Consider two stories—both of them horrible. The first involves a woman who was a member of the opposition Democratic Party in Albania. In the mid-1990s, she was arrested at a political rally and held without charge. Shortly thereafter, she was kidnapped by four masked men on the street and forced into prostitution. Her captors, one of them a policeman, told her that she was being "punished" for her involvement with the Democratic Party and branded her leg, in case she escaped. Eventually, she was rescued by her fiancé and returned home, but, after her fiancé was attacked by masked men, they left the country—first to Italy, then to the United States.

The second story involves another young Albanian woman, "Ariana," who was kidnapped with her cousin in 1999 by two masked men, raped at gunpoint, and then driven to a vacant building with five other women where they were raped and beaten for a week. The women were then dragged onto headed to Italy, where they would be forced into prostitution, but the boat was intercepted and Ariana was freed. She filed a police report, but nothing ever came of it. Soon after, her family started receiving regular threats over the phone by men who knew Ariana's name. After her cousin was kidnapped yet again, and gunmen broke into her home, she fled to the United States.

What's the difference here? Well, the first woman was granted asylum by the Bureau of Immigration; the second was not. In both cases, the immigration judges found the women's stories credible, but, in the second, the judge ruled that Ariana had been "randomly targeted," and had no proof that the kidnappers had any "personal animus" toward her. Of course, there did seem to be personal animus at work—Ariana and her cousin were repeatedly targeted by men who knew her name—but in any case, that's irrelevant. For asylum purposes, the question should have been whether she would be targeted for persecution if she returned to Albania. And the answer is: Yes, very likely.

Now, anyone can nitpick with an individual judge's decision in this or that case. What we want to know is if there's a systemic problem here. The Center for Gender and Refugee Studies (CGRS) has just put out a report examining nearly 100 cases in which trafficking was a key part of an applicant's claim to asylum. The center found that the trend is disturbing—decisions by immigration courts are "heavily weighted toward denials." In fact, most of the cases seem to run counter to the criteria for asylum laid out in U.S. law:

Refugee status may be established by showing past persecution or a well-founded fear of future persecution. An applicant seeking to qualify for asylum based on past persecution must show: "(1) an incident, or incidents, that rise to the level of persecution; (2) that is 'on account of' one of the statutorily-protected grounds; and (3) is committed by the government or forces the government is either 'unable or unwilling' to control."

The problem here is that most immigration judges seem to treat women fleeing trafficking as victims of ordinary crime. On the other hand, if a woman can prove that she was persecuted in part because of her political beliefs—as with the first case I mentioned—then judges will look more favorably on their cases. But this isn't always easy to prove. A huge plurality of cases involves Albania, which has a serious trafficking problem, and there's good reason to think that many women are targeted for political reasons. But they can't always prove it. Strikingly, Albanian women seeking asylum are rejected at a much higher rate than average.

Reading through the case studies, it looks like judges also tend to argue that many women aren't being targeted for trafficking because they belong to a specific social group, and hence don't qualify for asylum. CGRS argues that this is misguided: Previous court rulings have declared that sex and gender can constitute a "protected" group, and various international rulings have, rightly, pointed out that women are often targeted for trafficking because of "their vulnerability in certain social settings." There's something wrong with this picture, and even if immigration judges aren't to blame, then the policy seems badly conceived.