The Evolution of Social Mobility: Norway during the Twentieth Century

• DOI: http://doi.org/10.1111/sjoe.12205
• Author: Kjell G. Salvanes,Tuomas Pekkarinen,Matti Sarvimäki
• Published date: 01 January 2017
• Date: 01 January 2017

Scand. J. of Economics 119(1), 5–33, 2017

DOI: 10.1111/sjoe.12205

The Evolution of Social Mobility: Norway

during the Twentieth Century∗

Tuomas Pekkarinen

VATT Institute for Economic Research, FI-00101 Helsinki, Finland

tuomas.pekkarinen@vatt.fi

Kjell G. Salvanes

Norwegian School of Economics, NO-5045 Bergen, Norway

kjell.salvanes@nhh.no

Matti Sarvim¨

aki

Aalto University, FI-00076 Helsinki, Finland

matti.sarvimaki@aalto.fi

Abstract

We document trends in social mobility in Norway using intergenerational income elasticities,

the associations between the income percentiles of fathers and sons, and brother correlations.

The results of all approaches suggest that social mobility increased substantially between

cohorts born in the early 1930s and the early 1940s. Father–son associations remained stable

for cohorts born after World War II, while brother correlations continued to decline. The

relationship between father and son income percentile ranks is highly non-linear for early

cohorts, but it approaches linearity over time. We discuss increasing educational attainment

among low- and middle-income families as a possible mechanism underlying these trends.

Keywords: Intergenerational mobility

JEL classification:D31; J31; J62

I. Introduction

The debate on the consequences of income inequality has drawn attention

to cross-country differences in social mobility. A large body of research

has shown that countries that are known for redistributive welfare state

institutions and low cross-sectional income inequality, such as the Nordic

countries, have a much lower degree of intergenerational income persistence

than, for example, the US or UK.1These cross-country differences have

∗We thank two anonymous referees and seminar participants at EALE, FEA, HECER, NHH,

SOLE, and the Conference on Social Mobility held at the University of Chicago for insightful

comments and suggestions, and the Academy of Finland and the Norwegian Research Council

for funding.

1See Black and Devereux (2011) and Corak (2013) for recent surveys.

CThe editors of The Scandinavian Journal of Economics 2016.

6Evolution of social mobility: Norway during the 20th century

led to speculation about their potential causes and implications. Yet, it is

difficult to draw conclusions from a pattern present at a single point in

time. As a response, recent research has shifted towards a complementary

approach of documenting within-country changes in social mobility.

In this paper, we examine the evolution of social mobility in Norway

for children born between the early 1930s and the mid-1970s using newly

digitalized data and alternative measurement approaches. These birth co-

horts are of particular interest because they cover a period in which the

Norwegian economy underwent dramatic structural change and much of the

Norwegian welfare state was built. The last few cohorts included in our

sample were born into one of the world’s richest countries, with extensive

redistributive institutions and a high level of intergenerational mobility. In

contrast, our earliest birth cohorts grew up in a relatively poor and unequal

country. We show that they also experienced less social mobility than did

subsequent birth cohorts.

We contribute to the earlier body of literature across several dimensions.

First, we use high-quality register data augmented with military records

from the early 1950s and newly digitalized municipal tax records from

1948. These data allow us to present precise estimates, even for those co-

horts born before World War II (WWII). Moreover, we use three different

measurement approaches – intergenerational income elasticities, associa-

tions between the income percentile ranks of fathers and sons, and brother

correlations – in order to assess the robustness of patterns over time. We

also examine non-linearities in father–son associations and, in particular,

evidentiary changes in these across birth cohorts. Finally, we document

the changes in the association between educational attainment and family

background.

Our paper adds to the growing body of literature on historical trends in

intergenerational mobility. Previous work examining Nordic countries in-

cludes Pekkala and Lucas (2007), who examine trends in intergenerational

income elasticity in Finland, and Bj¨

orklund et al. (2009), who investigate

the evolution of brother income correlations in Sweden. Both of these stud-

ies present evidence on the increasing mobility between cohorts born in the

1930s and 1950s, and stable or decreasing social mobility for later birth

cohorts. Modalsli (2017) documents a substantial increase in intergenera-

tional occupational mobility in Norway between 1865 and 2011. In con-

trast, Lindahl et al. (2015) focus on the descendants of a single generation

of schoolchildren in one Swedish city and find no evidence of changes

in intergenerational income mobility. Clark (2012) examines the persis-

tence of surnames among elite occupations and argues that rates of social

mobility in Sweden have remained roughly stable since the pre-industrial

era. Finally, and in line with our results, Bratberg et al. (2005) find that

CThe editors of The Scandinavian Journal of Economics 2016.