Understanding the dynamics of the survival prospects of a given population as measured by life expectancy is vital in multiple research and policy areas, for instance, in public and private health care planning (e.g., developing preventive actions, in measuring human development and policy outcomes, in preparing for long-term care needs, in anticipating and understanding epidemiological episodes), in demographic analysis (e.g., population projections, ageing assessment), in pension systems design, reform and solvency analysis (e.g., updating the retirement age to longevity increases), in the pricing and risk management of longevity-linked life insurance contracts, private (individual, occupational) retirement income schemes and novel capital market instruments (e.g., longevity bonds, longevity swaps, mortality forwards). 

To measure longevity developments in a population, the metrics most adopted summarize average outcomes, such as the life expectancy at birth (or at retirement ages) or age-standardized mortality ratios. For instance, period life expectancy at birth measures, at a point in time, the average number of years a newborn is predicted to live given the survival (mortality) conditions observed at that time, i.e., assuming that newborn will experience throughout life the same conditions observed when he/she was born.  

Abstracting from the well-known deficiencies of period life expectancy to measure expected longevity when compared to birth cohort measures (see, e.g., Ayuso, Bravo & Holzmann, 2021; Bravo et al., 2021), life expectancy and similar markers are an average, hides the lifespan differences among individuals in a population, neglecting the significant variation in the ages at death among them.  

In recent years, a substantial body of empirical evidence has emerged displaying growing heterogeneity (inequality) in average life expectancy between socioeconomic groups (see, e.g., Chetty et al., 2016; Ayuso et al., 2017a,b), broadening previously analysis mostly focused on sex differences in longevity. Moreover, life expectancy heterogeneity and lifespan inequality has been worsening in many countries. The gap between the average lifespan of individuals belonging to different socioeconomic groups has been attributed to differences in income, wealth, education, lifestyles, between various demographic groups.  

Since life expectancy metrics serve to evaluate policy outcomes (e.g., pension policy designs and reforms such as automatically indexing the retirement age or defining sustainability factors, health policies), to compare national and subnational populations development, to early identify public health emergency threats (e.g., pandemics) or to set public health objectives (longevity increases) targets, it is important to know that average metric may hide substantial inequality that counteracts policy intentions. For instance, longevity heterogeneity perverts the redistributive objectives of pension schemes and distorts individual lifecycle labor supply and savings decisions which, in turn, risk invalidating current reform approaches targeting, for instance, a closer relationship between benefits and contributions and life expectancy-indexed retirement ages (Ayuso et al., 2017a,b). 

We believe that to fully assess longevity developments in a population, it is not sufficient to measure life expectancy. We need also to monitor lifespan inequality. The concept of lifespan inequality (also known as lifespan disparity or age-at-death variation) measures how ages at death are spread out across individuals in a population. At the individual level, lifespan disparity quantifies the uncertainty in the timing of death which is not necessarily negative and can be attributed to biological (genetic) diversity, cultural and social diversity (e.g., different lifestyles, nutrition habits), the exposure to different hazards (e.g., hazardous jobs) or simply accident mortality.  

In aggregate terms, lifespan inequality usually suggests heterogeneity in the population's health levels. Lifespan inequality is measured using markers such as the life disparity (e-dagger), the life table entropy, the variance or the coefficient of variation of the remaining lifetime of an individual, the Gini coefficient, or the Theil index. 

Lifespan inequality is off course negative when it relates to inequality of opportunity or, in the more extreme scenario, when it expresses unfair and avoidable differences in the access to, for instance, health services, to education services, to job opportunities, or to affordable housing. At the societal level, lifespan (and economic) inequality are likely to trigger populism, to undermine social cohesion, to demotivate democratic participation, to represent a waste of talent and economic underperformance. 

It is thus important to go beyond traditional metrics of human longevity, measuring to what extent the longevity of an individual from a subgroup with low socioeconomic status is longer or shorter than that of an individual from a high socioeconomic status subgroup. While life expectancy seizes the size of average longevity improvements, lifespan inequality captures the (in)equality in survival developments across different groups. 

Lifespan inequality has been described as the most important of all inequalities because every other type of inequality (e.g., economic) is conditional upon being alive (van Raalte, Sasson, & Martikainen, 2018). Lifespan inequality is an ultimate manifestation of health and living conditions disparities, with empirical studies showing that the most deprived groups in the society experience the lowest life expectancy and the highest amount of variation in age at death, with inequalities still growing over time. 

