The fundamental aim of most self-direction programs is not to save money but to give people with disabilities greater control over the services and supports they receive and when, by whom, and how they are delivered. In virtually all instances, however, the increase in personal control is accompanied by requirements that total service costs are not to exceed the costs that a community provider agency would incur in delivering the same array of services and supports. In some instances, the upper limit on self-directed support plans is set at 100 percent of the cost of provider-controlled services and supports; in other programs, a discount factor is applied to self-directed support plan allocations (e.g., 90% of provider agency costs) to be held as a "risk pool" of funds that can be used by the state or provider agency to meet unanticipated cost increases over the course of the year. As a result of such policies and the variability among self-directed programs across and within states, it is difficult to draw valid comparisons between the costs of self-directed versus agency-directed services.

Head and Conroy reported a median reduction of 8 percent in the cost of serving 70 participants in a self-determination demonstration program for people with intellectual and developmental disabilities in Michigan. Comparisons of expenditures on behalf of these individuals were made before they entered self-directed programs and again three years following their enrollment in the program. The savings increased to 14 percent when expenditures were adjusted for inflation over the three-year period, with the median public cost per participant declining from $67,322 to $56,778 in inflation-adjusted dollars. The study also found that participants reported that they had more and better choices, less professional domination, and a higher overall quality of life. The study did not analyze control or comparison group data, and therefore the authors warn against generalizing from the findings of this small, single-state study.³⁶

When personal care cost data of participants in the Cash and Counseling demonstration program were compared with those of a control group receiving agency-directed personal care services, researchers discovered that participants incurred higher costs primarily because program enrollees received more of the care they were authorized to receive than control group members. In addition, the increased personal care costs were partially offset by lower institutional and other long-term care outlays on behalf of Cash and Counseling participants. The evaluation team concluded that, if a state carefully designs and monitors its Cash and Counseling program, self-directed services should not cost any more than traditional, agency-provided services.³⁷  The Arkansas Cash and Counseling program saved $5.6 million after nine years of operation, not including the additional savings associated with reduced nursing home utilization.³⁸

In examining the experiences of states operating consumer-directed support programs for people with intellectual and developmental disabilities (I/DD), Walker found that cost savings usually are built into a state’s funding assumptions. Typically, a state either pays a set fraction (e.g., 90%) of the total amount allowed for traditional agency-directed services, or establishes a lower allowance for self-directed administrative/overhead costs than for agency-directed administrative/overhead costs.³⁹

Walker also points out that some states have created consumer-directed support programs with tight spending caps that are aimed at stabilizing families and preventing emergency out-of-home placements of individuals on a waiting list for full-time residential supports. By dampening demand for residential placements, this comparatively low-cost option allows a state to extend services to additional wait-listed individuals, thus reducing the gap between supply and demand. One I/DD program administrator estimated that his state was saving more than $1 million a year by offering low-cost self-directed support options to families caring for loved ones with an I/DD in their homes.⁴⁰

In summary, studies of the cost-effectiveness of self-directed services are few in number and generally have yielded inconclusive and sometimes contradictory results. The limited utility of such studies in shaping public policy can be attributed in large part to the cost assumptions underlying most existing programs and the methodological problems involved in conducting such research (e.g., accounting for (and weighting) all of the relevant cost variables that influence outcomes).