Missing women

In the context of human demographics, the term "missing women" indicates a shortfall in the number of women relative to the expected number of women in a region or country. It is most often measured through male-to-female sex ratios, and is theorized to be caused by sex-selective abortions, female infanticide, and inadequate healthcare and nutrition for female children. It is argued that technologies that enable prenatal sex selection, which have been commercially available since the 1970s, are a large impetus for missing female children.
File:Schnorr von Carolsfeld Bibel in Bildern 1860 083.png|thumb|right|Benjaminites seize wives from Shiloh in this 1860 woodcut by Julius Schnorr von Karolsfeld. There were not enough women available for marriage due to the high losses in the Battle at Gibeah.
The phenomenon was first noted by the Indian Nobel Prize-winning economist Amartya Sen in an essay in The New York Review of Books in 1990, and expanded upon in his subsequent academic work. Sen originally estimated that more than a hundred million women were "missing" or "gone". Later researchers found differing numbers, with most recent estimates around 90–101 million women. These effects are concentrated in countries typically in Asia, the Middle East and northern Africa. Economists such as Nancy Qian and Seema Jayanchandran have found that a large part of the deficit in China and India is due to lower female wages and sex-selective abortion, or differential neglect. However, the disparity has also been found in Chinese and Indian immigrant communities in the United States, albeit to a far lesser degree than in Asia. Some countries in the former Soviet Union also saw declines in female births after the revolutions of 1989, particularly in the Caucasus region. Also the Western world saw a dramatic drop in female births since the 1980s.
Researchers have also argued that other diseases, HIV/AIDS, natural causes, and female abduction are also responsible for missing women. However, son preference, as well as associated reasons to care for male well-being over female well-being, is still considered to the primary cause.
In addition to the health and wellbeing of women, the missing women phenomenon has led to an excess of males in society and an imperfectly balanced marriage market. Because of the association of missing women with female neglect, countries with higher rates of missing women also tend to have higher rates of women in poor health, leading to higher rates of infants in poor health.
Researchers argue that increasing women's education and women's employment opportunities can help decrease the number of missing women, but the effects of these policy solutions differ greatly between countries due to differing levels of ingrained sexism between cultures. Various international measures have been instituted to combat the problem of missing women. For example, to bring awareness to the problem of missing women, the OECD measures the number of missing women through the "son preference" parameter in its SIGI index.

The problem and prevalence

According to Amartya Sen, even though women make up the majority of the world's population, the proportion of women in each country's population varies drastically from country to country, with various countries having fewer women than men. This runs contrary to research that females tend to have better survival rates than males, given the same amount of nutritional and medical attention. To capture this divergence from natural sex ratios, the count of "missing women" is measured as a comparison of a country's male-to-female sex ratio compared to the natural sex ratio. Unlike female mortality rates, estimates of missing women include counts of sex-specific abortions, which Sen cites as a large factor contributing to the disparity of sex ratios from country to country. Furthermore, female mortality rates fail to account for intergenerational effects from female discrimination, while a comparison of a country's sex ratio to natural sex ratios would.
Sen's original research found that while there are typically more women than men in European and North American countries, the sex ratio of developing countries in Asia, as well as the Middle East, is much higher. For example, in Mainland China, the ratio of men to women is 1.06, far higher than most countries. The ratio is much higher for those born after 1985, when ultrasound technology became widely available. Using actual numbers, this means that in Mainland China alone, there are 50 million women "missing" – that should be there but are not. Adding up similar numbers from South and West Asia results in a number of "missing" women higher than 100 million.
According to Sen, "These numbers tell us, quietly, a terrible story of inequality and neglect leading to the excess mortality of women."

