Table of Contents
When I wrote Creating Public Value, my intention was to offer a challenge and a source of inspiration for people who wanted to provide leadership in the public sector. At the time, there was, quite rightly, some criticism along the lines of: ‘Well, fine. One can readily talk about a concept of public value at a very high level of abstraction. But can we define the concept more concretely so that we can recognise objectively and empirically whether we are producing it?’ This seemed like an important challenge to address.
To respond to this criticism, I began to talk about the challenge of ‘recognising public value’ as well as imagining it and working to produce it. The word recognising seemed particularly suitable in talking about the challenge of measuring public value because it carried two different connotations – one intuitive, and one more technical.
The intuitive idea behind recognising public value is that we can collectively recognise a particular event or result as publicly valuable because we share a common intuitive notion of what we mean by public value. This is the meaning of ‘recognition’ suggested by United States Supreme Court Justice, Potter Stewart, who, when asked to define pornography, said, ‘I know it when I see it’. So, one idea of recognising public value suggests some emergent, communal social process through which a community comes to understand (perhaps through some kind of continuing dialogue with an organisation) what valuable effects an organisation produces. Where such a common view exists, an organisation can take advantage of the shared understanding of the public value it delivers and simply point to those effects, as in: ‘Look at that! We reduced crime in your neighbourhood,’ or, ‘See? The garbage has been removed from the curb’.
The second idea of recognising public value is far more technical. It derives from the field of financial accounting. The most basic requirement for financial accounting is the development of an administrative system that allows the organisation’s managers to recognise expenditures of the organisation’s assets and the receipt of financial assets by the organisation). These technical, administrative systems are constructed to render objective and consistent the movement of material and financial assets across the boundaries of an organisation and enable the use of those assets in productive activity. While standard financial accounting systems can capture financial flows through an organisation together with the costs expended by an organisation in producing particular products and services, it should be clear that if we want to measure the public value an organisation produces, we will have to construct some other technical system that can allow us to record when public value is being produced. Generally speaking, that requires us to develop some conceptual and operational basis for defining what constitutes the public value the organisation produces.
I will begin, then, by introducing the concept of ‘recognising public value,’ in both its abstract sense and its concrete manifestations.
One frequently cited example of the importance of developing the systems through which a community and an organisation can come to recognise public value comes from the New York City Police Department (NYPD). In January, 1994 Mayor Rudolph Giuliani announced the appointment of William Bratton as head of the NYPD.[1] Bratton came to the job saying he wanted to bring bottom-line management techniques to the NYPD. Correspondingly, like the CEO of a firm seeking to attract investment, he declared, first to the mayor and then to the public, that he was going to reduce crime by 10 per cent in New York City over the course of the next year. Somewhat surprisingly, this was a kind of promise that no police commissioner had ever made before.
Having promised this reduction, he exposed himself and his organisation to a substantial risk of failure. Necessarily, then, he began looking around for ways to engage other people in sharing the burden. He found a way to distribute responsibility for reducing crime throughout the NYPD by delegating responsibility for reducing crime to each of the NYPD’s 72 precinct commanders. He and his deputy commissioners then called those managers to account for their crime reducing performance in public meetings in which precinct commanders were castigated for their failures to perform and challenged to come up with new approaches, or praised for strong performance with respect to the goal of reducing crime. They measured crime control effectiveness by looking at trends in reported crime statistics disaggregated by precinct and crime type. Essentially, Bratton followed the basic principles of ‘Management 101’: define your operational goals, delegate responsibility for achieving them, set up a system for monitoring the performance of managers in achieving the desired results, and sanction the managers on the basis of their performance.
Amazingly, after establishing this system and laying out some new guidelines for patrol officers, reported crime rates fell substantially in the city. In fact, during the two-year tenure of Bratton and his management team, the murder rate in New York City dropped 50 per cent and the overall crime rate dropped 39 per cent. So, there was a rather astonishing moment when it looked as though we might be able to get significantly improved performance from police departments, or maybe even public sector organisations more generally, just by being sufficiently tough-minded and straightforward about measuring the performance of organisations with respect to their core mission. And thanks in part to this, the use of performance measurement as part of a performance management system in public sector organisations is now fundamental to the mainstream teaching of public administration.
Despite the apparent success of this initiative, and despite the widespread commitment to these simple uses of performance measurement, I want to move the discussion of performance measurement up a level or two, because I do not think the simple lessons of this case are the right ones to draw. I am quite confident that the way that Bratton chose to measure the public value created by a police department is wrong, and wrong in ways that will apply to many more public agencies than police departments. And I am also reasonably confident that the method he chose for using performance measurement to animate and guide his organisation is not necessarily the best. But it is hard to argue with success – particularly when I am basically in agreement that we need more and better ways to measure the performance of public agencies. What I propose to do, then, is to back up a few steps from these simple lessons and ask three harder questions about performance measurement in the public sector.
The first is: Why should a public manager measure performance? Perhaps another way to put the question is, ‘Why should public managers load a gun that others will use to shoot them?’ I regard this basic motivational question as very important for two simple reasons. On one hand, building a strong performance measurement system requires a huge investment. On the other, having such a system inevitably exposes a manager and his organisation to the risk of a clearly documented failure. Given these facts, any moderately self-protective manager might reasonably ask, ‘Why should I make this big effort when all it does is carry risk for me and for my organisation?’ Even if all the rest of us citizens thought public managers ought to measure performance as part of their general accountability to us, it would help if we could find a compelling reason for managers to commit themselves to doing rather than resisting it.
The second question is: Where, along what I call the ‘value chain,’ is it best to measure performance? More particularly, the question might be, ‘Should we measure inputs, processes, outputs or outcomes?’
The third question is: Should we measure customer satisfaction as an important dimension of public value creation, and, if so, who are the customers and what do they want?