Most elected officials take more pleasure in boasting about new money- saving initiatives than they do in telling anyone whether those strategies actually saved a nickel. We tested this proposition--in an admittedly non-scientific way--by doing an Internet search using the words "government efficiency savings." We discovered that lots of state and local officials--from both legislative and executive branches--extolled the anticipated benefits of their efforts to make government run "leaner and meaner." Hardly any revealed what happened to those predicted savings.

Sounds like politics as usual, right? But it's indicative of a more pernicious problem: One of the easiest ways to balance a budget is to simply predict savings that may or may not ever come to pass.

Mark Funkhouser, the well-respected auditor of Kansas City, Missouri, tells a story that's right on point. Back in the mid-1990s, the Kansas City budget was in trouble. The state Supreme Court had invalidated a use tax the government was collecting, and that left officials with an unexpected $13 million hole in the budget. City leaders mulled over their difficult choices. In the end, though, the city manager simply announced that he would achieve $10 million to $12 million in efficiency savings. "Efficiency savings! Of course, they never materialized," Funkhouser says. "We had the usual midyear budget crunch and had to scramble."

Episodes in which green eyeshades are traded for rose-colored glasses are common. In 2003, one of the ways leaders in Maine were able to balance the books for the state's Medicaid program was by relying on reforms of the program rather than by eliminating beneficiaries. Although the state made progress in some areas, such as reducing the growth of drug costs, the total savings turned out to be far less than leaders had hoped. Partially as a result, budgeters were left confronting more than $100 million in Medicaid-related shortfalls for both 2004 and 2005.

Medicaid is an area in which accurately predicting savings from reforms is particularly difficult. There are so many instances in which one action that appears to have distinct dollar savings results in unanticipated consequences that wind up costing money. "There is always an imaginary component to making Medicaid budget impact estimates," explains Lou Rossiter, health economist at the College of William and Mary. "You say 'Let's do something to cut the cost of emergency rooms,' and you end up increasing the costs of hospitalizations. Likewise, 'Let's change long-term care' and you see an increase in home health services."

While there are no hard and fast rules, agency leaders as well as budget officers seem far less likely to lean in the direction of overly optimistic estimates than are others in and around state government. One way or another, the budget office and the agency are going to have to get expenses in line with revenues by year's end. The pain of that process is an incentive to budget conservatively at year's beginning.

Edmund J. McMahon, senior fellow for tax and budget studies at the Manhattan Institute, theorizes that it's often consultants, vendors or advocacy groups that push the more dramatic savings figures. "Outsiders tend to overestimate savings, insiders tend to underestimate savings," he told us. "If you want to know what your realistic savings are, take what the outsiders suggest you'll save, take what the insiders say, and throw a dart precisely between the two."

McMahon suggests that the problem of over-optimism can often be found among state lawmakers and city council members. "Legislators tend to overestimate savings as well as revenues, especially when it's their idea," he says. Moreover, he notes that "some who haven't been in government or corporate organizations are likely to do simple arithmetic and assume it adds up." By way of example, McMahon looks at how some legislators might calculate the savings if they decided to cut their government's workforce by 20,000 positions. They don't, he says, "figure how long it actually takes to save money doing that. They add up the salaries and assume they're saving that much money the first day the pink slips are sent out."

One last thought. The real problem here may go somewhat deeper than the simple human tendency to take the easiest way out of a bad situation. Rather, it's that people aren't held accountable for the predictions they make when they turn out to be wrong. Worse yet, in many instances, nobody even checks to see whether they were right or wrong. Attention is paid when reforms or efficiency-related measures are developed and authorized. But once they're up and running, there's often no mechanism that requires a second look.

And that's dangerous. As the old saying goes, "It is good to hope, but bad to depend on it."