Estimates

Since Sen's original research, continued research in the field has led to varying estimates on the total numbers of missing women. Much of this variation is because of underlying assumptions for "normal" birth sex ratios and expected post-birth mortality rates for men and women.
Sen's original calculations using 1980s and 1990s data for missing women were indexed using the average sex ratio in Western Europe and North America as the natural sex ratio, through assuming that in these countries men and women received equal care. After further research, he updated these numbers with Sub-Saharan African sex ratios. Using these countries' sex ratios as the baseline and male-female populations from other countries as the data, he concluded that over 100 million women were missing, primarily in Asia. However, later authors pointed out that Europe tended to have higher rates of male mortality due to multiple wars and generally risky behavior. Due to male workers migrating from rural to urban regions, immigration, and world war, a culture of "high masculinity" existed in these countries, while on the other hand, in other countries such as India, traditions regarding the discriminatory treatment of female children were stronger from the late 1950s to mid-1980s.
As a result of this disparity between countries, American demographer Coale re-estimated the Sen's original numbers of missing women using a different methodology. Using data from his Regional Model Life Tables, Coale found that the natural male-to-female sex ratio, accounting for different country fertility rates and circumstances, had an expected value of 1.059. Using the number, he then arrived at an estimate of 60 million missing women, much lower than Sen's original estimate. However, a few years later, re-calculated the count of missing women using Coale's methodology with updated data. He found 69.3 million missing women, which was higher than Coale's original estimate. He also noted a problem with the Regional Model Life Tables; they were based on countries with higher female mortality, which would bias Coale's numbers of missing women downwards. Furthermore, Klasen and Wink noted that both Sen's and Coale's methodologies were flawed because Sen and Coale assume that optimal sex ratios are constant across time and space, which they are often not.
Klasen and Wink conducted a study in 2003 with updated census data. Using life expectancy to instrument for sex ratio at birth, they estimated 101 million missing women across the world. Overall they found trends that showed that while West Asia, North Africa and most of South Asia saw more equal sex ratios, China's and South Korea's ratios worsened. In fact, Klasen and Wink noted that China was responsible for 80% of the rise in missing women from between 1994 and 2003. Sex-selective abortions were given as reasons for the lack of improvement in India and China, while women's growing educational and employment opportunities were cited as reasons for the ratio improvement in other previously low ratio countries such as Sri Lanka. Klasen and Wink also noted that similar to both Sen's and Coale's results, Pakistan had the world's highest percentage of missing girls relative to its total pre-adult female population.
Later estimates have tended to have higher numbers of missing women. For example, a 2005 study estimated that over 90 million females were "missing" from the expected population in Afghanistan, Bangladesh, China, India, Pakistan, South Korea and Taiwan alone. On the other hand, Guilmoto in his 2010 report uses recent data, and estimates a much lower number of missing girls in Asian and non-Asian countries, but notes that the higher sex ratios in numerous countries have created a gender gap – shortage of girls – in the 0–19 age group. A table summarizing his results is below:

Country	Gender gap 0-19 age group	% females
Afghanistan	265,000	3
Bangladesh	416,000	1.4
China	25,112,000	15
India	12,618,000	5.3
Nepal	125,000	1.8
Pakistan	206,000	0.5
South Korea	336,000	6.2
Singapore	21,000	3.5
Vietnam	139,000	1

Differences within countries/states

Even within countries, the prevalence of missing women can vary drastically. Das Gupta observed that the preference for boys and the resulting shortage of girls was more pronounced in the more highly developed Haryana and Punjab regions of India than in poorer areas. This prejudice was most prevalent among the more educated and affluent women and mothers in those two regions. In the Punjab region, girls did not receive inferior treatment if a girl was born as a first child in a given family, when the parents still had high hopes for obtaining a son later. However, subsequent births of girls were unwelcome, because each such birth diminished a chance of the family having a son. Because more affluent and educated women would have fewer offspring, they were therefore under more acute pressure to produce a son as early as possible. As ultrasound imaging and other techniques increasingly allowed early prediction of the child's sex, more affluent families opted for an abortion. Alternatively, if the girl was born, the family would decrease her chance of survival by not providing sufficient medical or nutritional care. As a result, in India there are more missing women in developed urban areas, than in rural regions.
On the other hand, in Mainland China, rural areas have a larger missing women problem than urban areas. Mainland China's regional differences lead to different attitudes towards the one-child policy. Urban areas have been found to be easier to enforce the policy, due to the danwei system, a generally more educated urban population – understanding that one child is easier to care for and keep healthy than two. In more rural areas where farming is labor-intensive and couples depend on male offspring to take care of them in old age, males children are preferred to females.
Even developed countries face problems with missing women. The bias against girls is very evident among the relatively highly developed, middle-class dominated nations and the immigrant Asian communities in the United States and Britain. Only recently and in some countries have the development and educational campaigns begun to turn the tide, resulting in more normal gender ratios.