Skip to content

Is long-term earnings inequality growing? Evidence from German baby-boomers and their parents

9 January 2015
by Guest author
Neither cohort looks too happy

Neither cohort looks too happy

Today’s post is by Giacomo Corneo of the Free University of Berlin, following a talk given in the OECD Directorate for Employment, Labour and Social Affairs Seminar Series in October 2014.

Modern welfare economics suggests that lifetime income is a main determinant of how well individuals fare in economic terms. However, most analyses of income inequality are based on yearly data that might be poorly correlated with lifetime incomes. Typically, these analyses include individuals of different age and educational attainment, whose incomes in a given year may be little representative of their long-term incomes. As sample composition changes over time, it is unclear whether the increase of inequality that is often found in cross-sectional analyses corresponds to a similar evolution of long-term inequality or is simply due to sample changes.

In a recent study, we pinned down the evolution of intra-cohort lifetime earnings inequality in Germany. Starting with the cohort born in 1935, we computed the distribution of lifelong earnings among employees who were born in a same year. Our analysis exploits a rich dataset of the German social security system that includes monthly information about earnings, employment status, sickness and other variables of interest for some 240,000 individuals. Based on this, we built a sample that covers about 80 % of the West German labor force in a typical year.

For the cohorts born between 1935 and 1952, we can compute for each individual his or her lifetime earnings, defined as discounted earnings received between age 17 and age 60. For these cohorts we can thus compare lifetime inequality to annual inequality using the Gini coefficient. We find that the distribution of lifetime earnings for these cohorts is rather compressed, with a Gini coefficient that is less than two thirds of the average value of the Gini coefficients of the distributions of yearly earnings. In other words, yearly earnings show far greater inequality than lifetime earnings. This big difference is caused by the mobility of the individuals in the distribution of yearly earnings during their life cycle – the fact that the same individual may rank low in the distribution of annual earnings in some years and rank high in others. Over a lifetime, the ups and downs offset each other to some extent and make the income distribution less unequal.

But how is lifetime inequality evolving over time? Is it increasing, similarly to what is happening in terms of annual inequality?

It is instructive to compare the lifetime inequality experienced by the baby-boomers – who in Germany were born in the 1960s – with their parents. The parents correspond to the oldest cohorts in our sample, while the baby-boomers are now entering their fifties, so that their lifetime earnings cannot be measured yet. In order to gauge the evolution of long-term inequality up to the baby-boomers, we generalized the concept of lifetime earnings to one of up-to-age-X earnings (UAX). These are defined as the present value of earnings received until some age X and discounted to the year when the individual turned age 17. Lifetime earnings are thus a special case of UAX for X equal to 60.

The figure below shows the Gini-coefficient of representative UAX distributions for all cohorts in our sample, the youngest being born in 1972. It shows an upward trend of lifetime inequality, with a secular rise from the cohorts born in the mid-1930s to those born in the early 1970s. (The figure refers to men; our findings for women are qualitatively similar, albeit less strong.)

Gini coefficients of UAX for cohorts 1935-1972, men only


Source: Research Data Centre of the German Pension Insurance System (FDZ-RV), own calculations using weighted data.

In quantitative terms, the intergenerational change that we have detected is considerable. Take for instance the cohort born in 1935 (fathers) and the cohort born in 1963 (sons). The Gini-coefficient of the UA-45 distribution for the fathers equals almost 0.13. The Gini-coefficient of the UA-45 distribution for the sons equals 0.23. This implies a rise of inequality by 85 % and substituting the 1963 cohort with the 1972 cohort yields a rise of inequality by about 100 %. This marks a deep difference between the baby-boom generation and its parents.

Our rich dataset allows us to detect further striking intergenerational change in terms of labor-market outcomes. An important dimension of it relates to pay uncertainty – i.e. the extent to which employees of different cohorts could count on the labor market in order to achieve stable living standards. Statistically, pay uncertainty can be measured by the transitory variance of wages measured when the cohort was 40-years-old. The earnings histories of the baby-boomers and their fathers reveal that pay uncertainty has doubled from one generation to the next.

Another striking intergenerational mutation pertains to exclusion from the labor market. This can be captured by the fraction of a cohort that experienced more than twelve months of unemployment before reaching age 40. For those born in 1935, only 2.3 % of them experienced more than one year of unemployment before they turned 40. For the cohort born in 1963, 28.2 % had that experience. This means that in contrast to their fathers, a substantial fraction of the baby-boom cohort lacked a full integration in the labor market.

In sum, from one generation to the next, Germany has moved from having a rather homogeneous workforce to having a quite heterogeneous one. Compared to their parents, German baby-boomers are substantially more unequal in terms of long-term earnings, are subject to a much stronger pay uncertainty, and are considerably more likely to experience long spells of unemployment. Arguably, this intergenerational change has lowered the cohesion of the workforce and its members’ feeling of sharing a common fate – with potentially far-reaching social and political implications.

Useful links

Income inequality from a lifetime perspective, Giacomo Corneo, Freie Universität Berlin School of Business & Economics Discussion Paper, Economics, 2014/30

OECD work on income inequality and poverty

OECD Income distribution database

OECD Earnings database

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS

Optimization WordPress Plugins & Solutions by W3 EDGE