DRD2 gene. But since there’s so much published material that references the DRD2 gene, the old name lingers on, and everyone now seems to just call it the DRD2/ANKK1 gene if they’re talking about the mutation we’ll get to in just a second.

Variants of genes are called alleles. Since you have two copies of every gene—one copy from mom and one copy from dad—sometimes variants aren’t the same. If your mom has green eyes, the gene from her would code for green eyes. But if your dad has blue eyes, then the gene from him could code for blue eyes. In your DNA, then, you’ve got two variants: a green eye allele and a blue eye allele. Which wins? Whichever one is dominant.

The DRD2/ANKK1 gene comes with two alleles, too: one is called the A1 allele and the other is called the A2 allele. The A2 is normal, while the A1 allele is abnormal.

This A1 allele—the abnormal one—is called the Taq1A allele. Geneticists refer to this abnormal version as the DRD2/ANKK1 Taq1a.

If you’ve got the A1 allele, then the D2 receptors at the ends of your nerves—the entry and exit points for dopamine—don’t work as well. One group of researchers discovered a variance as high as 28% in the ability of D2 receptors to bind to dopamine.[37]

As a result, the brain can’t as effectively use dopamine as a feedback mechanism, especially as a metric for judging pleasure or understanding reward. This can lead to extreme or excessive behaviors.

Why does this matter? Because dopamine levels affect behavior. The brain uses dopamine (among other things) as a feedback mechanism–especially as a metric for judging pleasure and understanding reward. For people with the A1 allele, this mechanism is partially broken. They are worse at avoiding errors and more likely to suffer from alcoholism,[38] obesity,[39] and other problems. They need more drinks or more food to get the same feelings as pleasure as people with the A2 allele. When people with the A1 suffer from a brain injury, it takes them longer to recover.[40] People with the A1 allele are more likely to experience a greater high from cocaine, have a harder time overcoming an addiction, and are more likely to relapse if they do.[41] In general, the A1 allele disrupts people’s ability to process rewards, and use the feelings from this reward to modify future behavior.[42]

Interaction between BDNF val66met SNP and the DRD2/ANKK1 Taq1A variant

You guessed it: the DRD2/ANKK1 Taq1A makes people less loss averse, too.

Gesine Voig from the University of Bonn, along with her colleagues, discovered that in a sample population:

  • People had a mean loss aversion ratio of 2.08.
  • People who had only the BDNF val66met SNP had a loss aversion ratio of 1.88.
  • People who had only the DRD2/ANKK1 Taq1A variant had a loss aversion ratio of 2.19 (it went up!).
  • But people who had both the BDNF val66met SNP and the DRD2/ANKK1 Taq1A variant had a loss aversion ratio of 1.69.

In other words, people with both the BDNF val66met SNP and the DRD2/ANKK1 Taq1A allele were 18.75% more loss averse than people without both variants.[43]

To put it in coin-toss terms, people with no SNP or variant would accept a $10 loss in exchange for a gain of $20.80 or more. People with both the BDNF val66met SNP and the DRD2/ANKK1 Taq1A variant would accept a $10 loss in exchange for a $16.69 gain.

To summarize: there isn’t a loss aversion gene. But there are two genes related to dopamine regulation that can make someone less loss averse.

It’s worth pondering that if you are among the sizable minority of the population who has both the BDNF val66met SNP and the DRD2/ANKK1 Taq1A variant, then you are willing to accept a smaller gain than most people in exchange for a potential loss, and you are willing to do so not on accord of your own free will or from rational, calculated decision-making, but because of how two genetic variations affect dopamine levels in your brain—genetic variations which first occurred without your input or awareness hundreds of generations before you were born.

We’ve now determined that two genes (and possibly more) can contribute to loss aversion, and that variations of the genes make people less loss averse.

This raises two questions:

  1. Where did these genes come from?
  2. Why do they persist?

Let’s tackle each of these questions.

