Carla Jacobs and other mental health advocates in California celebrated when "Laura's Law" was signed in 2002 by then-Governor Gray Davis. The law -- named for a young woman who was shot and killed while working at a public mental health clinic -- allowed the courts to require that people with severe mental illness accept outpatient treatment. It was hailed as landmark legislation that would plug "a huge hole in California's safety net" and a "critical step in helping the mentally ill and their families."
But advocates such as Jacobs have had to put away their party hats. Six years later, the law has been fully implemented in only one county and is a pilot in another. The state didn't fund Laura's Law, relying on its 58 counties to do so. And they haven't. "Having a law on the books that isn't implemented," Jacobs says, "is like having no law at all."
This is only one example of a much-heralded governmental effort that never gets very far. There are a variety of reasons -- both managerial and fiscal -- why this happens. Gary VanLandingham, director of Florida's Office of Program Policy Analysis and Government Accountability cites three good ones: funding failure, problematic planning, and simple muck-ups in the actual implementation.
The lack of funding isn't rare. Sometimes, as with Laura's Law, the money was never there in the first place. In other instances, it disappears prematurely. In Michigan, for example, where the number of state troopers has fallen off, plans for a "trooper school" with 50 new recruits in 2007 were cancelled when the state police ran over budget and had to lay people off. (The state is trying again with a plan this year to train 100 new troopers.)
At least when funding is the issue, there's always some hope that cash will come in eventually. It may be even worse when inadequate or ineffective planning dooms an idea that seemed to be worthwhile. The Government Evaluation Act in Maine, for example, gave policy committees the role of doing a performance review of agencies on a regular schedule. Apparently, the act's writers didn't realize that such committees don't have enough time to undertake this task in a meaningful way. So, while reviews are done, they are pro forma and haven't lived up to the legislation's original intent.
Even if an effort seems to be reasonably planned, a host of implementation gremlins can appear. For instance, a classic implementation problem comes when you need lots of cooperation between different entities. "The coordination tasks are burdensome and time-consuming," says VanLandingham, "and getting everyone to agree on something is virtually impossible." So, when Texas hired IBM for a data consolidation effort, it carefully took into account the need for IBM to have solid institutional wisdom about the state's agencies. Plans were made to transfer 500 state IT positions to IBM. But the IT directors of some agencies weren't happy about losing their best and brightest. They were true to the letter of the contract -- but not its intent: About 200 of the positions transferred didn't have employees in them; they were vacant. Not surprisingly, a just-released audit revealed a host of problems going on with the system.
When the federal government is one of the entities, things can get particularly hairy. In 2006, Michigan began pursuing $100 million in federal funding for a Medicaid waiver to expand access to health insurance. Both the executive branch and the legislature gave approval. With fingers crossed, the paperwork was sent to Washington, D.C., where it still resides some two years later. The feds never said no. But they never said yes, either. So, the effort languishes.
The classic work on many of these issues was written 35 years ago, but is still very much up to date. "Implementation: How Great Expectations in Washington are Dashed in Oakland," by Jeffrey Pressman and Aaron Wildavsky, focuses on the serious obstacles that emerge from "perfectly ordinary circumstances" even when there is little political controversy or disagreement. "People now appear to think that implementation should be easy; they are, therefore, upset when expected events do not occur or turn out badly," they wrote in their original preface. "We would consider our effort a success if more people began with the understanding that implementation under the best of circumstances is exceedingly difficult. They would therefore be pleasantly surprised when a few good things really happened."