Issue 070 - War and demography - Mortality rate and life-years
What does a wartime death toll mean demographically?
Recent Russia-Ukraine war coverage has continued to report deaths and casualties in ways that are hard to interpret without a demographic baseline. Consider a hypothetical country of 40 million people experiencing 100,000 direct war deaths in one year.
Public systems and economicsAbout 1 minute
Sources checked October 6, 2026. Figures and circumstances may have changed.
The problem
In a 40-million-person country, about what share of normal annual mortality would 100,000 direct war deaths in one year represent?
Then estimate the years of life lost and how the answer changes if the deaths are concentrated among young adults.
Because Fermi problems target an order of magnitude, I normally use no more than two significant digits and write most calculations in scientific notation; the Fermi reference explains both conventions.
Grounding facts
The useful baseline is the crude death rate: deaths per year per 1,000 people. For a Ukraine-like demographic comparison, World Bank data show Ukraine's 2019 crude death rate at about 13 deaths per 1,000 people, before the pandemic and full-scale invasion.
World Bank data also put Ukraine's 2021 life expectancy near 72 years, with later values in the low-to-mid 70s. Years of life lost are calculated by multiplying deaths by the expected remaining years at the age of death.
As a separate comparison, the U.S. Census Bureau described U.S. deaths increasing by about 19% from 2019 to 2020 during the COVID-19 pandemic.
After checking sources
Checked answer and calculation
Using a rounded crude death rate of 13 per 1,000 people:
ordinary annual deaths
~= 40,000,000 x 13 / 1,000
~= 5.2 x 10^5 deaths/year
~= about 500,000 deaths/yearThen compare the hypothetical war deaths with that baseline:
war deaths as share of ordinary deaths
~= 100,000 / 520,000
~= 0.19
~= about 20%So 100,000 direct deaths would not double a country's normal annual mortality, but it would be a huge shock: roughly a one-fifth increase over normal yearly deaths.
For years of life lost, the age distribution matters more than the count alone. If those deaths followed the country's usual death pattern, many would occur among older people and the average remaining life might be only around 10 to 20 years:
typical-age life-years lost
~= 100,000 x 10 to 20 years
~= 1 x 10^6 to 2 x 10^6 yearsIf the deaths are concentrated among young adults around age 30, and a rough life expectancy is in the 70s, the average remaining life could be around 40 to 45 years:
young-adult life-years lost
~= 100,000 x 40 to 45 years
~= 4 x 10^6 to 4.5 x 10^6 yearsThe death count looks like a roughly 20% mortality shock. The life-years calculation makes it feel larger when the victims are young, because the same number of deaths can remove several million additional years of human life.
Before checking sources
Matt's original estimate
This is the unverified estimate Matt wrote before checking sources, not the checked answer.
I assumed the average life span of about 85, and I assumed I could roughly approximate the expected annual deaths would be 1/85th of the population, given an even distribution of people across those ages, which probably isn't accurate. That gives about 470,000 expected deaths each year for a population of 40 million.
100,000 excess deaths is about 21% of that number, so it would be 21% greater than a typical year. That's less than the bump we saw during COVID, I think. Next, I expect the war casualties average age would skew younger, maybe about 30 years old, so each death is approximately 55 years of lost life potential, or 5.5 million years of lost life. That feels much more tragic.
Reasoning score
Matt's reasoning score: 90 / 100
Higher is better: earn points for useful facts, a sound reasoning approach, correct math, and a final estimate close to the sourced answer. The owl meter shows percent full of it: 100 minus the reasoning score.
Useful facts: 20/30. The death-rate shortcut landed close to the sourced mortality baseline, though the assumed 85-year average lifespan was high for the Ukraine-like comparison.
Reasoning approach: 30/30. Comparing deaths with ordinary annual mortality, then switching to years of life lost for age structure, was the right model.
Math: 10/10. The arithmetic was clean.
Final estimate: 30/30. The 21% mortality-shock estimate was almost exactly the checked answer, and the young-adult years-lost estimate was in the right range.
Post-check reflection
Matt's reflection
Looks like this was pretty close, and the COVID comparison was closer than I anticipated as well. The effect that the hypothetical war has had in terms of deaths is similar to the effect that COVID had in the US, as a percentage of population, or a percentage in growth in annual deaths.
Recommended memory peg
For a country with an older population, a crude death rate around 10 to 15 deaths per 1,000 people per year is a useful baseline. In a 40-million-person country, that means roughly half a million ordinary deaths per year.
Reader results
Bars show how submitted estimates sort into the answer choices from the gut-check prompt.