Origins of the BDNF val66met SNP

In 2004, Eiji Shimizu from Chiba University in Japan noted that the val66met SNP was found in 50.3% of the population of Japan, 43.2% of the population of Italy, and 27.1% of the population of the United States.[44] A later study by Tracey Petryshen and her colleagues from the Harvard Medical School found the val66met SNP in 19.9% of Europeans and 43.6% of Asians. They also noted that only 0.55% of populations from sub-Saharan Africa had the val66met SNP, and that it was “virtually absent” in Native American populations as well (except, strangely, a group of Native Americans in Arizona, where approximately 40% have it).[45]

From this basic data, we can infer that the val66met SNP originated sometime after the first homo sapiens migrated out of Africa 100,000 years ago. We know this because had it occurred before humans migrated out of Africa, we would see more instances of the SNP in Africa. This means the SNP originated sometime in the past 100,000 years.

Origins of the DRD2/ANKK1 Taq1a allele

The DRD2/ANKK1 Taq1a allele has a much older and more interesting history. In fact, this history predates modern humans.

Anatomically modern humans split from our closest hominid relatives, Neanderthals and Denisovans, roughly 400,000 to 600,000 years ago.[46] Then, roughly 390,000 years ago, Denisovans and Neanderthals separated from each other.[47] Before modern humans, Neanderthals, and Denisovans was homo erectus, the likely common ancestor of all of us. And before homo erectus was the last common ancestor of our closest living primate relatives, including chimps, gorillas, orangutans, baboons, and rhesus monkeys.

What other primate species have the Taq1a allele? It turns out our closest living cousins don’t. Chimps, gorillas, orangutans, baboons, and rhesus monkeys have a plain old ordinary ANKK1 gene.

But that’s not true of our closest non-living hominid cousins, Neanderthals and Denisovans. Of those we know about, all Neanderthals have the Taq1a allele, and some of the Denisovans probably did.

What does this tell us?

It means the Taq1a allele originated prior to the existence of modern humans, and at least as far back as the divergence between Neanderthals, Denisovans, and modern humans, roughly 400,000 to 600,000 years ago.[48]

(It also means Neanderthals, all of whom had the Taq1a allele, were probably less loss averse than humans.)

Why did these two variants get passed down?

Evolution is sometimes described as survival of the fittest. But that’s only half the story. The other half is that a mutation needs to confer a benefit that not only helps you survive, but also helps you find a mate—so the two of you can pass your genes to the next generation.

This doesn’t mean the benefit the val66met SNP conferred was the reason it spread so rapidly. For example, let’s say I have a gene that makes me stronger, smarter, and faster than everyone else. That’s a benefit. But let’s say I also have a gene that makes me live, on average, one year less than everyone else. That’s a bummer. But on balance, I still come out ahead: the gene that makes me stronger, smarter, and faster than everyone else makes me more likely to find a mate; the downside of having that lifespan-shortening gene doesn’t help, but it doesn’t hurt much either. As a result, I pass my genes to the next generation, thanks for the beneficial genes, and the lifespan-shortening gene hops along for the ride.

That could be what’s happening with the DRD2/ANKK1 Taq1a allele and the val66met SNP. In fact, it’s almost certain it is. Remember that these genes are both located on the same chromosome, not far from each other. They’re part of genes on chromosome 11 called the NTAD cluster which help regulate dopamine production and reward processing. The acronym NTAD is formed from the first letters of the four genes on the cluster: NCAM1, TTC12, ANKK1, and DRD2. Together, these genes work together in some pretty interesting ways. For example, one of the things the ANKK1 gene does is encode a protein–a set of instructions–for how a different gene in the next-door DRD2 gene gets expressed. It’s a circular connection that makes it more likely for these genes to get passed to the next generation, warts and all.[49]

What this means is that an SNP, such as the val66met SNP, or the DRD2/ANKK1 Taq1a allele, can more easily hop along for the ride as the NTAD cluster gets passed to offspring either as a unit or not at all. The cumulative benefits of the gene cluster outweigh the cumulate impacts of the SNPs in the clusters. This means SNPs persist a great deal longer in a gene cluster than in a singular gene, and perhaps why the two SNPs that make people less loss averse have persisted so long, too.

