James K. Hammitt
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
James K. Hammitt has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
Jin-Tan Liu
Department of Economics, National Taiwan University and NBER, 21 Hsu-Chow Road, Taipei 100, Taiwan, Republic of China
4 shared publications- Lisa A. Robinson
Harvard T.H. Chan School of Public Health
4 shared publications Jin-Long Liu
Institute of Industrial Economics, National Central University, Chungli, Taiwan
3 shared publications- Jin-Tan Liu
National Taiwan University
2 shared publications Jin‐Tan Liu
Department of Economics , National Taiwan University and NBER , Taiwan
2 shared publicationsKevin Haninger
U.S. EnvironmentalProtection Agency, 1200 Pennsylvania Ave., NW, Washington, DC 20460, USA
2 shared publications
Latest Publications
- 10 April 2025
- Published by Wiley in Health Economics
- Vol. 34 (7) , 1350-1364
Fair Innings: An Empirical Test
The fair innings principle states that fairness requires allocating life‐saving treatments to younger rather than older patients when each would gain the same extension in longevity. It is motivated by the notion that older patients have already benefited from a longer life and so have less claim to scarce treatment resources than younger patients who have not yet lived their “fair innings.” The principle can be theoretically justified by a prioritarian social welfare function applied to lifetime wellbeing. We conducted an online survey to test whether there is support for the principle in the general population (in France). We find substantial but not universal support. When choosing to allocate a treatment that would provide the same life extension to an older or a younger patient, about one‐half the respondents would allocate the treatment to the younger patient while about one‐third are indifferent to which patient is treated and about one‐fifth would allocate treatment to the older patient. Holding the life extension to the older patient fixed, decreasing the life extension to the younger patient decreases (increases) the fraction of respondents that would allocate treatment to the younger (older) patient. These results highlight the tension between principles of equal treatment and of giving priority to those who are worse off that confound healthcare policy.
- 10 April 2025
- Published by Wiley in Health Economics
- Vol. 34 (7) , 1350-1364
Fair Innings: An Empirical Test
The fair innings principle states that fairness requires allocating life‐saving treatments to younger rather than older patients when each would gain the same extension in longevity. It is motivated by the notion that older patients have already benefited from a longer life and so have less claim to scarce treatment resources than younger patients who have not yet lived their “fair innings.” The principle can be theoretically justified by a prioritarian social welfare function applied to lifetime wellbeing. We conducted an online survey to test whether there is support for the principle in the general population (in France). We find substantial but not universal support. When choosing to allocate a treatment that would provide the same life extension to an older or a younger patient, about one‐half the respondents would allocate the treatment to the younger patient while about one‐third are indifferent to which patient is treated and about one‐fifth would allocate treatment to the older patient. Holding the life extension to the older patient fixed, decreasing the life extension to the younger patient decreases (increases) the fraction of respondents that would allocate treatment to the younger (older) patient. These results highlight the tension between principles of equal treatment and of giving priority to those who are worse off that confound healthcare policy.
- 10 May 2023
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 67 (3) , 239-269
Monetary values of increasing life expectancy: Sensitivity to shifts of the survival curve
Individuals’ monetary values of decreases in mortality risk depend on the magnitude and timing of the risk reduction. We elicited stated preferences among three time paths of risk reduction yielding the same increase in life expectancy (decreasing risk for the next decade, subtracting a constant from or multiplying risk by a constant in all future years) and willingness to pay (WTP) for risk reductions differing in timing and life-expectancy gain. Respondents exhibited heterogeneous preferences over the alternative time paths, with almost 90 percent reporting transitive orderings. WTP is statistically significantly associated with life-expectancy gain (between about 7 and 28 days) and with respondents’ stated preferences over the alternative time paths. Estimated value per statistical life year (VSLY) can differ by time path and averages about $500,000, roughly consistent with conventional estimates obtained by dividing estimated value per statistical life by discounted life expectancy.
