In this report, we go beyond the question of how much money is spent on public services and ask what the Government – and the public – is getting for that money. We do this by examining key datasets across nine different public services, to analyse how efficiently public money is turned into services that people use, and how this has changed over time.

What we measure

We define the efficiency of each public service in terms of the amount of ‘output’ the service produced for each pound spent. Outputs are essentially ‘the things that a service produces’ – for example, the number of pupils achieving certain exam grades or the number of people having a consultation with a GP. As these examples suggest, we are interested in both the amount of output a service provides and the quality of that output. This is sometimes referred to as ‘technical efficiency’.*

The efficiency of a service therefore depends on two things:

  • How much did the service inputs – such as staff, equipment and buildings – cost? If the Government or the people running services manage to drive down the costs of these inputs – by renegotiating maintenance contracts, for example, or limiting pay increases for staff – we say that they have made economies.
  • How many (and what sorts of) inputs were needed to produce each output? If the Government or the people running services manage to get more output (of the same or better quality) for each input – for example, by increasing the ratio of pupils to teachers – we say that they have made productivity improvements.

The efficiency of a service can, therefore, be boosted either by buying inputs more cheaply or by using inputs more productively to produce more with less.

This is a widely used framework for defining public service efficiency. The same basic approach has been used, for example, by the civil service’s Public Sector Efficiency Group.[1]

Understanding the performance of public services depends not only on understanding how efficiently a service converts money into outputs but also on whether the outputs produced are consistent with the aspirations that the Government has for the service and sufficient to meet the demands placed on them. Performance Tracker therefore also examines how demand for each public service has changed over time and whether new government policies have limited or increased these demands.

For each service, we look only at the parts of the UK where the Westminster Government is responsible. As most public services are devolved to the nations of the UK, this means that most of the chapters in this report only cover England. The exception is Chapter 4 on law and order, which includes the police, criminal courts and prisons, and which covers England and Wales.

 *    Technical efficiency – how money is translated into outputs, such as the number of hospital appointments – is distinct from ‘allocative efficiency’, which goes one step further – covering the process of converting money into outcomes for the public. Outcomes include, for example, levels of public health, crime rates or the rule of law. Performance Tracker focuses on outputs, rather than outcomes, because there are many factors (aside from the performance of one particular public service) that can affect outcomes. In the remainder of this report, the term ‘efficiency’ is used to mean technical efficiency.


How we do it

The analysis in this report is drawn from more than 150 data series – mostly produced by government itself – alongside other information gathered from inside and outside government. The exact information presented varies between different services, depending on what data is available and the nature of the service. But the analysis for each service follows the same basic structure.

For each service described in this report, we examine:

  • how day-to-day (‘current’) spending has changed since 2009/10* – throughout this report we describe spending in real terms, deflating cash figures using the Gross Domestic Product (GDP) deflator, which is a measure of economy-wide inflation**
  • how demand for the service has changed since 2009/10 – where possible, we have tried to identify how the underlying demand for a service among the population has changed, not just the demand that arrives at the door of the service
  • the type and quantity of inputs that each service uses and how this has changed over time – we focus in particular on the number of staff employed, as staff costs make up the single largest component of spending in most of the public services we consider
  • what volume and type of outputs have been produced by each service and – where possible – the quality of these.

Based on this information about spending, demand, inputs and outputs, we draw conclusions about whether each service has become more efficient over time – that is, producing more outputs with the same or less spending. Where possible, we also point out whether this has been made possible because of an economy drive (that is, reducing the cost of inputs) or through productivity improvements (that is, producing more output with every input, either by introducing new ways of working or by making existing inputs work harder).

We also ask whether these efficiency improvements have been sufficient to make up for any gap that has emerged between growth in spending and demand. If there are signs that some demand is going unmet, that people are waiting longer to access a service or that actual spending is consistently higher than budgeted spending, it indicates that the service is struggling to manage within its budgets. In these cases, initiatives to improve efficiency, divert demand away from the service or increase spending are likely to be needed.

*    For some services, consistent spending figures are not available as far back as 2009/10. In these cases, our analysis covers a shorter time period.

**    For some services, the price of the inputs they use may have risen markedly more or less quickly than economy-wide inflation. Where this is relevant and alternative measures of service-specific inflation are available, we mention these in the text. However, for consistency and ease of comparison between services, we present all the headline spending figures deflated by economy-wide inflation.