The classical health transition theories suggest that longevity increases go in tandem with the so-called "mortality compression" or (outer) "rectangularization hypothesis" (Fries, 1980). The empirical studies in high-income countries show that period life expectancy increases are strongly inversely correlated with lifespan variation when one considers the entire human lifespan (Wilmoth & Horiuchi, 1999; Aburto et al., 2020).  

Figures 1 and 2 confirm these results representing, respectively, the observed and forecasted values of the life table Gini coefficient for the female and male populations of selected countries. Lifespan disparity has decreased mainly as a result of survival improvements at premature ages, which shifted deaths toward the end of the lifespan.  

Figure 1. Observed and forecasted values of the life table Gini coefficient for female populations of selected countries, 1960-2075 

Figure 2. Observed and forecasted values of the life table Gini coefficient for male populations of selected countries, 1960-2075 

This empirical evidence could suggest continuing increases in life expectancy are reducing lifespan inequality. Yet, when observed from a birth cohort perspective, and across adult (retirement) and more advanced age ranges only, the compression in period life tables may hide mortality shifting and expansion and stagnant or increasing lifespan inequality in birth cohorts due to heterogeneous mortality regimes. 

Lifespan inequality substantially reduces if lives are saved at infancy and teenage years but increases if lives are saved at older ages (Aburto et al., 2020). This means the relationship between life expectancy and lifespan inequality is more complex than previously thought. In developed countries, where recent trends in life expectancy have been attributed to reductions in old-age mortality, we may see increases in lifespan inequality due to the heterogeneous composition of the population.  

To fully understand longevity, we need to monitor and forecast the relationship between life expectancy and lifespan inequality markers, identify the specific ages and causes of death that explain inequality, analyze the development of life expectancy and lifespan inequality by socioeconomic groups, and discuss how to intervene to minimize the perpetuation of social inequalities and inequality of opportunity.  

Reducing average mortality rates may not be enough to reduce lifespan inequality, it is important to know which policies matter the most when it comes to mitigating disparity in the distribution of the ages at death. Increasing retirement ages automatically with life expectancy may redistribute pension wealth from the poorer in the society to the richer. Uncertainty in the timing of death distorts labour market, retirement and consumption and saving decisions, affecting social relationships. 

Acknowledgments  

Mercedes Ayuso is grateful to the Spanish Ministry of Science and Innovation for funding received under grant PID2019-105986GB-C21, and to the Secretaria d'Universitats i Recerca del departament d'Empresa i Coneixement de la Generalitat de Catalunya for funding received under grant 2020-PANDE-00074. Additionally, Jorge M. Bravo was supported by Portuguese national science funds through FCT under the project UIDB/04152/2020-Centro de Investigação em Gestão de Informação (MagIC).  

References  

Aburto, J. M., Villavicencio, F., Basellini, U., Kjærgaard, S., & Vaupel, J. W. (2020). Dynamics of life expectancy and life span equality. Proceedings of the National Academy of Sciences of the United States of America, 117(10), 5250-5259. https://doi.org/10.1073/pnas.1915884117 

Ayuso, M., Bravo, J. M. & Holzmann, R. (2021). Getting life expectancy estimates right for pension policy: period versus cohort approach. Journal of Pension Economics and Finance, 20(2), 212–231. https://doi.org/10.1017/S1474747220000050. 

Ayuso, M., Bravo, J. M., & Holzmann, R. (2017a). On the heterogeneity in longevity among socioeconomic groups: Scope, trends, and implications for Earnings-Related Pension Schemes. Global Journal of Human Social Sciences - Economics, 17(1): 31-57. 

Ayuso, M., Bravo, J. M., & Holzmann, R. (2017b). Addressing longevity heterogeneity in pension scheme design. Journal of Finance and Economics, 6(1): 1-21. 

Bravo, J. M., Ayuso, M., Holzmann, R. & Palmer, E. (2021). Addressing life expectancy gap in pension policy. Insurance: Mathematics and Economics, 99, 200-221. 

Chetty, R., Stepner, M., Abraham, S., Lin, S., Scuderi, B., Turner, N., Bergeron, A., & Cutler, D. (2016). The association between income and life expectancy in the United States, 2001-2014. The Journal of the American Medical Association, 315 (14): 1750-1766. 

Fries, J. F. (1980). Aging, natural death, and the compression of morbidity. New England Journal of Medicine, 303, 130-135. 

van Raalte, A. A., Sasson, I., & Martikainen, P. (2018). The case for monitoring life-span inequality. Science, 362(6418), 1002-1004. 

Wilmoth, J. R., Horiuchi, S. (1999). Rectangularization revisited: Variability of age at death within human populations. Demography, 36, 4, 475-495.