Civil Service in the news
Understanding public sector productivity
Research underway at the London School of Economics (LSE) explores changes in public sector productivity over the last twenty years.
The preliminary results of the Public Policy Group study, released on 8 July at the LSE conference Understanding Public Sector Productivity Change, suggest that productivity in the area of customs processing has improved since 1988, but hasn't in the areas of tax collection and benefits payment, despite large-scale service investment and restructuring during this time.
However, discussions at the conference of civil servants, academics and Office of National Statistics (ONS) researchers also showed just how difficult it is to accurately measure public sector productivity. Even so, the LSE researchers Professor Patrick Dunleavy, Dr Leandro Carrera and Simon Bastow say that as 25% of final UK consumption is spent by government, it is vital to overcome these difficulties and determine how well public money is being used.
There have long been measures of private sector productivity but, Professor Dunleavy said, until the late 1990s ‘productivity in the public sector was assumed to be flat’ because public sector outputs were given the same value as inputs – a simple way of coping with the fact that public sector outputs do not have a price in the way that private sector outputs do. Concerned with this approach's limitations, the Government asked the ONS to explore more sophisticated ways to measure public sector productivity.
In 2003, the Government commissioned the Atkinson Review, led by Sir Tony Atkinson of Oxford University, to explore how successful the ONS had been in their work. The final report in 2005 accepted the methodology of the ONS overall, which consists of measuring the specific number of outputs within a public service and weighting them by the unit costs of producing them to elaborate a cost-weighted index of outputs. The Atkinson Review also suggested applying this approach to a wide range of public sector services in health, education, social protection, and public order and safety.
The United Kingdom Centre for the Measurement of Government Activity (UKCeMGA) was created by the ONS in July 2005 to do this, and since then has been working to further improve measures of government output and monitor the performance of the main government services.
UKCeMGA measures public sector productivity through segmenting public service inputs and outputs and analysing and assigning values to each segment, before calculating the ratio of outputs to inputs to give the productivity. Director Aileen Simkins says it is important to look at the whole picture and account for both quality and quantity when analysing productivity, as services ‘can make one area better by making another worse.’ It is also difficult to measure the impact of investment, as there can be some very long lags before you see improvements – 10 years in education and 40 years in health, for example.
The preliminary research of the LSE Public Policy Group study similarly applied the methodology sketched out in the Atkinson Review to measure the productivity changes over the two decades from 1988 to 2007 in the processing of goods for customs, tax collection and the payment of social security benefits.
As the researchers say,
‘total government expenditure in these three public service outputs is above 13% of GDP, so how efficiently they are carried out has major implications for national prosperity.’
Professor Dunleavy said,
‘We're focusing on areas that are perennial services, where there's unlikely to be a great change in efficiency or quality…and looking at productivity in terms of inputs to outputs, not at all in terms of effectiveness.’
Only customs processing showed significant increases in productivity. How this could be the case, when in the private sector the expectation is always that productivity keeps increasing, was discussed at length in the conference.
According to Professor Dunleavy, ‘one of the things that might be fairly continually damaging to productivity is government intervention‘. He questioned whether our short-termist political system reduces productivity because frequent legislative change means that services have to regularly reorganise themselves and retrain staff. Of the three areas studied, there has been significantly less legislation in customs, which came out best in terms of productivity.
However, some people challenged the customs results at the conference: could customs seem to be increasingly successful at making seizures (as a percentage of their checks) simply because an increasing percentage of people coming through customs are carrying contraband?
The idea of effectiveness, and of how you assign value to public service outputs, is also perhaps why measuring productivity is more difficult and contentious than measuring private sector productivity. How do you measure social good? If crime has been falling but fear of crime increasing, should the police focus on making people safer or on them fearing crime less? This is why the LSE research seeks to look firmly at concrete outputs such as the number of customs seizures made or benefits claims processed rather than delve into the emotive political issue of effectiveness.
Deciding on how good public services are when calculating productivity is also difficult because, Dunleavy says, societal expectations of government are changing too – people expect a more personalised service, for example. ‘Although it looks like there's been a massive change from within the Civil Service, it's not so easy to capture this from outside it.‘
Then there’s the question of whether technology changes – for example the move from paper forms to online forms – are simply basic modernisation or increase the quality of public services. The conference made it clear that all of these issues need to be considered when analysing public service productivity and finding out how well public money is being spent.
The Public Policy Group hope to publish their next set of findings in the autumn.
Further information:
- Innovation Research Programme at LSE [External website]
- United Kingdom Centre for the Measurement of Government Activity (UKCeMGA) [External website]
