It must have sounded like a good idea at the time: rats were destroying sugar cane crops in the Caribbean Islands and in 1872 someone decided to introduce a foreign predator, the small Indian Mongoose, to kill the pests. Problem was, mongooses (not "mongeese," according to my Oxford English Dictionary) do most of their hunting during the day and the rats were most active in the sugar cane fields at night. Instead of staying up late to take care of the rats, the mongooses found it easier to kill chickens and snakes instead. (The latter, ironically, were already doing their part to keep the rats in check.) Today mongooses cause millions of dollars of damage in the Caribbean and are a major source for rabies and leptospirosis.

The point of all this is that sometimes the solution, no matter how well-intended, turns out to be worse than the problem.

A few days ago, the Government Accounting Office released a long-awaited report, Horse Welfare: Action Needed to Address Unintended Consequences from Cessation of Slaughter. Two years in the making, the 60-plus-page report manages to have something for people who favor the reintroduction of horse slaughter in the United States, something for people who do not, and probably nothing that will make anyone on either side of the fence entirely happy.

For me, the most telling parts of the report are the numbers—the horses slaughtered in the United States before the last processing plants in Illinois and Texas were closed in 2007 and the horses shipped across the borders to Mexico and Canada for slaughter since then. According to the GAO, horse slaughter in the U.S. peaked in 1990 (345,900 animals), dropped consistently until 1994, held steady for several years, then dropped again. The low point (41,134 horses) came in 2000, and then the numbers started to rise again. In 2006, the last full year for slaughterhouse operation in the States, the number topped 100,000 for the first time in more than a decade.

Not surprisingly, the number of horses being shipped to Mexico and Canada for slaughter increased dramatically when the U.S. plants closed. In 2010, according to the GAO report, 137,984 horses were exported to the north and to the south for slaughter. This was about the same number of horses slaughtered in the U.S., Mexico, and Canada combined in 2006. (Despite the apparent precision of the counting, the 2010 number is an estimate only. The actual figure could be higher, could be lower. No one knows for sure because official trade export figures for equine traffic between the United States and its neighbors do not specify either the quantity or value of horses exported for slaughter.)

So let’s assume that the GAO figure is reasonably accurate. This means that more horses were slaughtered in Mexico and Canada last year than were slaughtered in this country in any single year since 1993. The undeniable conclusion is that ending horse slaughter in the United States has not put even a small dent in the total number of horses slaughtered every year in North America. If that was the hope when the last U.S. plants closed in 2007, it has not been realized.

And considering the abysmal conditions under which horses bound for slaughter are shipped to Mexico and Canada, and the treatment of the animals once they reach their final destinations, ending slaughter in this country has been a pyrrhic victory at best. Opponents of slaughter have managed to push the problem outside our borders, but that doesn’t mean that the problem has vanished. Instead, more horses have been slaughtered, almost certainly under worse conditions, since the plants in Illinois and Texas closed.

Stopping slaughter in this country might be a valuable first step toward a solution—but only if legislation preventing the export of horses for slaughter follows. And we know how well that has worked out in Congress and in state legislatures. Until there is an enforceable ban on exports, and without some meaningful way to deal with overpopulation of horses in this country, ending domestic slaughter is turning out to be a solution that may be worse than the problem it was supposed to solve.