Challenging metric power

Modern public service is a world of targets, indicators, financial and economic analyses, and statistical data and projections. The choice of the numbers determines what we measure and what is considered important. The process of providing numbers dominates policy thinking and advice. The things we measure exercise a power over the agenda, controlling how advice is given and flowing across media and political debate. You can think of the reporting of hospital waiting list in the UK, or, worldwide, of the fixation with Gross National Product or the financial exchanges. When we see these issues dominating our lives, it may be good to remember that it was not that long ago that maintaining the gold standard was the big thing…

As well as the broad political hold numbers have, they also shape more subtly the life of bureaucracy and in turn those who engage with bureaucracy. Advice typically has to make a financial and cost benefit case for action – such business cases are dominated by numbers rather than vision and values and finding the numbers can both distort the goals and what is measured in the activity. This distortion of purpose can then be reinforced in audit regimes and political or public scrutiny.

The spread of managerialism in bureaucracy means that targets and KPIs are now applied universally to direct service provision and track delivery. Setting numerical targets for delivery systems or contracts always lends itself to unintended consequences or gaming rather than keeping the system focused on what the policy intention really is. It divests the measured activity of meaning and value and replaces trust in common purpose in public service with a blunt stick.

Agencies often have their own technical approaches which are similarly numbers driven. They are also sources of projections and models where the numbers which are chosen for input shape future policy direction. I think especially of the projections of motorised transport growth in the 20th century which led not to a questioning of the direction but to massive expansion in road building to provide for it – a self-fulfilling prophecy which facilitated ever greater growth and transformed the places in which we live. We seem to have the same fascination now with developing AI and megadata, though at least with some voices questioning the moral implications.

Even in the area of sustainability, the climate change debate became an almost abstract argument about numerical targets rather than taking urgent measures, where long-term goals served to replace activity. This was coupled with arguments about economic impacts and optimal timing of interventions, often forgetting that emissions are long-lasting, and the reduction curve needs to be frontloaded.

We need to have many more ways to measure what is good and possible beyond the target, the calculated economic number or the model projection. None of these are able to cope with the genuine uncertainty of our complex interacting systems and none of them speak to the concerns and values of people in their communities. The numbers serve as technical justification and reinforce a fixation with the economic. Simply starting to seek out and apply local values and knowledge in decision-making (without converting these into ersatz numbers) is a good starting point for change.