The Currency of Death: A Field Guide to Micromorts

15 Mar, 2026

A marathon will kill you about as fast as a skydive. A day in hospital is like motorcycling from London to Edinburgh. Giving birth in the US is twice as deadly as in the UK. And climbing Everest costs roughly one micromort per foot of elevation.

These comparisons sound unhinged. They are also precise.

The Missing Unit

Money has units. Energy has units. Temperature has units. Risk, for most of human history, had nothing -- just a soup of vague adjectives ("low risk," "dangerous," "pretty safe") that made comparing hazards across domains essentially impossible. Is surgery riskier than skydiving? Is cycling more dangerous than driving? Without a common denominator, these questions don't even parse.

In 1979, Ronald Howard, a decision analyst at Stanford, proposed the missing unit: the micromort. One micromort equals a one-in-a-million chance of death. That's it. A probability small enough to fit most everyday risks, large enough to produce whole numbers you can compare at a glance. Howard's core insight was almost embarrassingly simple: give people a unit, and suddenly they can think.

He published the concept formally in 1989 as "Microrisks for Medical Decision Analysis," arguing that doctors and patients needed a common language for surgical risk. But the unit turned out to be far more general than medicine. It works anywhere death is a possible outcome -- which is to say, everywhere.

The Tables

A micromort needs an exposure to accompany it: per trip, per mile, per day, per event. With that caveat, here is a tour of mortality.

How You Move

Motorcycling is 41 times riskier per kilometer than driving. Driving is 27 times riskier than flying. The safest distance-normalized mode of transport is the commercial jet by a wide margin. The most dangerous is one that most jurisdictions let sixteen-year-olds operate.

How You Play

A marathon runner who refuses to skydive "because it's too dangerous" has no statistical basis for that fear -- the numbers are nearly identical. Base jumping, at 430 micromorts per jump, is roughly 50 times more dangerous than skydiving and in a different category entirely. And Everest -- at about 1 micromort per foot of elevation -- sits in a zone that most people cannot intuitively process at all.

An Australian study by Keage and Loetscher (2018) asked 284 people to rate 20 activities on a risk scale. Participants rated Everest as about 9 times riskier than a 27-km walk. The actual ratio is 38,000 to 1.

How Medicine Touches You

Going under general anesthesia -- just the anesthesia, not the operation -- carries the same risk as a single skydive. A day in hospital exposes you to 75 micromorts, equivalent to riding a motorcycle about 675 kilometers. And giving birth in the United States is nearly twice as risky as in the United Kingdom, a gap that compresses an entire public health failure into a single number.

Howard's Original "One Micromort" List

Ronald Howard compiled a list of activities that each cost approximately one micromort. It remains one of the most striking risk comparison tables ever assembled:

The nuclear plant line tends to get a reaction. You would need to live within 20 miles of a nuclear power plant for 150 years to accumulate a single micromort. By contrast, driving to visit that plant -- say, a 370 km round trip -- costs you the same micromort in an afternoon.

The Background Hum

Just being alive carries a daily micromort cost that rises exponentially with age, doubling roughly every eight years after childhood (the Gompertz law):

At 20, you spend about 1 micromort per day on the background risk of being human. At 90, you spend 450. The meter runs faster and faster.

The Extremes

Living in New York City during the worst weeks of spring 2020 carried a daily risk comparable to being a US soldier in Afghanistan. A full tour of duty in Helmand province accumulated roughly 6,000--8,500 micromorts. A Space Shuttle astronaut absorbed about 15,000 in a single flight. RAF bomber crews over Germany faced 25,000 per mission -- and were expected to fly 30.

But These Numbers Lie

Here is where the micromort framework breaks. Smoking 1.4 cigarettes costs one micromort. That is a true statement. It is also profoundly misleading.

One micromort is the acute risk -- the chance that this specific cigarette kills you right now, today, through some immediate event. By that measure, smoking is less dangerous than getting out of bed at age 45 (about 6 micromorts from baseline mortality alone). A naive reading of the micromort table would suggest that cigarettes are among the safest items on the list.

And yet smoking kills roughly 8 million people per year.

The problem is that micromorts measure only the gun that might fire right now. They are blind to the slow poison -- the cumulative damage that shortens life without ever producing a single dramatic moment of risk. Smoking doesn't kill you on any particular day. It kills you over decades, shaving minutes off each end of your life until the total is measured in years.

This is not a minor footnote. It is a fundamental structural limitation. There are two modes of dying, and the micromort only sees one of them:

• Acute risk: the skydive, the surgery, the motorcycle ride. A discrete event with a yes/no outcome. You survive or you don't. The micromort handles this perfectly.
• Chronic risk: the cigarette habit, the sedentary lifestyle, the air pollution. A slow erosion of life expectancy with no single identifiable moment of danger. The micromort is nearly useless here.

The Microlife

David Spiegelhalter, a Cambridge statistician who spent a decade as the Winton Professor for the Public Understanding of Risk, recognized this gap and in 2012 proposed a companion unit: the microlife. One microlife equals 30 minutes of life expectancy.

A young adult has roughly one million half-hours of life remaining -- hence one million microlives. Every day, 48 of them vanish just from the passage of time (aging). But habits can accelerate or slow the clock:

Now the cigarette's true cost becomes visible. Not 0.2 micromorts of acute risk, but half a microlife -- 15 minutes of life expectancy -- per cigarette. A pack-a-day habit costs roughly 10 microlives daily, which accumulates to about 8 years over a lifetime.

The distinction is clean and structurally important:

Together, the two units form a complete vocabulary for mortality risk. Micromorts for the acute. Microlives for the chronic. Two lenses on the same phenomenon, each blind where the other sees clearly.

The Price of a Micromort

One final puzzle. When researchers infer from everyday behavior how much people implicitly pay to avoid one micromort -- through car safety features, slightly safer routes, protective equipment -- the figure comes out to roughly $50. Meanwhile, the US Department of Transportation, using its Value of Statistical Life ($6.2 million), prices a micromort at $6.20. The UK's Department for Transport arrives at about £1.60.

There is a tenfold gap between what individuals reveal through their choices and what governments budget in policy. Whether this means people are irrationally cautious or governments are irrationally cheap depends on your priors. What the micromort gives you is the ability to see the gap at all.

Howard put it best in his original paper:

Precision in language permits the soundness of thought that produces clarity of action and peace of mind.

Most of us will never need to calculate micromorts. But the framework changes how you see risk once you've encountered it. The numbers are everywhere, quietly running in the background of every trip, every meal, every medical decision. The meter is always on.

Links: Microrisks for Medical Decision Analysis (Howard, 1989) | Small but Lethal (Plus Maths / Spiegelhalter) | Microlives (Spiegelhalter, BMJ 2012) | The Power of the MicroMort (Spiegelhalter, BJOG 2014) | Estimating Everyday Risk (Keage & Loetscher, 2018) | Micromorts Interactive (micromorts.rip)

#decision-theory #health #mathematics #risk #statistics