Demographic Transition and Fertility Rebound in Economic Development*

• DOI: http://doi.org/10.1111/sjoe.12391
• Date: 01 October 2020
• Published date: 01 October 2020
• Author: Asako Ohinata,Dimitrios Varvarigos

Scand. J. of Economics 122(4), 1640–1670, 2020

DOI: 10.1111/sjoe.12391

Demographic Transition and Fertility

Rebound in Economic Development*

Asako Ohinata

University of Leicester, Leicester LE1 7RH, UK

ao150@le.ac.uk

Dimitrios Varvarigos

University of Leicester, Leicester LE1 7RH, UK

dv33@le.ac.uk

Abstract

Recent evidence on the “fertility rebound” offers credence to the idea that, from the onset of

early industrialization to the present day, the dynamics of fertility can be represented by an

N-shaped curve.An overlapping generations model with parental investment in human capital can

account for these observed movementsin fertility rates during the different stages of demographic

change.A demographictransition with declining fertility emergesat theinter mediate stage, when

parents engage on a child quantity–quality trade-off. At later stages, however, the process of

economic growth generates sufficient resources so that households can rear more children while

still providing the desirable amount of education investmentper child.

Keywords: Demographic transition; fertility rebound; human capital

JEL classification:J11; O41

I. Introduction

The relation between economic growth and demographic change has been

at the forefront of research on the economics of development for at least

three decades. Motivated by recent evidence on a phase of demographic

change that researchers have termed “fertility rebound”, this study is the

first to develop a theory that offers a joint account for three empirically

observed phases of changes in fertility trends, during the various stages of

the development process.

Until recently, the conventional wisdom with regard to fertility dynamics

seemed to favour the view that, from the onset of early industrialization,

population changes in Northern Europe and the Scandinavian Peninsula

can be categorized into two distinct stages. As a means of illustrating this

*We would like to thank two anonymous reviewers for their constructive comments and

suggestions, which helped us improve several aspects of our study. Any remaining errors are

our own responsibility.

C

The editors of The Scandinavian Journal of Economics 2019.

A. Ohinata and D. Varvarigos 1641

point, in Figure 1 we present trends in the total fertility rate (TFR)1in four

western regions between 1850 and 2015.2Focusing on Northern Europe

and the Scandinavian Peninsula, we can observe what has also been pointed

out by many researchers, such as Dyson and Murphy (1985), Galor (2005),

and Mokyr and Voth (2010), among others. During the first stage (i.e., up

to around 1870), fertility rates and population growth increased in these

regions,3whereas the second stage is a prolonged one, during which these

regions, together with Southern Europe and the Western “Offshoots” (see

Figure 1), have witnessed a striking decrease in fertility rates.4

Given these observations, a number of studies have endeavoured to offer

theoretical frameworks that account for both these phases of demographic

change (e.g., Tamura, 1994, 1996; Galor and Weil, 2000; Lagerl¨of, 2003;

Tabata, 2003; Strulik, 2008; Strulik and Weisdorf, 2008), while other studies

have focused mainly on the second stage of the aforementioned changes

(e.g., Becker et al., 1990; Galor and Weil, 1996; Blackburn and Cipriani,

2002; Varvarigos and Zakaria, 2013).5

However, more recent data indicate that some developed countries are

undergoing a new phase of demographic change. As a means of illustrating

and clarifying this point, in Figure 2 we focus on a sample from 1980

to 2015. These scatterplots indicate that the process of fertility reductions

might have been reversed in most of the countries in the sample. Rather than

pointing to an outcome that was previously unknown, however, the main

1The TFR is defined as the number of children that wouldbe bor n to a womanif she were to live

to the end of her childbearing years and bear children in accordance with current age-specific

fertility rates.

2Data on the TFR are taken from Gapminder (https://www.gapminder.org/). We plot separate

trends for Northern European countries (Austria, Belgium, Germany, Ireland, Netherlands,

Switzerland, and the United Kingdom), countries of the Scandinavian Peninsula (Denmark,

Finland, Norway, and Sweden), Southern European countries (France, Greece, Italy, Portugal,

and Spain), and theWestern “Offshoots” (Australia, Canada, and the United States). These figures

include both the scatter plots as well as the fitted local polynomials with degree one.The ker nel

function for each figure is chosen to be alternative Epanechnikov, although the choice of kernel

function does not significantly affect the fit to the data. The bandwidth is chosen bythe r ule-of-

thumb bandwidth selector.The cor responding bandwidth is stated at the bottom of each figure.

3In fact, recent empirical research has shown that, in some countries, a post-Malthusian

regime entailing a positive correlation between income and fertility rates emerged prior to

industrialization. See Møller and Sharp (2014) for evidence from the United Kingdom, and

Klemp and Møller (2016) for evidence from Scandinavian countries.

4As is evident from Figure 1, for many regions the demographic transition was temporarily

interrupted by the “baby boom”. Wewill elaborate on this point later in our analysis.

5Pestieau and Ponthi`ere (2014) construct a model of multiple reproductive periods to argue that

optimal fertility can be a source of endogenous cycles. Borck (2011) studies endogenous fertility

under different regimes regarding the provision of public education, whereas Dioikitopoulos

(2014) employs a model of endogenous fertility to examine the optimal allocation of public

expenditure between health and education.

C

The editors of The Scandinavian Journal of Economics 2019.