- 10 May 2023
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 67 (3) , 239-269
Monetary values of increasing life expectancy: Sensitivity to shifts of the survival curve
Individuals’ monetary values of decreases in mortality risk depend on the magnitude and timing of the risk reduction. We elicited stated preferences among three time paths of risk reduction yielding the same increase in life expectancy (decreasing risk for the next decade, subtracting a constant from or multiplying risk by a constant in all future years) and willingness to pay (WTP) for risk reductions differing in timing and life-expectancy gain. Respondents exhibited heterogeneous preferences over the alternative time paths, with almost 90 percent reporting transitive orderings. WTP is statistically significantly associated with life-expectancy gain (between about 7 and 28 days) and with respondents’ stated preferences over the alternative time paths. Estimated value per statistical life year (VSLY) can differ by time path and averages about $500,000, roughly consistent with conventional estimates obtained by dividing estimated value per statistical life by discounted life expectancy.
- 3 May 2023
- Published by Wiley in Health Economics
- Vol. 32 (9) , 1964-1981
Consistent valuation of a reduction in mortality risk using values per life, life year, and quality‐adjusted life year
The monetary value of a reduction in mortality risk can be accurately characterized using the alternative concepts of value per statistical life (VSL), value per statistical life year (VSLY), and value per quality‐adjusted life year (VQALY). Typically, each of these values depends on the age and other characteristics of the affected individual; at most one of the values can be independent of age. The common practice of valuing a transient or persistent risk reduction using a constant VSL, VSLY, or VQALY yields systematic differences in the calculated monetary value that depend on the age at which the risk reduction begins, its duration, time path, and whether future lives, life years, or quality‐adjusted life years are discounted. Mutually consistent, age‐dependent VSL, VSLY, and VQALY are derived and the large differences in valuation of illustrative transient and persistent risk reductions that can result from assuming age‐independent values of each of the three concepts are illustrated.
- 3 May 2023
- Published by Wiley in Health Economics
- Vol. 32 (9) , 1964-1981
Consistent valuation of a reduction in mortality risk using values per life, life year, and quality‐adjusted life year
The monetary value of a reduction in mortality risk can be accurately characterized using the alternative concepts of value per statistical life (VSL), value per statistical life year (VSLY), and value per quality‐adjusted life year (VQALY). Typically, each of these values depends on the age and other characteristics of the affected individual; at most one of the values can be independent of age. The common practice of valuing a transient or persistent risk reduction using a constant VSL, VSLY, or VQALY yields systematic differences in the calculated monetary value that depend on the age at which the risk reduction begins, its duration, time path, and whether future lives, life years, or quality‐adjusted life years are discounted. Mutually consistent, age‐dependent VSL, VSLY, and VQALY are derived and the large differences in valuation of illustrative transient and persistent risk reductions that can result from assuming age‐independent values of each of the three concepts are illustrated.
- 1 December 2022
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 65 (3) , 239-260
Is survival a luxury good? Income elasticity of the value per statistical life
- 1 December 2022
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 65 (3) , 239-260
Is survival a luxury good? Income elasticity of the value per statistical life
- 4 October 2022
- Published by Springer Nature in Theory and Decision
- Vol. 95 (1) , 1-10
Downside risk aversion vs decreasing absolute risk aversion: an intuitive exposition
- 4 October 2022
- Published by Springer Nature in Theory and Decision
- Vol. 95 (1) , 1-10
Downside risk aversion vs decreasing absolute risk aversion: an intuitive exposition
- 7 June 2022
- Published by Elsevier in Journal of Health Economics
- Vol. 84, 102643
Prevention, treatment, and palliative care: The relative value of health improvements under alternative evaluation frameworks
- 7 June 2022
- Published by Elsevier in Journal of Health Economics
- Vol. 84, 102643
Prevention, treatment, and palliative care: The relative value of health improvements under alternative evaluation frameworks
- 17 December 2020
- Published by Cambridge University Press (CUP) in Journal of Benefit-Cost Analysis
- Vol. 12 (1) , 64-84
Accounting for the Distribution of Benefits and Costs in Benefit–Cost Analysis
Benefit–cost analysis (BCA) is often viewed as measuring the efficiency of a policy independent of the distribution of its consequences. The role of distributional effects on policy choice is disputed; either: (a) the policy that maximizes net benefits should be selected and distributional concerns should be addressed through other measures, such as tax and transfer programs or (b) BCA should be supplemented with distributional analysis and decision-makers should weigh efficiency and distribution in policy choice. The separation of efficiency and distribution is misleading. The measure of efficiency depends on the numéraire chosen for the analysis, whether monetary values or some other good (unless individuals have the same rates of substitution between them). The choice of numéraire is not neutral; it can affect the ranking of policies by calculated net benefits. Alternative evaluation methods, such as BCA using a different numéraire, weighted BCA, or a social welfare function (SWF), may better integrate concerns about distribution and efficiency. The most appropriate numéraire, distributional weights, or SWFs cannot be measured or statistically estimated; it is a normative choice.
