In theory, both risk aversion and value of life year relative to income are mainly set by utility function concavity. However, as both also show puzzlingly large empirical variation, there is apparently a lot more going on than a simple context independent utility function. But what?

While pondering this question, I saw some nature videos of cute young mammals, who start out as physically and emotionally fragile, and prone to crying, but who grow to be tough and emotionally steady like their parents. I wondered: might it be useful to frame this as a transition from risk-aversion to risk-tolerance?

Of course youngsters aren’t always risk averse. Sometimes they try to play with a scorpion, or rough-house too close to a cliff. In such cases, mom may restrain their risk-taking. But when kids suffer some mild depredation, such as wanting food or other help from mom, that’s when they act most emotionally needly and fragile:

Oh my, this problem I face is way too hard for my fragile emotions. I’m young, innocent, and nearly dying of fright to think about it. Won’t someone who loves me come to protect me from this terrible anxiety and suffering?

Plausibly parents are built to find it hard to resist wanting to help such puppy-dog eyes, to get them to take care of their kids. And when parents do tend to help when kids suffer, that actually makes kids more risk-tolerant regarding choices to take precautions to prevent future suffering.

Humans are famously neotenous, retaining more childlike features further into adulthood. And living in large social groups that share food and other resources, even human adults have incentives to show puppy-dog eyes, and to feel sympathetic to those eyes when they have something to share. So it seems plausible that human nature would have adapted the fragile-tough child-adult dynamic to apply to the helpee-helper relations in groups of adults.

That is, I’m suggesting that evolution has built we humans to strategically (if unconsciously) make a key attitude choice regarding each particular situation we face: do we project a cool tough self-reliant risk-tolerance, or a more emotionally-expressive risk-averse fragile vulnerability. In other words, do we choose to act vulnerable but sympathetic, or do we choose to act more “emotionally mature”? I’m not saying this is the only factor that influences risk-tolerance, but it may be one of the largest.

Yes, people often talk as if emotional maturity is the product of learning from experience, with perhaps some help from an admirable moral will. But if emotional maturity were always the better strategy, why would evolution have ever encoded in us any other tendency? Evolution could have made the very young take on emotionally mature attitudes if it had wanted. And in fact, it does sometimes want that, such as when kids “grow up too fast” due to getting little help in taking on adult-sized problems.

The key maturity choice here is of whether to “own” each problem that we face. It makes sense to own a problem, and act more risk-tolerant toward it (and more risk-averse re preventing it), when we seek to impress others with our confidence in handling the problem, when we bid for parent/leader roles, and when we want to avoid the embarrassment of others seeing that no one comes when we ask for help. However, when getting help with our problem seems more important, and likely enough, or when showing submission to and dependence on others seems important, it can make more sense to try to get others to own our problem by acting more risk-averse and hurt by the problem. We can suggest that others are to blame for causing our problem, and even if not they are responsible to help fix it.

So does my theory fit the data? In one key study, “Psychosocial maturity proved a better predictor of risk-taking behaviour than age.” Which is striking because age (at least below age 65) is one of our two usual best predictors of risk-tolerance, the other being male gender. Some data suggests that the following groups are also more risk-tolerant: the tall, the married, and those with more kids. People who work in finance, insurance, and real estate seem more risk-averse, and buying insurance on a risk is a strong sign that one is averse to it. Those with mid-level wages and extreme high or low wealth also seem more risk tolerant. The self-employed seem more risk averse.

Emotional maturity tends to increase on average with age, range of experiences, intelligence, self-esteem, and work performance, positive attitudes toward childcare, and not being an orphan. Here are some descriptions of the concept:

An emotionally mature individual gives off a sense of ‘calm amid the storm.’ They’re the ones we look to when going through a difficult time because they perform well under stress. … When you’re less mature, the world is full of minor annoyances, and you’re unaware of your own privileges. Think about how often a day you complain about others or different situations. (more)

Accept[s] the sorrows of life whole heartedly and … show[s] distress when there is occasion to be worried, without feeling a requisite to use a false facade of bravery. (more)

If my theory is right, and much of the variation in risk aversion (and value of life) we see results from strategic context-dependent choices to act risk-tolerant or risk-averse, this makes it harder to used measured risk-aversion (and value of life) to inform policy. Yes, if true risk aversion were higher, that would justify paying more to save lives, including via stricter regulation, and also justify more redistribution and social insurance. But if much of what we see people do is done for show, then we have to try to infer the real level of risk aversion behind all that show.

My guess is that on average in a social species with lots of sharing, free-riding is a bigger problem than excess autonomy. If so, we more often try to seem more needy to gain more help, than we try to seem less needy to gain respect. And thus typical behavior will exaggerate our real overall degree of risk aversion (and value of life). But I don’t yet know how to show this. It does seem worth further study; we may well figure out some way to see.

Added 3p: Related datum: “women are more sensitive to pain and less tolerant of pain than men.”