In an October Slate column, Steven E. Landsburg deduced from an array of data that parents, on average, prefer sons over daughters. His evidence lay in a few recent studies that show that daughters have a slight but marked tendency to break up (or else forestall) marriages while sons tend to keep them together. But it turns out there's a fascinating fork in the statistical trail of bread crumbs.
For years, it's been common currency in adoption circles that girls are far more popular than boys among adoptive parents. Now there's data to confirm it, which has prompted another round of speculation about gender preference among parents—an issue that is bound to rouse more interest, and concern, as the era of assisted reproduction progresses.
This past August, the Census Bureau released an unprecedented report comparing adopted, biological, and stepchildren based on results from the 2000 Census—amazingly, the first census to differentiate between these groups. First of all, the report found that there are about 105 boys for every 100 girls in the general population of biological children under the age of 18. Adopted children, it turns out, present a very different picture, with a "sex ratio"—the sociologists' term—of 89 boys for every 100 girls. What's more, adopted children under the age of 6 constitute a group where there only are 85 boys for every 100 girls. (The Census Bureau reports that stepchildren—a sizable population whose sex ratio is closer to the norm—are usually adopted at later ages than orphans are. Hence the under-6 drop-off.)
Last but not least, the sex ratio of adopted children goes still furtheroff-kilter if you look only at international adoptions. The Immigration and Naturalization Service (now part of the Department of Homeland Security) has kept up excellent data on international adoptions over decades of processing visa paperwork. Its word: Girls make up about 64 percent of all children adopted by Americans outside the United States. That's a mere 56 * boys for every hundred girls.
What explains the disproportion? If we didn't know better, the most obvious conjecture would be that these numbers simply reflect an imbalance in supply. After all, America's leading source of adoptees is China, where the legacy of female infanticide is the grimmest hallmark of that country's overwhelming preference for males. The organization Families With Children From China reports that about 95 percent of children available for adoption in China are girls. Other Asian adoption hubs (like Korea, the erstwhile lead supplier) have orphan sex ratios that tend in the same direction. So Americans adopt more girls because other countries don't want them, right?
Wrong. Unlike biological parents, who must simply make do with what the procreative coin toss affords them—as in a market determined solely by supply—adoptive parents get to be upfront about their gender preferences. And a look at those preferences suggests that, in fact, the adoption market in China represents a happy coincidence of supply and demand.
Numbers vary, but it's pretty safe to say that somewhere between 70 percent and 90 percent of parents looking to adopt register some preference for a girl with an agency. It doesn't matter if they're adopting from China, where girls far outnumber boys; from Russia, where the numbers are about even; or from Cambodia, where there is typically a glut of orphan boys and a paucity of girls. Everywhere, demand tends to favor the feminine.
And, as the case of Cambodia suggests, demand can in fact exert an influence on supply—and not a happy one. In the late '90s, Cambodia became a popular source for American adoptions, thanks to a relatively quick, cheap, and tidy process. But for whatever reason (some cite a Cambodian tradition that girls are expected to take care of their parents when they get older), Cambodia didn't offer the standard Asian profile of adoptable children. Boys outnumbered girls by a healthy margin. So what happened was what you would expect to happen in an underpoliced free market: Market pressure built up, until certain enterprising Cambodian adoption suppliers, or "facilitators," stepped in and found a way to meet demand.
Evidence of child-trafficking came to light in late 2001 and early 2002, when several poor Cambodian women stepped forward saying they had been approached by someone from an "NGO" who offered them a sum of money—significantly more for a daughter than for a son, though never more than $200—in exchange for their children. When that "NGO" turned out to be an orphanage, the U.S. Embassy and the then-INS slammed the gates on all U.S. adoptions out of Cambodia. They haven't reopened the gates yet.
Scholars inside the adoption community are quick to admit that the historical aura of secrecy surrounding adoption has hobbled research efforts to account for the decided preference among parents for girls. Still, there are a few decent indicators. First, there are certain norms and stereotypes peculiar to the world of adoption that have been wafting around since adoption became a modern institution. Take, for example, the following quote, an excerpt from the 1916 annual report of the Spence Alumni Society, one of the very first American adoption agencies: "Why do so many people prefer girls! The majority seem to feel that a girl is easier to understand and to rear, and they are afraid of a boy."