Cross-sectional Average Length of Life Entropy: Cross-Population Comparisons and Decomposition

Wen Su, Australian National University
Vladimir Canudas-Romo, Australian National University

Research on inequality in mortality needs to consider both the length of lifespan and the variation around the age at death for a population. Keyfitz-Leser’s life table entropy was proposed to serve as a standardized inequality in mortality measure. We present an index that incorporates the history of survival of all cohorts presented at a given time, namely the Cross-sectional Average Length of Life Entropy, or CAL entropy. We decompose cross-population differences of CAL entropy into the contribution of longevity and contribution of lifespan variation, and the change of those differences across time. Our illustrations show that lifespan variation holds a significant share in both differences of entropy among the population selected and the contribution to changes over time in differences, mostly among low-inequality Western European populations. Longevity was once a significant influence on population-relative disparities and its change, but its influence has receded over the years. The US demonstrates a unique trend where it is performing worse across time compared to the Western European level and lifespan variation has played a major role in this process. This study signals the importance of lifespan variation in reducing inequality in mortality among developed and longevous populations.

Keywords: Decomposition analysis/methods, Methodology, Mortality

See paper.

  Presented in Session 100. Socioeconomic Inequality and Mortality Differentials