Trends in Life Expectancy by Income and the Role of
Speciﬁc Causes of Death
† and J
†Swedish Institute for Social Research, Stockholm University ‡University of Helsinki, VATT
Institute for Economic Research and Swedish Institute for Social Research, Stockholm University
Final version received 7 October 2016.
This study explores how life expectancy at age 35 has evolved across the income distribution in Sweden
over time. We examine individual income for men 1970–2007 and family income for both men and women
1980–2007. During this period, income inequality increased in most western countries, but especially so in
Sweden. Drawing on a large sample of the Swedish population, our results show that the gap in life
expectancy between the richest and poorest ﬁfths of the income distribution also increased. This was the
case both for individual and family income. The increase was larger for men than for women, but the only
group with stagnant life expectancy at age 35 was women in the lowest income quintile group. Between
1986 and 2007, the diﬀerence between the lowest and highest family income quintiles increased by about
one year for women and by almost two years for men. The causes of death that most signiﬁcantly
contributed to the increased disparities among women were circulatory and respiratory diseases. For men,
circulatory disease mortality alone caused most of the increased disparities.
That life expectancy is positively associated with socioeconomic status is a well-
documented phenomenon (Antonovsky 1967; Cutler and Lleras-Muney 2006; Elo 2009).
People with higher earnings and more education can expect to live longer than less
advantaged individuals. For example, employed men in the lowest income quintile group
in Sweden have more than twice the mortality risk of men in the highest income group,
while the diﬀerence is smaller among women but still signiﬁcant (Torssander and Erikson
2010). Furthermore, social inequalities in mortality seem to have increased during the
last decades in Sweden (Kondo et al. 2014; Shkolnikov et al. 2012), as well as in other
Nordic countries (Moe et al. 2012; Brønnum-Hansen and Baadsgaard 2012; Tarkiainen
et al. 2012) and European countries (Mackenbach et al. 2003, 2015).
During this time, there have also been large changes in income distributions in many
parts of the world. In particular, inequality of income has increased substantially. In the
historically egalitarian Sweden, the increase in inequality has been more rapid than in
any other OECD country (OECD 2011).
Because of increasing income inequality in
Sweden and elsewhere, the trend in life expectancy diﬀerences across income groups is of
particular interest from a general equity perspective. For a longer time period than has
been considered before, from 1970 to 2007, we describe how diﬀerences in life expectancy
across income groups have developed in Sweden, and examine which causes of death that
contributed to any changes.
Existing studies show that life expectancy disparities by education have increased in
Sweden over the last decades (Shkolnikov et al. 2012; National Board of Health and
Welfare [Socialstyrelsen] 2013), while previous studies of the trend in the income–
mortality association are fairly inconsistent. One study reports that income inequalities
in mortality have increased for both women and men (Kondo et al. 2014), while another
reports that they have increased for women but decreased for men (Wamala et al. 2006).
By using administrative register data covering a large part of the Swedish population, we
© 2017 The London School of Economics and Political Science. Published by Blackwell Publishing, 9600 Garsington Road,
Oxford OX4 2DQ, UK and 350 Main St, Malden, MA 02148, USA
Economica (2018) 85, 606–625