Son preference and the demographic transition

• Published date: 01 February 2022
• Author: Anna‐Maria Aksan
• Date: 01 February 2022
• DOI: http://doi.org/10.1111/rode.12831

32 

|

     wileyonlinelibrary.com/journal/rode  Rev Dev Econ. 2022;26:32–56.

© 2021 John Wiley & Sons Ltd

Received: 10 June 2021 

|

  Revised: 26 August 2021 

|

  Accepted: 31 August 2021

DOI: 10.1111/rode.12831  

REGULAR ARTICLE

Son preference and the demographic transition

Anna- MariaAksan

Department of Economics, DMH 326, 

Fairfield University, Fairfield, CT, USA

Correspondence

Anna- Maria Aksan, Department 

of Economics, DMH 326, Fairfield 

University, 1073 North Benson Road, 

Fairfield, CT 06824- 5195, USA.

Email: aaksan@fairfield.edu

Abstract

In  a  general  equilibrium  model  of  fertility  with  higher 

economic returns  to sons  relative to  daughters, parents 

choose  overall  fertility  and  the  gender  composition  of 

their children. Son preference is partially endogenized to 

reflect how relative scarcity  of females raises their value 

even while  social norms and lack  of economic opportu-

nities lessen their value. These competing factors lead to 

an oscillating  sex ratio.  Model simulations demonstrate 

that son preference increases fertility,  but that sex selec-

tion reduces  fertility  in the  presence of  son preference. 

The results suggest that effectively banning sex- selective 

abortions  in  places  such  as  India,  which  has  struggled 

to enforce its existing  ban on the practice, is insufficient 

and may have  unintended consequences  such  as slow-

ing fertility decline, reducing quality investment in girls, 

and slowing human capital accumulation and economic 

growth.  Improving  economic  opportunities  for  women 

will increase the value placed  on daughters, thereby im-

proving the  sex  ratio and  human capital  investment  in 

all children.

KEYWORDS

demographic transition, sex selection, son preference, quantity– 

quality tradeoff

JEL CLASSIFICATION

J11; J13; J16

|

33

AKSAN



|

INTRODUCTION

Since 1980 India’s child mortality rate (CMR) fell by 71%, from 166 deaths per 1,000 live births to 

47.7 in 2015, while the  total fertility rate (TFR) declined from 4.8 to 2.4  children per woman. In 

China the CMR  fell by 82%, from 61  deaths per 1,000 live  births to 10.7 in  2015, while the TFR 

declined from 2.6 to 1.6 (World Development Indicators, 2017). Yet this apparent progress is ob-

scured by a severe sex ratio imbalance in  both countries. The most recent Indian census data of  

2011 indicate that the current sex ratio at birth (SRB) was 1.12 males per female, up from 1.10 in 

2001, and has been estimated to be as high as 1.3– 1.5 in some regions of the country (Guilmoto, 

2009; Rajan  et  al., 2017).  In  China the  SRB rose  from 1.09  in  1982 to  1.18  in 2012  (UNICEF, 

2017). These  imbalances  are in  large part  attributed to  the  technologically aided  sex- selective 

termination of  pregnancies, even though the  practice has been outlawed  in both countries. We 

examine theoretically the  potential consequences for  the well- being of  both boys  and girls, the 

demographic transition, and economic growth.

A preference for sons arises from economic considerations and cultural norms (Mayer, 1999). 

Stronger patrilocal and patrilineal traditions coincide with a higher sex ratio in parts of Asia, the 

Middle East and North Africa (Basu, 1999; Jayachandran, 2015). In the absence of social security 

programs and  an efficient  financial system  by which  to save  for old  age, parents  rely on their 

children to  care for them.  In India,  China, and  South Korea traditionally  the son  supports his 

parents when they age, inherits the property, and continues the family line (Chung & Das Gupta, 

2007; Jayachandran, 2015). Hindu and Confucian traditions emphasize the son’s imperative role 

in caring  for his  parents in  old age  and in their  afterlife (Bhaskar,  2011; Chung &  Das Gupta, 

2007; Jayachandran, 2015). In contrast,  raising a daughter is considered  akin to “watering your 

neighbor’s garden"  since she  eventually moves in  with her  husband’s family  and cares  for his 

parents (Arnold et al., 1998; Guilmoto, 2009). Moreover, dowries in South Asia have increased in 

real value over time and are a financial burden on the daughter’s family (Basu, 1999).

The practice of  sex- selective abortion magnif ies the demographic effects  of more traditional 

methods of  gender control, specif ically stopping behavior whereby  parents have children until 

the  desired  number  of  sons  is  born,  and  even  infanticide  or  neglect  of  newborn  daughters 

(Bongaarts, 2013). Rapid increases in the SRB have followed geographical patterns of ultrasound 

technology diffusion, particularly throughout the 1980s in India, South Korea, and China. While 

South Korea’s sex  ratio has  returned to normal  since its  peak of  1.15 in  the early  1990s, India 

remains less  economically developed and  its SRB  continues to worsen  (Guilmoto, 2009; Rajan 

et al., 2017). As India modernizes and its demographic transition proceeds, in many regions par-

ents prefer to have fewer children, potentially intensifying the sex ratio imbalance (Das Gupta & 

Bhat, 1997; Dharmalingam  et al., 2014; Guilmoto, 2009).  Parents can effectively choose to  have 

more sons and fewer daughters. China’s experience has been similar, although some of its fertil-

ity decline can be attributed to state- mandated restrictions on family size via the so- called “one- 

child policy.” Falling fertility combined with son preference led to rapid increases in SRBs during 

the 1990s in  the South Caucasus,  reaching 1.17 in Azerbaijan  in 2002, 1.19  in Georgia in 1998, 

and 1.16 in Armenia in 2001 (Guilmoto, 2009).

Theoretically, falling  desired  fertility  could  improve  sex  ratios  if  the  ideal  number  of  sons 

declines more quickly than desired fertility (Das Gupta & Bhat, 1997). For example, Hindus con-

sider it sacramental  to give away one  daughter in marriage, while  sons, on the  other hand, are 

perceived as productive assets (Bhat & Zavier, 2003). Thus as ideal family size declines, the ideal 

number of  daughters  changes little,  while the  ideal number  of sons  changes more.  Bongaarts 

(2013) illustrates a positive relationship between desired fertility and son preference (i.e. desired