How serious is son preference in China?

At the time of the census in 2005, almost 121 boys were born for every 100 girls. Last year's census showed that sex ratio at birth (SRB) had improved somewhat. But it is still too early to celebrate, in Eklund's view: the narrowing of the gap does not necessarily mean that girls are valued more highly.

Because of the high SRB, there has been a tendency to picture China as a country where son preference is strong and possibly increasing since the 1980s. However, Eklund argues in her PhD thesis that using SRB as a proxy indicator for son preference is problematic. She has therefore developed a model to estimate what she calls "son compulsion," where data on SRB and total fertility rate are used to estimate the proportion of couples who wants to give birth to at least one son and who take action to achieve that goal. When looking at variation in son compulsion over time and between regions, Eklund finds that new patterns emerge that do not surface when using SRB as a proxy indicator. Contrary to popular belief, son compulsion remained steady in rural China (at around 10 per cent) while it increased in urban China in the 1990s (from 2.8 per cent to 4.5 per cent).

"This doubling concurred in time with cuts in the state welfare system in the cities, which meant that adult sons were given a more important role in providing for the social and financial security of the elderly," she says. Her findings call into question the assumption that son preference is essentially a rural issue. They also have implications for comparative perspectives and her findings suggest that son compulsion may be higher in other countries even though they expose lower SRB.

When it emerged that far more boys than girls were being born in China, the Chinese government launched the Care for Girls Campaign to improve the value of the girl child and to prevent sex-selective abortion. Nonetheless, the imbalance between the sexes continued to increase. Eklund's findings suggest that the campaign may actually have done more harm than good. Families receive extra support if they have girls and in rural areas exceptions are made from the one-child policy if the first child is a girl.

"By compensating parents of girls in various ways, the government reinforces the idea that girls are not as valuable as boys," says Eklund.

Eklund further challenges the notion that families in rural areas want sons because sons are expected to take over the farming.

"That is a weak argument," says Eklund. "Young people, both men and women, are moving away from rural areas. Of those who stay, women provide just as much help as men. In fact, it is the elderly who end up taking greater responsibility for the agriculture."

However, there are also other reasons why sons are seen as more important for families. Traditionally, a girl moves in with her husband's family when she gets married and she thus cannot look after her own parents when they grow old. Boys also play an important role in ancestor worship, and they ensure that the family name lives on.

Eklund further finds that there is a stubbornness in both popular and official discourses to view son preference as a matter of parents and grandparents without looking at structural factors that help underpin the institution of son preference.