In American jurisprudence, two justifications have traditionally been put forth to support the government's social control of persons with mental illness: police power and parens patriae. As public mental hospitals became less available as loci in which to exercise these functions, governments sought alternative means to achieve the same ends. One prominent but quite controversial means is involuntary outpatient treatment (IOT). While the concerns about IOT have been myriad, one often alluded to but never documented is that of "net-widening." That is, once IOT became available, it would be applied to an ever greater number of individuals, progressively expanding the margins of the designated population to whom it is applied, despite the formal standard for its application remaining constant. We tested the net-widening belief in a naturalistic experiment in Massachusetts. We found that net-widening did not occur, despite an environment strongly conducive to that expansion. At this time, whatever the arguments against IOT might be, net-widening should not be one of them.