It’s also possible that SNPs that make someone less loss averse confer a benefit we’re not aware of. This happens with lots of genes. In the 1940s, Anthony Allison, an Oxford student working in Kenya, noticed that people with sickle cell disease were resistant to malaria. This resistance is caused by a mutation on chromosome 11. In parts of the world where the risk of getting malaria is high, this mutation confers a benefit. The risk of sickle cell disease is a small price to pay in exchange for getting malaria. But in parts of the world where there’s no risk of getting malaria—Michigan, for example—the mutation that causes sickle cell disease comes with too high a cost. There’s no longer any benefit. So the mutation gets passed down in regions of the world where there’s malaria, but it’s deselected in areas where there’s no malaria–where the costs outweigh the benefits.

There’s another well-known example on chromosome 7. A mutation of the CFTR gene causes cystic fibrosis, but it also makes someone less susceptible to typhoid. Both are dangerous, but the mutation persists, because, across a population, protection against typhoid confers slightly more benefit than protection against cystic fibrosis.

The two genetic mutations that make people less loss averse could work the same way. While these mutations might cause you to make slightly worse decisions, they might also confer an unknown benefit that makes up for it.

Is loss aversion rational?

So far, I’ve been assuming that loss aversion is not only irrational, but it also bad. But is that really true?

On the one hand, if you’re constantly making incorrect decisions, even when the odds in your favor, then that’s bad. If you have a chance of gaining $11, but there’s a possibility you’ll lose $10 instead, then it would be irrational to pass on the bet. Since most people do pass on that bet, most people are irrational, and that’s bad.

On the other hand, it’s also possible you take take a bet where the odds are in your favor and lose–and thereby lose the ability to continue betting. If you bet the house and lose your house, then what? No more betting.

In fact, if you make any decision where one possible outcome could restrict your ability to make sound decisions in the future (or to ever make a decision again), then you should not take a bet, even when the odds are in your favor.

When you put it that way, loss aversion starts to sound pretty rational.

This gets at a fundamental aspect of human decision-making: decisions aren’t made in a vacuum. They’re made through time, in a series, one after another. Each decision affects your ability to make your next decision. Nassim Nicholas Taleb presented a clever illustration of this in his book Skin in the Game: Hidden Asymmetries in Daily Life. Imagine you are asked to buy insurance against a 1 percent probability of losing $10,000. What would you pay? Generally, people would pay more than $100, which, in the technical sense, is irrational.

Taleb writes that calling this kind of behavior irrational is unfair, because “you cannot possibly ignore all the other financial risks he is taking: if he has a car parked outside that can be scratched, if he has a financial portfolio that can lose money, if he has a bakery that may risk a fine, if he has a child in college who may cost unexpectedly more, if he can be laid off, if he may be unexpectedly ill in the future.”[50]

If you lose a bet where the odds are in your favor, your loss still affects the kind of bet you make in the future. And if you string a bunch of these bets together, if you make only a few more bad bets than good ones, this will restrict your ability to make any bets in the future.

What’s more, even though I’ve been using a coin toss to illustrate loss aversion, most decisions come with unknown probabilities; life isn’t just a series of 50/50 coin toss games. Often you don’t know the probability, or the probability of the next decision either.

It’s hard to fathom that until the past century or so, most of the world lived in fear of constant famine, or disease, or massacre from rival tribes.

People who weren’t loss averse might have done well in the short term. But it only takes one mistake, or one unpredicted invasion, or one instance of the plague before none of it matters.

Because you’re dead.

