Intergenerational Mobility in Norway, 1865–2011

• Date: 01 January 2017
• Published date: 01 January 2017
• Author: Jørgen Modalsli
• DOI: http://doi.org/10.1111/sjoe.12196

Scand. J. of Economics 119(1), 34–71, 2017

DOI: 10.1111/sjoe.12196

Intergenerational Mobility in Norway,

1865–2011∗

Jørgen Modalsli

Statistics Norway, NO-0033 Oslo, Norway

mod@ssb.no

Abstract

There are large differences in intergenerational mobility between countries. Little is known,

however, about how persistent such differences are, and how they evolve over time. This paper

constructs a data set of 835,537 linked father–son pairs from census records and documents

a substantial increase in intergenerational occupational mobility in Norway between 1865

and 2011. The increase is most pronounced in non-farm occupations. The findings show that

long-run mobility developments previously described for the US and UK are not necessarily

representative for other countries, and that high mobility in a given country today need not

reflect high mobility before industrialization.

Keywords: Economic history; intergenerational mobility; mobility measurement; occupations;

JEL classification:J62; N33; N34

I. Introduction

The spread of the Industrial Revolution from its core areas to other parts

of the world from the mid-1800s onward led to massive increases in eco-

nomic growth and human welfare. This development was accompanied by

a decrease in income and wealth inequality in most Western countries,

culminating in historically low income inequality in the 1960s. However,

information on economic inequality and growth does not fully characterize

the distribution of welfare across families. If social mobility is low, mean-

ing that individuals’ positions are to a large extent determined by those of

their parents, then not all members of society will be able to make use of

the increased opportunities made available by industrialization. Economists

often conceptualize this as a “dynasty utility function”, where individuals

have preferences not only over their own welfare, but also that of their

descendants. Hence, the distribution of economic utility depends on inter-

generational mobility. The extent of such mobility changes over time, as

∗I would like to thank Rolf Aaberge, Karin Hederos Eriksson, Magne Mogstad, Steven

Nafziger, Gunnar Thorvaldsen, Ola Vestad, Daniel Waldenstr¨

om, participants at workshops

and seminars, and two anonymous referees for helpful comments and discussions. Support

from the Research Council of Norway is acknowledged.

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

J. Modalsli 35

documented by Long and Ferrie (2013): since the late nineteenth century,

intergenerational mobility has decreased in the US and remained relatively

stable in the UK.1

Constructing estimates of social mobility for time periods earlier than

the late twentieth century is a challenging task. To examine the evolution

of economic characteristics across generations, it is necessary to have data

that measure these characteristics consistently over time, along with the

possibility of linking individuals across generations and time periods. To

study the change in mobility over time, an even longer time-span of obser-

vations is needed. However, prior to the 1970s, few population registries

were computerized, and even fewer large-scale surveys were conducted.

Some existing studies of trends in mobility rely on databases constructed

from preserved records from specific rural regions or small towns.2How-

ever, because industrialization and economic development coincided with

large population movements from the countryside to cities, estimates of so-

cial mobility in such small geographic regions cannot be easily generalized

to infer trends in mobility for society as a whole. The use of country-

wide, census-based data sets has so far been restricted to analyses of Great

Britain and the US.3

In this paper, full-count digitized historical census data for Norway are

combined with modern administrative data to construct a database of the

occupations of a total of 835,537 father–son pairs spanning 146 years, from

1865 to 2011. Intergenerational occupational mobility is found to increase

over the period studied, with the increase being driven by a decrease in the

father–son persistence of non-farm occupations.

This paper contributes to the literature in three ways. It presents the first

nationwide, long-run data set on intergenerational occupational mobility

outside the US and UK; this is also the first study to use a consistent

methodology for the nineteenth, twentieth, and twenty-first centuries. In

order to analyze this data set, new methodology on the decomposition

of measures of intergenerational mobility is developed, highlighting the

differential trends in mobility in and outside farming. While the increas-

ing mobility in Norway was driven by decreasing non-farm father–son

1There is also substantial variation in intergenerational mobility across present-day countries

(Corak, 2013).

2Van Leeuwen and Maas (2010) review the historical sociological literature on intergener-

ational occupational mobility based on such regional databases. Lindahl et al. (2015), who

have studied three generational transitions in the city of Malm¨

o, find no large changes in in-

tergenerational mobility in earnings. Using data from five rural parishes in southern Sweden,

Dribe et al. (2012) find some evidence of increased occupational mobility over time.

3See Long and Ferrie (2007, 2013), Ferrie (2005), and Long (2013). For studies based on

other types of sources, see, for example, Clark and Cummins (2015) (wealth estates) and

Boberg-Fazlic and Sharp (2013) (family reconstitution data).

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

36 Intergenerational mobility in Norway, 1865–2011

persistence, the decrease in mobility in the US can be attributed to an in-

crease in father–son persistence in farming. Supplementing the Norwegian

mobility matrices with occupational mean income provides an economic in-

terpretation of the increase in occupational mobility, and shows the relative

contribution to welfare equalization of changing relative mean occupation

incomes and intergenerational mobility. Finally, this paper explores the role

of regional economic differences in the change in social mobility over time.

Few systematic differences in mobility are found across Norwegian regions,

and the extent of neighborhood effects has changed little. Individuals who

moved from one region to another experienced higher intergenerational

occupational mobility than non-movers.

Long-Run Changes in Social Mobility

There is a small but expanding body of literature on the long-run devel-

opment of social mobility in Western countries. Long and Ferrie (2013)

demonstrate that intergenerational occupational mobility decreased in the

US between the nineteenth and twentieth centuries, using comparable sets

of census or survey data for both periods. For England and Wales, mobil-

ity was lower than in the US in the nineteenth century, and it remains at

about the same level today.4Because of the distinct characteristics of the

history of these countries (early Industrial Revolution in Great Britain and

large geographic expansion and immigration in the US), it is not clear how

the results from these countries generalize to other Western countries. By

constructing a comparable data series for Norway, it becomes clear that the

stability of intergenerational mobility in Great Britain is not representative

of Europe as a whole.

Social mobility in Norway between 1800 and 1950 has previously been

discussed by Semmingsen (1954). Reviewing legal changes and the devel-

opment of the cross-section income distribution, Semmingsen argues that

the move towards a more fluid society started in the eighteenth century

and accelerated through economic liberalization reforms in the nineteenth

century. Social circulation is said to have increased from around 1850 on-

ward, driven by industrialization and the increasing integration of Norway

into the world market. Moreover, technological advances led to increasing

population growth, putting old social structures under pressure. In agricul-

ture, rates of self-ownership were high – by 1900, nearly all farms were

run by owner-occupiers and there were no large estates of the type seen in

4Because of data limitations, most historical studies of intergenerational mobility use occu-

pation information. However, using estate data, Clark and Cummins (2015) examine wealth

mobility in the UK and find strong and stable persistence in the correlation between father

and son wealth between 1858 and 2012.

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