What is the economic value of a human life?

If there are multiple methods to use in calculating, what are they, which method is preferred, and why?

Considerations/sub-questions on methods

A. Is the calculation purely based on output value (i.e. wages, goods produced, etc.), or are there intangibles that must be included (value of innovations that cause progress for society, NPV of progeny, etc.)?

B. As a result of #1, are there things which reduce economic value in one's life? For example, when one has gone on unemployment, is one's net economic value negative for that period of time?

C. Is the economic value of life dependent on one's country/state/locale or the sector one performs in? For example, is it correct to say the economic value of a life is different in the US than in South Africa (due to average output in real terms?

  • 1
    $\begingroup$ Could you define economic value? There are several interpretations (not methods) because lives are not traded, so it is not obvious who the evaluators (buyers and sellers, payers and receivers) would be. (You, your family, the state, society, etc.) $\endgroup$ – Giskard Nov 23 '15 at 7:18
  • 1
    $\begingroup$ That's a good point. Is that clarification really necessary from the question side though? Or would that clarification perhaps make the rest of the answer somewhat trivial? I'm thinking it over and trying to decide if that is the real question in the first place (and the rest follow from that decision)... $\endgroup$ – LightCC Nov 23 '15 at 18:17
  • 1
    $\begingroup$ Suppose I ask you the value of a car. There are probably several methods to determine its market value. And this should not be different from what you are willing to pay for it. If your subjective valuation was lower, you could still sell the car afterwards, and if it was higher, this is still the lowest price on the market so it's a good purchase. But the whole subjective valuation is different if I ask you to sell me your car that you love. You might go higher than the 'objective market' price, as your car has no perfect substitute. Subjective valuation of a life is even more difficult. $\endgroup$ – Giskard Nov 23 '15 at 18:45
  • 1
    $\begingroup$ In the environmental economics literature, they conduct "contingent valuation studies" that seek to determine how to monetize various non-market values, including the dollar value of life. Here is a simple reference. $\endgroup$ – emeryville Nov 24 '15 at 1:42
  • $\begingroup$ here is a survey analysis of studies which estimate the value of life: mdm.sagepub.com/content/20/3/332.short From there, you should be able to find more references. $\endgroup$ – HRSE Nov 25 '15 at 10:08

It depends on the context, of course, but most often in policy analysis "the value of a life" has nothing (directly) to do with output, etc, but instead means the maximum amount that people would want the government to spend in order to save a randomly chosen life.

So in a country of 300,000,000, the question is: What, to you, is the monetary equivalent of a 1/300,000,000 chance of death? Because 1/300,000,000 is a very small number, we don't have to worry terribly much about willingness-to-pay versus willingness-to-accept. (Theory tells us that for small changes, the two willingnesses are effectively equal.)

Returning to the question: How much would you be willing to pay to avoid a 1/300,000,000 chance of death? Now multiply that value by 300,000,000. That's the value of one life, and arguably the amount we'd want our government to spend to preserve a randomly chosen life.

Obviously there are problems with heterogeneity (you and I might not answer the question identically). But as a general rough rule, the estimates tend to come in somewhere under $10 million.

  • 2
    $\begingroup$ Interesting, since the 7 million to 9 million USD figure I'm seeing a few different places (Wikipedia entry for one) do add up to roughly $120,000 times 50-60 years of productive economic life. Granted, that's well over the median and mean incomes, in the US, anyway. But one would expect the value created exceeds the amount paid in wages, especially including intangibles. So I'm not sure I buy that the estimates, however they start, don't end up converging on "economic value"... $\endgroup$ – LightCC Nov 28 '15 at 8:03
  • 1
    $\begingroup$ It's worth adding that for policy related to health interventions, where you are likely to know the age and life expectancy of the recipient, and any health problems likely to reduce quality of life, that analysis is done in terms of quality adjusted life years rather than lives. This justifies spending more to save the life of a young child who will be healthy afterwards than an 80-year-old with multiple other health problems. $\endgroup$ – Dan Apr 18 '18 at 12:42
  • $\begingroup$ Is there a simple theoretical reason to believe that willingness to pay is linear in death probability $p$ at small $p$? If it scales as another power, then the statistical value of a life is not "scale invariant" (or more precisely, does not become scale invariant in the limit of large population size) and therefore depends on the level of government you're considering (local, state, federal, etc.). That would seem to make the concept less natural to me. $\endgroup$ – tparker Sep 14 '18 at 22:00
  • 3
    $\begingroup$ @tparker: To follow up --- A theorem of von Neumann and Morgenstern says that under very general hypotheses, a person will act to maximize the expected value of some utility function $U$. We can normalize so that $U(death)=0$. For a positive real number $x$ write $U(x)$ for the utility of being alive with wealth $x$. Let $W$ be your initial wealth. Let $A(p)$ be the amount you're willing to pay to avoid a probability $p$ of death. Then $(1-p)U(W)=U(W-A(p))\approx U(W)-A(p)U'(W)$ so $A(p)\approx pU(W)/U'(W)$ is linear in $p$ (for small $p$ where the approximation holds). $\endgroup$ – Steven Landsburg Sep 15 '18 at 15:33
  • 1
    $\begingroup$ @tparker: Also, let $B(p)$ be the amount you'd accept to take on a probability $p$ of death. Then $(1-p)U(W+B(p))=U(W)$ and a similar calculation shows that $B(p)\approx pU(W)/U'(W)\approx A(p)$, which both shows linearity for $B(p)$ and justifies the assertion that when $p$ is small, we don't have to worry about philosophical questions like whether willingness-to-pay or willingness-to-accept is the more appropriate measure. $\endgroup$ – Steven Landsburg Sep 15 '18 at 15:38

You can take a look at the approach that insurances uses to pay out life insurance:

In life insurance parlance, "Human Life Value" or HLV, represents the amount that ensures a family's standard of living does not get affected if the one who earns for the family dies or is unable to continue earning.

This source has several links for different methods (such as the income replacement method or the needs-based method).


In New Zealand, \$1,478,507. This is based on a total tax revenue of \$86,597,000,000 divided by the population of 4,779,943 multiplied by our average life expectancy of 81.61 years.

The motivation of this answer is: "In the context of the government, what is the average value of each person in terms of gross tax revenue?"

  • $\begingroup$ There seems to be very little motivation for this calculation. $\endgroup$ – Giskard Mar 17 '19 at 11:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.