In the late 1990s, as an alderman in Somerville, Massachusetts, Joseph Curtatone was perpetually frustrated by the budgets his local legislature was supposed to be helping shape and approve. "Budget time really used to get me," he recalls. "It was a straight line-item budget. There might be a small paragraph for each department briefly describing what they do, but there was nothing that told you how much we spent on what -- no inputs, outputs or outcomes."
It was classic best-guess budgeting, coupled with a typical tactic: Any department that had any money left in its account at year's end was guaranteed to get a budget cut. "If you're the DPW director and you have $15,000 left in your account, then that's how much we cut your budget for the next year," Curtatone recounts. "So the message to all our departments was, 'Spend down your budget.'"
While this was one of the most significant points of aggravation, Curtatone also was nettled by how the city managed its decisions in general. "Here we were, a multimillion-dollar operation with absolutely no real-time information on even the most basic services. We weren't measuring anything. How many potholes were we filling? How were we filling them?"
The simple fact that Curtatone harbored such frustration, though, made him an unusual breed of elected legislator. Nationally, those who follow budgeting and managing for results in the public sector note that his impatience with a lack of information about what the city was trying to accomplish, and whether it was getting important jobs done efficiently, set him apart from most legislators -- at the local and state levels. Even in states with results-based budgeting and management legislation on the books, and even in cities where mayors are preaching performance, the last public officials who seem to tune in are state legislators and city council members.
Those in the trenches agree. When asked to comment on the average city council member's familiarity with the concepts surrounding performance measurement, Jackie Nytes, an eight-year veteran of the Indianapolis City Council, lets out a long sigh. "Theoretically, if I ask city council members if they're in favor of using data to make decisions, they'll say, 'Of course.' But a lot of that is just lip service. A lot of council people are part timers, and the fact is that making budget decisions based on information about results is just harder."
But if her counterparts around the country are going to deal effectively with delivering services in times of tight resources, she thinks the way to do that is elevate the conversation. "If we're really going to get a handle on budgeting, then the question shouldn't be, do we have enough money for this many jail beds? The question should be, do we have the programs in place so that we don't need so many jail beds? If legislatures are acting responsibly, then they should be focusing on those kinds of outcomes."
Nytes is preaching that ethic not only locally but nationally. She's part of an effort being sponsored by the National League of Cities, aimed at tuning in elected officials to a much more performance-based view of government programming and budgeting. As chair of the CityFutures Finance Panel at NLC, Nytes is part of a team of local elected officials who are working with performance-measurement guru Harry Hatry of the Urban Institute to produce a manual and other materials by this fall. The aim is to help local elected officials to start focusing on the bigger, performance-informed picture.
The partnership is similar to one in which Hatry was involved a few years ago with the National Conference of State Legislatures. The result of that collaboration was "Legislating for Results," a basic primer that walks legislators through the fundamentals of data-driven decision making. According to Judy Zelio, who follows fiscal affairs for NCSL, there has been an uptick of interest in legislating for results in the wake of the book's publication. She also credits the buzz around work done in the field in states such as Louisiana, Oregon, Texas and Washington.
Mike Marsh, deputy director of the Oregon Department of Transportation, says it historically has been a mixed bag when it came to what legislators wanted to focus on when his department presented data on goals and outcomes. This year, however, in preliminary discussions with the chairs of the House and Senate budget committees in Salem, officials from his department were told to be ready to link dollars and outcomes. "They also said that if requests for additional money weren't related directly to key performance measures, don't even bother asking," Marsh adds.
Meanwhile, the committee chairs seem to be holding their own folks accountable for familiarizing themselves with ODOT's facts and figures on costs and accomplishments before they come to the budget hearings, Marsh notes.
Getting legislators to really buy in to legislating for results has never been easy. Even as Texas was being touted as a model of performance-based budgeting back in the late 1990s, legendary lawmakers and budget wranglers such as Lieutenant Governor Bob Bullock were referring to those efforts as "window dressing." In his view, the legislature knew from experience which agencies were doing a good job and which weren't.
Certainly some legislators have enough institutional memory and familiarity with their agencies to make informed judgments about performance. But especially in an era of term limits, and as departments become more technically adept at measuring what they do, advocates of data-driven government argue that something more than an agency's reputation for good or bad work ought to go into the spending equation.
There are some legitimate reasons why legislators might be uncomfortable using data to make big-dollar decisions, however. The most prominent among them: They don't trust agencies to honestly and accurately report what's going on.
