This pattern is repeated everywhere that breed bans have been enacted and enforced, or at least places [where] records of bites are kept. Initially, the number of dog bites do often go down, mostly because owners move, give up their pets, or have their dogs impounded or euthanized. Then (as people get dogs to replace the pets they had to give up or put down) the number of attacks levels off and then begins to rise, possibly because the breed ban creates a false sense of security: Certain breeds are banned because they are dangerous; I don’t own a dangerous breed, therefore my dog won’t hurt anybody.
Looking into the data on breed bans turns up an interesting finding: each region that has enacted breed-specific legislation of some sort appears to have also experienced significant, if not dramatic demographic changes over a relatively short period of time. For example, Denver, Colorado (a city that has one of the toughest breed bans in the United States — and routinely ranks among the highest in the nation in dog bites) has had breed-specific legislation in place since 1989, not long after the “energy bust” and associated migration dramatically changed the population of the city. Ditto for nearby Aurora, which put its own breed ban in place in 2005; Miami-Dade County, Florida, 1989; Yakima, Washington, 2001; Ontario, Canada, 2005; and so on.