Tuesday, March 22nd, 2011 - 13:10
There are a lot of good reasons for contracting out government services at the federal, state and local levels. And there are a lot of reasons why governments should not contract out services. Debates over the wisdom of this practice rage on – sometimes informed by cost-benefit analyses, sometimes by political whim, often by a desperate search for some way to alleviate fiscal pressures.
But that’s not the point of this column. Forget about whether or not cities, states or the federal government should use outsiders to perform any individual function. Our concern is this: When they do give up direct control over the delivery of some service, they need to make quite certain that the contractors deliver what was promised – and not just in terms of a set number of hours or effort, but rather the actual quality and results of the work done.
For some time now, the solution to this concern has seemed reasonably simple, on its face: transform traditional contracts into performance contracts that spell out the details of what’s really expected, and potentially tie incentives or disincentives to positive and negative outcomes. As the Government Accountability Office clearly explains it, “Agencies should structure performance work statements in contracts around the purpose of the work to be performed, that is, what is to be performed rather than how to perform it. For example, instead of telling the contractor how to perform aircraft maintenance or stating how many mechanics should be assigned to a crew, the contract should state that the contractor is accountable for ensuring that 100 percent of flight schedules are met or that 75 percent of all aircraft will be ready for flight.”
Truth, is though, that although there’s been a whole lot of talk about performance contracts, a surprising amount of it has amounted to little more than talk. “There is a big imbalance between how much writing there is about it and actually doing it,” says Jim Chrisinger, a former Iowa official and a now a consultant with the Public Strategies Group.
Performance contracting seems like such common sense, that we wondered why it wasn’t more prevalent. We spoke with a number of people who have observed the field and here were some of the obstacles they described:
- It’s not easy. It’s much easier to do contracts on the basis of inputs and outputs, and when it gets to measuring performance, the world becomes more complicated.
- Agencies, themselves, are sometimes nervous about being held to rigid performance expectations – even if it’s on the part of outside contractors.
- When a governmental entity goes to performance contracting, there’s heightened pressure on vendors. And many of those vendors have long-term, friendly, relationships with agency officials who are, in turn, less likely to want to rock the boat.
- If not negotiated properly, performance contracts can put financial strain on vendors.
- It can be very tricky actually specifying the outcome desired. Definitions of success are elusive at best, in many cases.
- Governments often don’t have the capacity and expertise necessary to actually negotiate these often-complex contracts
Of course, while these obstacles may exist, they’re far from insurmountable. Consider, for example, Franklin County, Ohio, with a population of 1.1 million, including Columbus. With the help of consultants at Weidner Inc., the county had long been engaged in building its capacity to do performance measurement. About ten years ago, the County decided to put about $25 million into 43 contracts with 30 different community and public sector organizations to deliver a variety of services connected to welfare reform. The following year those funds were cut dramatically, and yet work rolled on remarkably smoothly.
County officials attribute this to the use of performance contracts. “We based the continuation of all contracts on performance, according to Mary Lou Langenhop, who was the Director of Job and Family Services for Franklin County from 1998 to 2003. “For example, had the agency been meeting its job placement targets? If not, their funding was cut back, or eliminated . . .
“We gave the County Commissioners spreadsheets with the performance information, presented by agency. I met with each Commissioner individually, shared all of the performance information, and presented how the funding decisions were made. As one of them told me, ‘I don’t see how anybody can question this’.”
According to Susan Lewis Kaylor, vice president of performance and management for the Alcohol Drug and Mental Health Board in Franklin County, the old-fashioned model led to mission creep and often ended up in litigation. Since the county went to the performance contracting model, they’ve had no lawsuits in ten years. She attributes that to having the capacity to show providers the reasons for decisions. Moreover, the performance information gives political leaders the ability to defend their decisions, when they get pressured about a particular contracted-for expense.
One of the keys to the success of this effort, according to William Aaron, chief of consulting services and innovation at Weidner is that the measures used are reached through a consensus between the county and the vendors. “They bring in these folks and [discuss] provider stat measures with them as a way to ensure that they are getting what they need from the providers and as an early warning system,” he says.
That’s certainly one good way to get performance contracts to work well. But there are lots of others. For example, according to the GAO, agencies should “ensure that each standard is necessary, carefully chosen and not unduly burdensome. Failure to do so can result in unnecessarily increased contract costs. Agencies should also ensure that standards are not set so high that they could drive up the cost of service or too low that they may act as a disincentive to good contract performance.”
Some other good counsel from Weidners’ Aaron and others:
- Allow contractors to develop the most innovative and efficient methods to solve problems, rather than restricting them to approaches the government has used in the past.
- Have a balance of objective and more subjective measures, such as milestones, when results are difficult to quantify at the outset.
- Consider starting the process with a pilot When a few contractors can vouch for the benefits of performance contracting, it helps get others on board.
- Keep dialogue going in order to avoid having the performance contracts lead to a game of “gotcha,” with vendors.
- Make sure that both the programmatic agency and central procurement offices work together closely.
Picture courtesy of Franklin County Ohio Genealogy and History