In 1997, Maryland embarked on its "Managing for Results" process. "MFR was introduced to agency leadership as a strategic-planning process to help officials set goals, objectives and performance measures for programs and to assess the results of those programs," says a 2004 legislative auditor's status report.
The auditor's report, which was undertaken at the request of the chairs of the Senate and House budget and appropriations committees, noted some key reasons why the program might not be as effective as it could be. For one thing, the number of measures reported to the Department of Budget and Management was closing in on an astounding 9,000 for 64 state agencies. At the same time, the audit report pointed out that the Governor's Office itself had been doing a routinely poor job of linking budget requests with agency goals and results. Most significant, the report found that almost two-thirds of agency-reported data simply was not verifiable.
Despite such problems, Maryland Senator P.J. Hogan, who serves on the budget and tax committee, contends that he and his fellow committee members do pay close attention to the link between money and key results when they make budget decisions. "I find that MFR works very well for me as a budget person. I can look at an agency's budget and look at their goals and see how they're doing and make a better judgment as far as whether we're getting the right amount of bang for the buck," says Hogan. "Take a measure like infant mortality rates. Under our MFR guidelines, we can look and see what it was in the four previous fiscal years and at least see trends -- hopefully going in the right direction. And if not, then we can say, 'Hey, this program is not working, let's pull the funding.'"
Meanwhile, the executive side of Maryland government is promising to tighten up its MFR act under newly elected governor -- and former Baltimore mayor -- Martin O'Malley. Not surprisingly, O'Malley is taking the principles and tactics he practiced in creating and administering Baltimore's much-vaunted CitiStat and applying them to Maryland through "StateStat."
While legislation was pending to make StateStat official this session, O'Malley was already pushing it in three key agencies: corrections, juvenile justice and social services. Hogan doesn't see StateStat as either bolstering or overriding MFR. "I see it as complementing it," he says.
Meanwhile, Hogan concedes that most other legislators in Maryland have only passing familiarity with the whole MFR regimen, because they are less focused on the budgeting side of legislating and are more into "establishing new programs."
Despite the continuing state of general ignorance or indifference when it comes to using data to make important decisions, the ranks of governments that are turning to more results-based programming and budgeting seems to be slowly adding up, frequently because one or two people, such as Curtatone or Nytes or O'Malley, become frustrated enough to try moving government in a new direction.
For Curtatone's part, he took the most direct route possible to that position of influence: running for Somerville mayor. Upon taking office in January 2004, Curtatone promptly organized a series of field trips to Baltimore with top staff. The result was "SomerStat," which he says is now standard practice in his city. "For the first year, our aldermen were saying, 'Oh yeah, SomerStat. Explain that to me again.' " His early budgets had both line items and performance-related costing to ease them into the program. Now, using data to discuss budgeting has become standard operating procedure within his local legislature.
What hasn't happened in Somerville yet, Curtatone says, is taking the information being collected on government performance and results and pushing it out to citizens so that they can be more tuned in to what's being accomplished with their tax dollars. That's the city's next big step.
Some elected legislators think that is the best reason of all for making a clear connection between dollars and results. "Revenues are getting tighter and tighter, and we're not going to micromanage our way through this," Nytes says. "We have to reframe the discussion with the taxpayer. This isn't about looking for fluff in budgets, for waste. We're already efficient. The question is, are we efficient at the right things? So this is about what we want to budget for: What are our priorities and what do those cost, and then explaining to citizens the tough choices. If you want all these services, then you may need to increase taxes to pay for them."
Meanwhile, there is evidence that pushing performance -- and transparency in government -- can actually have significant political payoffs. Kansas City auditor Mark Funkhouser parlayed his reputation as a tough protector of taxpayers' interests into a successful mayoral run, winning a close election against a popular incumbent city councilman in March. Funkhouser's high-profile and tough performance audits were frequently so nettling to the mayors he worked for that he was directly threatened with being fired if he didn't back off. He ran on a campaign of government transparency and accountability.
And in Somerville, Curtatone says he ran for mayor on a straight platform of "performance-based budgeting and costing out and measuring activities and results." He adds with a laugh that a certain mayor just across the Charles River in Boston "doesn't believe in all this Activity-Based Costing stuff -- yet." But in Somerville, citizens were certainly ready to support a mayor who knows his ABCs.
These articles are part of a continuing series on public performance measurement focusing on citizen involvement. Support has been provided by the Alfred P. Sloan Foundation. Although the foundation may assist some of the programs described in these articles, it had no control or influence over the editorial content, and no one at Sloan read the material prior to publication. All reporting and editing was done independently by Governing staff.