In a slightly longer blog than usual, I set out the main findings from our recently completed ESRC project
Targets and performance indicators have become ubiquitous as techniques of governance. Governments and public service agencies have employed an array of such tools to steer, monitor and evaluate performance. Political leaders have also developed targets to signal their commitment to policy goals. Yet after more than three decades of performance measurement in public policy, most commentators agree that such tools have produced numerous adverse effects. Performance measurement techniques imply focusing on a limited range of quantitative features or goals, thereby narrowing down the focus of policy-making and political debate. They can create perverse incentives and encourage gaming. Not least, the use of such tools can erode trust within organizations, and even undermine confidence in political leaders and politics.
Given these short-comings, what explains the persistent appeal of performance measurement? And what effects do such techniques have on policy, politics, and public administration? The Politics of Monitoring was a three-year project sponsored by the Economic and Social Research Council (ESRC), which explored the factors shaping the use of targets and performance indicators in UK policy and politics; and examined their various effects on policy-making and public debate. The project focused on three policy areas: immigration, climate change, and defence procurement. Here I present some of the main findings of the project.
TARGETS AND INDICATORS AS AN ATTEMPT TO PRODUCE TRUST
Most analyses of targets and indicators suggest that they are favoured as a means of steering or controlling performance. Such techniques emerge as a response to the outsourcing of public services to semi-autonomous agencies and the private sector. Performance measurement is a means of enabling governments to govern ‘at a distance’. We suggest that this is only part of the story. Building on Michael Power’s work on auditing, we show how targets and indicators have an important symbolic role: they act as a ‘ritual of verification’ (Power) to those seeking to steer performance. They offer symbolic reassurance to the authors of targets and their audiences. Targets establish clear thresholds to guide decisions about bestowing and withholding trust. This helps assuage the anxiety of those attempting to steer policy-making and implementation. The use of performance measurement also helps bestow credibility on organizations and their political leaders, shoring up confidence in their capacity to control policy-making and implementation. In this way, targets and indicators help address problems of trust in the dealings between political leaders and those tasked with elaborating and implementing policy.
We also show how targets have an important outward-facing function. They are adopted to demonstrate to the public that the government is improving public services, and delivering on its pledges. Targets create an especially robust and precise tool of accountability. The appeal of this tool is particularly strong given the widely observed crisis of political disenchantment. The allure of targets thus lies not just in their promise of steering public administration, but also in their capacity to address problems of declining public trust in government.
Targets therefore have a dual function: they are adopted to address problems of trust between political leaders and their administrators; and to create public trust in the goals and conduct of political leaders. How far have targets succeeded in producing trust at each of these levels?
THE EFFECTS OF TARGETS ON PUBLIC ADMINISTRATION
The dual function of targets described above can create a number of organizational problems. In our study of asylum targets under the Labour government between 2000-2010, we found that technical targets designed to steer Home Office performance lacked political resonance, prompting politicians to resort to more politically appealing targets, notably Tony Blair’s prominent target of halving asylum applications. Yet these more politically compelling targets created various pressures and tensions in the Home Office. Targets aimed at mobilising public support were not well designed to address organizational problems (see Boswell 2015 for a full account, or this blog for a shorter version).
In our analysis of targets on defence procurement and asylum, we also found evidence of widescale ‘decoupling’: a gap between formal compliance with targets, and informal organisational practice, which deviated from the underlying rationale informing the targets. Literature on decoupling suggests that such forms of reinterpretation or subversion will be limited where organizations are subjected to clear, precise and robustly monitored targets. Yet in the 2000s, both the Ministry of Defence and the Home Office were successful in decoupling their informal practice from formal compliance with targets. Ultimately, it took highly intrusive political intervention by No.10 in the mid-2000s to bring about comprehensive reform of Home Office and UKBA practices. This achieved a closer alignment between formal and informal practices – although at a severe cost to organizational morale and trust between political leaders and civil servants.
In the Ministry of Defence, by contrast, No.10 was unable to impose the top-down reforms required to meet targets. This partly reflected the lower political salience of defence procurement. Asylum had become such a high profile public issue by the early 2000s that the Labour government felt its credibility was dependent on being seen to get a grip on asylum numbers. Defence was never similarly high profile. Moreover, No.10 and the Treasury felt they had less traction over the Ministry of Defence, given the technical complexity of procurement issues, as well as the distinct culture guiding the organization. You can read more about decoupling in the MoD and Home Office in Boswell and Fleming 2015.
