Challenges for the New Chief Performance Officer
Wednesday, January 7, 2009; 1:00 PM
President-elect Barack Obama's choice today of Nancy Killefer as our government's first chief performance officer (CPO) raises one obvious question: "What will a federal CPO do?" The details of Killefer's role and responsibilities remain fuzzy, but one key to her success is clear.
When it comes to government performance, one of the best ways to improve it will be to improve the way we measure it.
(For the sake of full disclosure, I should note that Killefer sits on my organization's board of directors.)
Right now, our federal government relies on lagging indicators to let us know how our government is doing. When Bernie Madoff's $50 billion Ponzi scheme hits the news, we realize the Securities and Exchange Commisssion needs to step up its enforcement. When toxic toys keep getting recalled, we know that the Consumer Product Safety Commission needs an overhaul. And when thousands of people are stranded at the Superdome after Hurricane Katrina, we know that FEMA is broken.
This pattern of discovering a federal agency isn't doing its job because something went terribly wrong needs to be reversed. The best way to do so is creating a system of effective, operational indicators to identify what parts of government are working in real time, and what needs to be fixed. In other words, we need leading indicators, not lagging ones.
Measuring performance isn't new to the federal government. Starting with the Government Performance and Results Act (GPRA) in 1993 through the Bush administration's Performance Assessment Rating Tool (PART), there has been a concerted push to improve measurement in government. While each of these initiatives had particular strengths, problems with design and implementation have limited their impact.
There are two fundamental problems with how our government currently approaches measurement.
First, the measures are not operationally useful. Most federal managers view our government's performance measures as a paperwork exercise to keep them from getting into trouble with the Office of Management and Budget. Even though PART is our government's primary performance management tool, only 23 percent of federal managers report having knowledge of its details, and just one in four members of this small subset say they use PART data when making decisions.
They don't use it, because they often aren't measuring the right things. In this regard, our federal government is very much like major league baseball in the pre-Bill James era. James was a numbers geek who proved that many of the most popular statistics used to measure a player's worth, such as wins for pitchers and runs batted in for hitters, are in fact poor indicators of his true value. Instead, teams should pay attention to statistics like a batter's on-base percentage and a hurler's strikeout-to-walk ratio. For the most part, our federal government is still looking at wins and RBIs.
The second big problem with the measurements we currently use is that they are not meaningful to the public and external constituencies. Currently, our government focuses on reporting the activities of particular agencies, such as how many inspections of slaughterhouses were conducted by the Agriculture Department last year or how many food companies were inspected by the Food and Drug Administration. It should instead emphasize public outcomes and results -- whether there were spikes in food-borne illness -- rather than programmatic activities.
There is little doubt that developing real performance measures for our federal government that are both outcome-oriented and in line with public expectations will be a major challenge. Creating performance measurement tools is much harder in the public sector than private business where one cannot simply look at profits as a yardstick of success. But it can and must be done.
There are some rules of the road the Obama administration and its chief performance officer could follow to enhance the chances of success.