The people who are loss averse might be making irrational decisions when those decisions are viewed in isolation. But those people happen to be more likely to stick around long enough to make another decision tomorrow.

In this sense, loss aversion protects our ability to make good decisions in the long run, even if it sometimes costs us in the short run: in the form of passing on opportunities that could benefit us, or by making us fall victim to marketers who take advantage of our aversion to loss.

It’s impossible to judge whether loss aversion is good or bad, rational or irrational. As with so many things, the answer is simply: it depends. Still, however you judge loss aversion, it’s the lever on our behavior that keeps us from losing everything. Loss aversion is what hedges against unforeseeable shocks to lives that are complex and an existence that is unpredictable.

Get more articles about cognitive biases, consumer psychology, and human behavior.

Sign up (it’s free!)


[1] Park, C., Jun, S., and Macinnis, J. (2000). “Choosing what I want versus rejecting what I do not want: An application of decision framing to product option choice decisions.” Journal of Marketing Research 37(2), 187-202.

[2] Genesove, D. and Mayer, C. (2001). “Loss Aversion and Seller Behavior: Evidence from the Housing Market.” The Quarterly Journal of Economics, Volume 116, Issue 4, 1233–1260.

[3] Tversky, A. & Kahneman, D. (1992). “Advances in prospect theory: Cumulative representation of uncertainty.” Journal of Risk and Uncertainty 5(4): 297–323.

[4] Walasek, L., Mullett, T., and Stewart, N. (2018). “A meta-analysis of loss aversion in risky contexts.” SSRN.

[5] Wang, M., Rieger, M., and Hens, T. (2017). “The impact of culture on loss aversion.” Journal of Behavioral Decision Making 30(2), 270–281.

[6] Bontempo, R. N., Bottom, W. P., & Weber, E. U. (1997). “Cross-cultural differences in risk perception: A model-based approach.” Risk Analysis, 17(4), 479–488.

[7] Wang, M., Rieger, M., and Hens, T. (2017). “The impact of culture on loss aversion.” Journal of Behavioral Decision Making 30(2), 270–281.

[8] Hofstede, G. (2011). “Dimensionalizing cultures: The Hofstede Model in context.” Online Readings in Psychology and Culture, 2(1).

[9] Eckel, C. and Grossman, P. (2008). “Differences in the economic decisions of men and women: Experimental evidence.” Handbook of experimental economics results vol. 1, C. Plott, V. Smith, eds., New York, Elsevier, 509–519.

[10] Schulreich S., Gerhardt H., and Heekeren H.R. (2016). “Incidental fear cues increase monetary loss aversion.” Emotion 16(3):402–12.

[11] Foellmi, R., Jaeggi, A., and Rosenblatt-Wisch, R. (2018). “Loss aversion at the aggregate level across countries and its relation to economic fundamentals,” Working Papers 2018-01, Swiss National Bank.

[12] Inesi, M. (2010). “Power and Loss Aversion.” Organizational Behavior and Human Decision Processes, 112, 58–69.

[13] Gächter S., Johnson E.J., and Herrmann A. (2007). “Individual-Level loss aversion in riskless and risky choices.” IZA Discussion Paper No 2961.

[14] Tanaka, T., Camerer, C., and Nguyen, Q. (2010). “Risk and time preferences: Linking experimental and household survey data from Vietnam.” American Economic Review, 100 (1), 557–71.

[15] Liu, E. (2013). “Time to change what to sow: Risk preferences and technology adoption decisions of cotton farmers in China.” Review of Economics and Statistics 95(4), 1386–1403.

[16] Schulreich S., Gerhardt H., and Heekeren H.R. (2016). “Incidental fear cues increase monetary loss aversion.” Emotion 16(3):402–12.

[17] De Martino, B., Camerer, C. F., & Adolphs, R. (2010). “Amygdala damage eliminates monetary loss aversion.” Proceedings of the National Academy of Sciences of the United States of America, 107(8), 3788–3792.