It has long been observed that policies can get lost in implementation. We explored some of the problems of implementation by comparing how targets were appropriated and applied in the areas of climate change, defence procurement and immigration.
In Boswell and Rodrigues (2016), we developed John Kingdon’s idea of ‘multiple streams’ to try to understand differential implementation across sectors. We suggested that implementation is most likely to be successful where the policy emanating from central government converges with organizational constructions of problems, and where such policies are sustained by strong political commitment from the centre.
We developed a model of different implementation scenarios, based on combinations of these conditions. We applied the model to the case of targets in our three policy areas between 2000-2010. CO2 targets took the form of ‘bottom-up’ implementation, where policies matched organisational problem constructions, but there was limited political commitment from the centre. The Departments for the Environment, Food and Rural Affairs (Defra) and of Energy and Climate Change (DECC) were keen to enhance their leverage by foregrounding the targets. Implementation of defence procurement targets, as we saw, was characterised by widespread decoupling, with the MoD resisting an approach it felt did not address organizational problems, and No.10 lacking sufficient political commitment to impose compliance. Asylum targets initially saw similar decoupling, but intrusive intervention from No.10 led to coercive implementation, which eventually produced an alignment of organizational and political goals.
The analysis showed how organisations can shift between modes of implementation over time, responding to changes in organisational problems and central political commitment to the policy. You can read our article here, or have a look at our policy briefing for Discover Society.
TARGETS AND PUBLIC TRUST
If targets have generally failed to steer public administration in the way intended, how well have they succeeded in building public trust? Targets promise to provide an especially precise and rigorous mechanism of accountability. But our 65 interviews with political actors suggest that targets rarely bring political dividends to their authors. For a start, they often set overly ambitious goals, creating clear political risks. Even when targets are met, they rarely yield the expected media attention or rise in public satisfaction. Our respondents stressed how disappointed Labour politicians and officials were that the media was not interested in reporting on successful targets.
This is confirmed by our analysis of media coverage of asylum and net migration targets between 2002-2014. We found that press reporting is largely impervious to data demonstrating performance to target. Instead, trust in political leaders is grounded in more impressionistic cues such as the perceived integrity of politicians. Thus although Blair’s government was able to meet its target of halving asylum applications, press reporting was largely sceptical of the government’s motives and methods of meeting the target. By contrast, Cameron’s inability to meet his ambitious net migration target was widely put down to external and intractable obstacles beyond his control.
The analysis challenges mainstream accountability models, according to which public support is based on an appraisal of the performance or conduct of politicians. Instead, it suggests that the types of judgements underpinning political support and trust are based on more impressionistic beliefs about the values, integrity and authenticity of those asserting claims. (This research is forthcoming – please see our project website for updates.)
TARGETS AND POLITICAL DEBATE
Targets may have a limited or even negative impact on public trust. Yet they can have a number of other, often anticipated, effects on political debate. One of these concerns the effects of quantitative measures on the way issues are framed. The use of numbers to describe policy problems and goals can carry particular authority in political debate. Statistical measures are associated with impartiality, rigour and precision. They can be embraced as offering particularly robust tools for holding political leaders to account.
Such quantitative framings also carry strong appeal to journalists, offering compelling and authoritative data, and a means of exposing government transgressions. And even where targets and indicators are contested, they can still provide political opponents with an excellent tool for critiquing a government’s record. So opposition parties end up inadvertently buying in to targets, even if they opposed them. At the same time, quantitative targets or indicators provide very simplifying and often distorting descriptions of policy problems. In order to quantify policy issues, we need to classify the entities involved as discrete and equivalent units. Such abstractions can compress important nuances and variations. These simplified classifications can also prove to be very rigid: once statistical categories are created, it may be difficult to adjust or retract them. For these reasons, the authors of targets can become locked in to quantitative formulations of policy problems.
We discuss some of these questions in a co-authored chapter on how targets constrain policy-making. You can also read about the distorting effects of targets in two of our working papers. Yearley and Rodrigues (2016) explore how targets on C02 emissions were effectively ‘blackboxed’: they became taken for granted as ways of framing the problem, thereby removing them from scientific and political scrutiny. Boswell’s (2016) paper on the effects of the net migration target discusses how the target influenced political debate through a ‘classification’ and a ‘measurement’ effect. The argument is also summarised in this blog, and in an article for the Herald.
You can download a PDF pamphlet of the findings here. For more information on the project, please see our website: www.skape.ed.ac.uk/research/projects/politics_of_monitoring