- 17 December 2020
- Published by Cambridge University Press (CUP) in Journal of Benefit-Cost Analysis
- Vol. 12 (1) , 64-84
Accounting for the Distribution of Benefits and Costs in Benefit–Cost Analysis
Benefit–cost analysis (BCA) is often viewed as measuring the efficiency of a policy independent of the distribution of its consequences. The role of distributional effects on policy choice is disputed; either: (a) the policy that maximizes net benefits should be selected and distributional concerns should be addressed through other measures, such as tax and transfer programs or (b) BCA should be supplemented with distributional analysis and decision-makers should weigh efficiency and distribution in policy choice. The separation of efficiency and distribution is misleading. The measure of efficiency depends on the numéraire chosen for the analysis, whether monetary values or some other good (unless individuals have the same rates of substitution between them). The choice of numéraire is not neutral; it can affect the ranking of policies by calculated net benefits. Alternative evaluation methods, such as BCA using a different numéraire, weighted BCA, or a social welfare function (SWF), may better integrate concerns about distribution and efficiency. The most appropriate numéraire, distributional weights, or SWFs cannot be measured or statistically estimated; it is a normative choice.
- Open Access
- 1 October 2020
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 61 (2) , 129-154
Valuing mortality risk in the time of COVID-19
In evaluating the appropriate response to the COVID-19 pandemic, a key parameter is the rate of substitution between wealth and mortality risk, conventionally summarized as the value per statistical life (VSL). For the United States, VSL is estimated as approximately $10 million, which implies the value of preventing 100,000 COVID-19 deaths is $1 trillion. Is this value too large? There are reasons to think so. First, VSL is a marginal rate of substitution and the potential risk reductions are non-marginal. The standard VSL model implies the rate of substitution of wealth for risk reduction is smaller when the risk reduction is larger, but a closed-form solution calibrated to estimates of the income elasticity of VSL shows the rate of decline is modest until the value of a non-marginal risk reduction accounts for a substantial share of income; average individuals are predicted to be willing to spend more than half their income to reduce one-year mortality risk by 1 in 100. Second, mortality risk is concentrated among the elderly, for whom VSL may be smaller and who would benefit from a persistent risk reduction for a shorter period because of their shorter life expectancy. Third, the pandemic and responses to it have caused substantial losses in income that should decrease VSL. In contrast, VSL is plausibly larger for risks (like COVID-19) that are dreaded, uncertain, catastrophic, and ambiguous. These arguments are evaluated and key issues for improving estimates are highlighted.