[18] Canessa, N., Crespi, C., Baud-Bovy, G., Dodich, A., Falini, A., Antonellis, G., and Cappa, S.F. (2017). “Neural markers of loss aversion in resting-state brain activity.” Neuroimage (146), 257–265.

[19] Sapolsky, R. (2017). Behave: The Biology of Humans at Our Best and Worst. New York, NY: Penguin Press.

[20] Canessa, N., Crespi, C., Motterlini, M., Baud-Bovy, G., Chierchia, G., Pantaleo, G., Tettamanti, M., Cappa, S.F. (2013). “The functional and structural neural basis of individual differences in loss aversion.” Journal of Neuroscience 33(36) 14,307–14,317.

[21] Sokol-Hessner P., Camerer C.F., Phelps E.A. (2013). “Emotion regulation reduces loss aversion and decreases amygdala responses to losses.” Social Cognitive and Affective Neuroscience 8(3), 341–50.

[22] Canessa, N., Crespi, C., Baud-Bovy, G., Dodich, A., Falini, A., Antonellis, G., and Cappa, S.F. (2017). “Neural markers of loss aversion in resting-state brain activity.” Neuroimage (146), 257–265.

[23] Canessa, N., Crespi, C., Motterlini, M., Baud-Bovy, G., Chierchia, G., Pantaleo, G., Tettamanti, M., Cappa, S.F. (2013). “The functional and structural neural basis of individual differences in loss aversion.” Journal of Neuroscience 33(36) 14,307–14,317.

[24] Linden, David J. (2011). The compass of pleasure: How our brains make fatty foods, orgasm, exercise, marijuana, generosity, vodka, learning, and gambling feel so good. New York, NY: Penguin Press.

[25] Sapolsky, R. (2017). Behave: The Biology of Humans at Our Best and Worst. New York, NY: Penguin Press.

[26] Sapolsky, R. (2017). Behave: The Biology of Humans at Our Best and Worst. New York, NY: Penguin Press.

[27] The technical term for this is a dopamine agonist

[28] The technical term for this is a dopamine antagonist.

[29] Clark, C. A., & Dagher, A. (2014). “The role of dopamine in risk taking: a specific look at Parkinson’s disease and gambling.” Frontiers in Behavioral Neuroscience, 8, 196.

[30] Samanez-Larkin, G. R., Gibbs, S. E. B., Khanna, K., Nielsen, L., Carstensen, L. L., & Knutson, B. (2007). “Anticipation of monetary gain but not loss in healthy older adults.” Nature Neuroscience, 10(6), 787–791.

[31] Lai, C.Q. (2006). “How much of human height is genetic and how much is due to nutrition?” Scientific American, December.

[32] Ridley, M. (2006). Genome: The Autobiography of a Species in 23 Chapters. New York, NY: Harper Perennial.

[33] The fancy word for this is synaptogenesis—which means it generates new synapses, the connection points between the ends of two nerve cells.

[34] In case you’re wondering, brain-derived neurotropic factor is one of four kinds of neurotropic factors, called neurotrophins. The other three are nerve growth factor, neurotrophin-3, and neurotrophin-4.

[35] Pruunsild, P., Kazantseva, A., Aid, T., Palm, K., Timmusk, T. “Dissecting the human BDNF locus: bidirectional transcription, complex splicing, and multiple promoters.” Genomics 90: 397-406, 2007.

[36] Voigt, G., Montag, C., Markett, S., & Reuter, M. (2015). “On the genetics of loss aversion: An interaction effect of BDNF Val66Met and DRD2/ANKK1 Taq1a.” Behavioral Neuroscience, 129(6), 801-811.

[37] Gluskin, B. S., & Mickey, B. J. (2016). “Genetic variation and dopamine D2 receptor availability: a systematic review and meta-analysis of human in vivo molecular imaging studies.” Translational Psychiatry, 6(3), 747.

