January 21, 2008

Police Pay

I must confess to a personal quirck here- I am fascinated by how you effectively tie pay to performance in the public sector. The Institute of Public Policy Research this morning has announced that in February it will publish a report covering police pay, unfortunately as we do not have the report itself but only an executive summary on the website of the Institute there isn't much we can say. However the reccomendations on that website- which one assumes will be central to the report raise serious questions about its contents. The central reccomendations, reported this morning by the BBC, concern the introduction of performance related pay into the police service. The IPPR point out that the rates of crime detection per officer in the UK have hardly moved since 1997 despite huge increases in pay for officers and that furthermore pay within the force does not reward performance but rewards seniority. They want to shift the balance of pay to reflect performance and to get officers to train more effectively.

All of that is laudable as an aim but there are some serious questions about it that deserve to be raised. Firstly there is the obvious structure and predictability of performance related pay- how it fluctuates for individual officers year on year. No doubt the IPPR would be keen to argue that it should not fluctuate too much- uncertainty of pay award is just the kind of thing to drive talented and therefore useful people away from the police force just at the moment when we most need them. But the real issue is a second one. The problem with performance related pay is never the concept but the metric. Its the way that you measure performance. A classic case can be seen in the IPPR's own research. They argue that police performance should on their website be measured in terms of arrests per officer- but of course there are other ways of measuring police performance. As advocates of the 'bobby on the beat' will often tell you the provision of a sense of security to the public is another measure of police performance for example. The idea of performance related pay risks skewing the performance of police to reflect one or two or three different metrics. The IPPR imply that performance's definition will be decided locally- in which case one has to ask why they use a central figure of police arrests to demonstrate failure and in which case one has to ask furthermore are they willing to see the price of localisation (local failure) to be paid.

None of this is to say that their report is neccessarily wrong- we can't, its not out yet (and quite what the IPPR are doing in releasing to the press a report that hasn't been published, getting publicity for the argument before they get criticism for the research I'm not sure) but it will be interesting to see how the institute has managed to square some of these circles. In particular it will be interesting to see how they manage to derive a concept of performance that doesn't warp the performance of police away from things that we want policement to do. In general within the public sector- the problem is also there with teachers and doctors- there are two great problems. One is that pay doesn't advance much until you move into management and therefore out of the job in the field which if you are talented is the place you are most needed in, and the other is to do with how you measure performance. Just asking for performance related pay is the easy bit, working out how it works is the hard bit. It will be interesting to see what the institute thinks about that!


Bretwalda Edwin-Higham said...

Imposing the business paradigm to the public sector is stupid. It's like trying to make the railways pay. Why bother - just provide a cheap, effective service for the people, that's all.