- 1 October 2020
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 61 (2) , 129-154
Open AccessValuing mortality risk in the time of COVID-19
In evaluating the appropriate response to the COVID-19 pandemic, a key parameter is the rate of substitution between wealth and mortality risk, conventionally summarized as the value per statistical life (VSL). For the United States, VSL is estimated as approximately $10 million, which implies the value of preventing 100,000 COVID-19 deaths is $1 trillion. Is this value too large? There are reasons to think so. First, VSL is a marginal rate of substitution and the potential risk reductions are non-marginal. The standard VSL model implies the rate of substitution of wealth for risk reduction is smaller when the risk reduction is larger, but a closed-form solution calibrated to estimates of the income elasticity of VSL shows the rate of decline is modest until the value of a non-marginal risk reduction accounts for a substantial share of income; average individuals are predicted to be willing to spend more than half their income to reduce one-year mortality risk by 1 in 100. Second, mortality risk is concentrated among the elderly, for whom VSL may be smaller and who would benefit from a persistent risk reduction for a shorter period because of their shorter life expectancy. Third, the pandemic and responses to it have caused substantial losses in income that should decrease VSL. In contrast, VSL is plausibly larger for risks (like COVID-19) that are dreaded, uncertain, catastrophic, and ambiguous. These arguments are evaluated and key issues for improving estimates are highlighted.
- Open Access
- 10 December 2019
- Published by Wiley in Risk Analysis
- Vol. 40 (4) , 674-695
Premature Deaths, Statistical Lives, and Years of Life Lost: Identification, Quantification, and Valuation of Mortality Risks
Mortality effects of exposure to air pollution and other environmental hazards are often described by the estimated number of “premature” or “attributable” deaths and the economic value of a reduction in exposure as the product of an estimate of “statistical lives saved” and a “value per statistical life.” These terms can be misleading because the number of deaths advanced by exposure cannot be determined from mortality data alone, whether from epidemiology or randomized trials (it is not statistically identified). The fraction of deaths “attributed” to exposure is conventionally derived as the hazard fraction (R – 1)/R, where R is the relative risk of mortality between high and low exposure levels. The fraction of deaths advanced by exposure (the “etiologic” fraction) can be substantially larger or smaller: it can be as large as one and as small as 1/e (≈0.37) times the hazard fraction (if the association is causal and zero otherwise). Recent literature reveals misunderstanding about these concepts. Total life years lost in a population due to exposure can be estimated but cannot be disaggregated by age or cause of death. Economic valuation of a change in exposure‐related mortality risk to a population is not affected by inability to know the fraction of deaths that are etiologic. When individuals facing larger or smaller changes in mortality risk cannot be identified, the mean change in population hazard is sufficient for valuation; otherwise, the economic value can depend on the distribution of risk reductions.
- 10 December 2019
- Published by Wiley in Risk Analysis
- Vol. 40 (4) , 674-695
Open AccessPremature Deaths, Statistical Lives, and Years of Life Lost: Identification, Quantification, and Valuation of Mortality Risks
Mortality effects of exposure to air pollution and other environmental hazards are often described by the estimated number of “premature” or “attributable” deaths and the economic value of a reduction in exposure as the product of an estimate of “statistical lives saved” and a “value per statistical life.” These terms can be misleading because the number of deaths advanced by exposure cannot be determined from mortality data alone, whether from epidemiology or randomized trials (it is not statistically identified). The fraction of deaths “attributed” to exposure is conventionally derived as the hazard fraction (R – 1)/R, where R is the relative risk of mortality between high and low exposure levels. The fraction of deaths advanced by exposure (the “etiologic” fraction) can be substantially larger or smaller: it can be as large as one and as small as 1/e (≈0.37) times the hazard fraction (if the association is causal and zero otherwise). Recent literature reveals misunderstanding about these concepts. Total life years lost in a population due to exposure can be estimated but cannot be disaggregated by age or cause of death. Economic valuation of a change in exposure‐related mortality risk to a population is not affected by inability to know the fraction of deaths that are etiologic. When individuals facing larger or smaller changes in mortality risk cannot be identified, the mean change in population hazard is sufficient for valuation; otherwise, the economic value can depend on the distribution of risk reductions.
- 15 June 2019
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 58 (2-3) , 167-186
Valuing mortality risk in China: Comparing stated-preference estimates from 2005 and 2016
- 15 June 2019
- Published by Springer Nature in Journal of Risk and Uncertainty
- Vol. 58 (2-3) , 167-186
Valuing mortality risk in China: Comparing stated-preference estimates from 2005 and 2016