[38] Olfson, E., & Bierut, L. J. (2012). “Convergence of GWA and candidate gene studies for alcoholism.” Alcoholism, Clinical and Experimental Research, 36(12), 2086–2094.

[39] “Genetics May Dull Brain’s Pleasure Response to Food, Causing Weight Gain,” 23andme blog.

[40] McAllister, T. W., Flashman, L. A., Rhodes, C. H., Tyler, A. L., Moore, J. H., Saykin, A. J., … Tsongalis, G. J. (2008). “Single Nucleotide Polymorphisms in ANKK1 and the Dopamine D2 Receptor Gene Affect Cognitive Outcome Shortly After Traumatic Brain Injury: A Replication and Extension Study.” Brain Injury 22(9), 705–714.

[41] Spellicy, C. J., Harding, M. J., Hamon, S. C., Mahoney, J. J., Reyes, J. A., Kosten, T. R., Nielsen, D. A. (2014). “A variant in ANKK1 modulates acute subjective effects of cocaine: a preliminary study.” Genes, Brain, and Behavior, 13(6), 559–564.

[42] Yacubian, J., Sommer, T., Schroeder, K., Gläscher, J., Kalisch, R., Leuenberger, B., Büchel, C. (2007). “Gene–gene interaction associated with neural reward sensitivity.” Proceedings of the National Academy of Sciences of the United States of America, 104(19), 8125–8130.

[43] Voigt, G., Montag, C., Markett, S., & Reuter, M. (2015). “On the genetics of loss aversion: An interaction effect of BDNF Val66Met and DRD2/ANKK1 Taq1a.” Behavioral Neuroscience, 129(6), 801-811.

[44] Shimizu, E. (2003). “Ethnic difference of the BDNF 196G/A (val66met) polymorphism frequencies: The possibility to explain ethnic mental traits.” American Journal of Medical Genetics 126B: 122–123.

[45] Petryshen, T., Sabeti, P., Aldinger, K., Fry, B., Fan, J., Schaffner, S., Sklar, P. (2010). “Population genetic study of the brain-derived neurotrophic factor (BDNF) gene.” Molecular Psychiatry, 15(8), 810–815.

[46] Scally, A. and Durbin, R. (2012). “Revising the human mutation rate: implications for understanding human evolution.” Nature Reviews Genetics 13, 745.

[47] Viviane Slon, Fabrizio Mafessoni, Benjamin Vernot, Cesare de Filippo, Steffi Grote, Bence Viola, Mateja Hajdinjak, Stéphane Peyrégne, Sarah Nagel, Samantha Brown, Katerina Douka, Tom Higham, Maxim B. Kozlikin, Michael V. Shunkov, Anatoly P. Derevianko, Janet Kelso, Matthias Meyer, Kay Prüfer & Svante Pääbo (2018). “The genome of the offspring of a Neanderthal mother and a Denisovan father.” Nature 561, 113–116.

[48] Mota, N. R., Araujo-Jnr, E. V., Paixão-Côrtes, V. R., Bortolini, M. C., & Bau, C. H. D. (2012). “Linking dopamine neurotransmission and neurogenesis: The evolutionary history of the NTAD (NCAM1-TTC12-ANKK1-DRD2) gene cluster.” Genetics and Molecular Biology, 35(4 Suppl), 912–918.

[49] Mota, N. R., Araujo-Jnr, E. V., Paixão-Côrtes, V. R., Bortolini, M. C., & Bau, C. H. D. (2012). “Linking dopamine neurotransmission and neurogenesis: The evolutionary history of the NTAD (NCAM1-TTC12-ANKK1-DRD2) gene cluster.” Genetics and Molecular Biology, 35(4 Suppl), 912–918.

[50] Taleb, N. (2018). Skin in the Game: The Hidden Asymmetries in Everyday Life. New York, NY: Random House.

Advertisements