Measuring public sector productivity has proved to be a challenging and daunting task. For one, the very definition of the term ‘productivity’ varies and depending on the definition you stick to, the variables are different too. Compounding this issue is the perceived complexity of measuring government outputs. In addition, available public sector productivity data is, more often than not, of questionable validity and reliability. So how is it done? The OECD simply defines productivity as to “how much output is produced for each unit of input, calculated as the ration of a volume measure of output to a volume measure of input used”. Simply put, good productivity means higher efficiency. You do more and better and less.

The Atkinson Review. The United Kingdom has been consistently enhancing its public sector productivity measurement since 1993. Beginning 1998, it already shifted from the traditional means of measuring public sector productivity (output = input) and has incorporated direct measures of the volume of government output in the national accounts. The Atkinson Review made a major step forward by recommending that to measure outputs, the total number of each of the activities performed by a given organization must be taken into account. Atkinson further recommended that these activities should then be weighed against each other according to the unit costs involved in producing them. In this step, the unit costs are used as proxies for the value of each of the different outputs produced, given that these are non-market outputs and thus do not have a price.

For national statistics purposes, where the level of analysis is often highly aggregated, Atkinson also recommended that output volumes should be adjusted by quality factors – a controversial and difficult to implement suggestion. The Atkinson Review included the following principles in measuring the output, input and productivity.


  • Output should be measured by incremental contribution to individual or collective welfare i.e. the added value by service concerned.
  • Start from services provided, and seek indicators that give full coverage.
  • Value should be seen as adjusted for quality.
  • Formal criteria should be established for extending direct output measures to further services.

Input and Productivity

  • Measures of inputs should be as comprehensive as possible, and should include capital services
  • Consideration should be given to the split between current and capital spending.
  • Criteria should be established for price deflators applied to input spending series.
  • Independent corroborative evidence should be sought on government productivity, as part of a process of “triangulation”.

The most important general conclusion of the Atkinson Review is that it is no longer possible and desirable to revert to an output = input perspective. This traditional perspective assumes zero productivity and in reality, this is unlikely to be true. Perhaps, the Atkinson Review’s most tangible role is giving information on the state of public services and the effects of reform and technological advancement in providing these services. In adopting the recommendations made by the Atkinson Review, the UK has made significant progress in measuring the output and productivity of their health, education, social care and social security services. While it is true that the Atkinson Review offers not a perfect method in measurement, it is a straightforward, consistent and mutually understood way of determining public sector productivity.