Performance management in public sector organisations

Public sector organisations
Public sector organisations come in many shapes and forms. The most obvious examples are schools and hospitals, police forces and local transport providers, but there are many less visible organisations such as regulatory bodies. The objectives of public sector organisations are very different from those of commercial organisations, and this can make performance management more complicated. The following factors in particular differentiate public sector organisations from commercial:

1. They have a broader group of stakeholders than commercial organisations. This can lead to greater conflicts. Commercial organisations are likely to be mainly concerned with shareholders, employees, customers and their lenders. Public sector organisations are likely to be interested in pleasing the providers of funding (the government), the users of the service and the taxpayer. In the case of schools, for example, parents would be happy to see more money spent on education – but, as taxpayers, they may not wish to pay more taxes.
2. Customers do not pay directly for the services they receive, and there may be little relationship between the costs of providing the service and the amount it is used. Consider a subsidised bus service, for example. The daily costs of running the buses are likely to be largely fixed, and do not depend on the number of passengers using them – at least in the short term. This makes it harder to decide how much should be spent on the service.
3. Many public sector organisations operate as monopoly providers. Even if customers are not happy with the service they receive, they cannot switch to an alternative supplier. In commercial organisations, this is generally not the case, and bad performance will lead to a loss of customers and, therefore, loss of funding.
4. The output of public sector bodies is often difficult to measure. How do you determine how much work a police force has performed? Statistics such as the number of crimes reported may be used. If the police force is doing a good job however, and crime is falling, the number of crimes reported may fall. So the lower number of crimes reported would wrongly suggest that the police force is not working so hard.

There is a perception that performance in public sector organisations is poorer than in the private sector, both in terms of efficiency and quality of service.

Greater use of targets in public sector organisations
Since the early 1980s the governments of some countries – notably the UK, Sweden, Australia and New Zealand – have undertaken reforms aimed at making public sector organisations become more accountable. These reforms have been driven by the need to:

• improve the overall efficiency and effectiveness of public expenditure
• reduce overall levels of expenditure
• improve accountability and transparency of the public sector
• enhance the responsiveness of public sector organisations to the needs of citizens.

Such reforms have been dubbed the ‘new public management’.

One of the key features of these reform programmes has been the increasing use of performance measures and targets to evaluate all aspects of the performance of an organisation’s activities. Typically, objectives are identified based on the mission statement. Targets are set for each objective. Managers must explain any variances between actual performance and the targets.

The targets are normally based on some output of the organisation. Typical targets might be the number of patients treated by a doctor, or the number of passports issued by the passport issuing authority. Qualitative targets may also be used, such as the level of patient care, which could be measured using surveys of patients.

This highly rational approach to performance management has its roots in the writings of Drucker and Argenti, who emphasised detailed planning and the use of quantified targets. They argued that this gives greater direction to the management and staff of an organisation.

The use of and publication of targets also increases accountability. In the public sector, the managers are the agents, who act on behalf of the principal. The principal would be the general public, although the role of principal is often played by the government on their behalf. Much discussion of accountability focuses on whether or not managers have acted ethically – ie have not stolen the funds provided to them, and can account for their use. An equally important aspect of accountability, however, is how well the agent has performed in terms of efficiency.

Linking reward schemes to targets
Performance-related pay schemes are usually introduced alongside the targets, whereby bonuses are paid if particular targets are achieved. In the UK , for example, the government introduced an incentive programme for doctors in 2004, whereby bonuses are awarded based on the achievement of 146 targets. These targets focus on a wide range of areas, including clinical care, practice organisation and patient satisfaction. The bonuses are on a sliding scale – so even if not all of the targets are achieved, some of the available bonus would be paid.

Having such reward schemes is designed to improve the motivation of the management and staff, and it is argued that this improves the overall performance of t he organisation.

Difficulties of using targets in the public sector
The use of targets in the public sector is not without problems, however:

1. The larger number of stakeholders makes it difficult to decide which metrics should be used. Often, governments focus on reducing costs. This has led to situations such as hospitals refusing to buy life saving medicines because they are too expensive, or police being removed from the beat to save money.
2. There may be less of a direct link between effort and outcomes in the public sector. In a hospital, for example, mortality rates may depend on many factors that are outside of the control of the hospital. Individual targets may not be a fair measure of performance in such situations; however, using a range of targets may overcome these problems.
3. It may be difficult to identify quantifiable outputs in the public sector. How does one measure the output of the local fire brigade, for example?
4. If systems are implemented in a very rigid way, without giving consideration to local issues, or special situations relating to the organisations being measured, then this may lead to problems such as manipulation of data, tunnel vision, sub optimisation and so on.
5. Many critics of targets in the public sector argue that their use has not resulted in lower costs or better quality of service. They claim that what has actually happened is that a higher portion of the organisation’s budget is spent on employing managers and accountants to set the targets and measure performance, and a lower portion has been spent on frontline services.

This final criticism does not appear to have been confirmed by the facts. In the UK National Health Service (NHS), for example, the period from 1999 to 2009 saw a large rise in the use of targets. During this period, the average increase in support staff was 3.6% per annum, while the average increase in total staff was 3% per annum, according to statistics released by the NHS information centre (www.ic.nhs.uk). There was clearly some increase in the portion of budgets spent on managers, but hardly a significant increase.

Empirical evidence
It is extremely difficult to assess the impact of the use of targets in public sector organisations, due to the fact that it is difficult to assess what would have occurred had they not been introduced. Unlike scientific experiments, where there is a control experiment, no such control experiment exists in the public sector. In some situations, governments introduce pilot schemes in an attempt to gain some sort of comparison, but it is often difficult to get precise comparisons.

Another problem is that numerous changes have been made to the practices of setting and monitoring targets over the years, making the picture even less clear. Debate has largely focused on anecdotal evidence rather than on proper studies of the outcomes.

Studies into the impact of performance-related reward schemes are fairly few and far between, but do appear to show a positive relationship between the use of performance-related pay and the performance of the staff of the organisation (5). While such studies have shown that individuals work harder, the impact on the provision of services overall is far from certain.

Use of benchmarking in public sector bodies
Benchmarking has also been used in many countries as a means of reducing the perceived gap between the performance of public sector organisations and their private sector counterparts, with the aim of improving the quality of service, and ultimately saving the taxpayer’s money.

Benchmarking is where one organisation compares its performance in a specific area with another organisation, the benchmark, to identify how much room there is for improvement. It then attempts to implement practices similar to the benchmark in an attempt to narrow the gap in performance.

The specific area for benchmarking could be a particular business process – such as inventory control, or it could be a broader area such as ‘quality of customer service’. The benchmark could be another organisation, or it could be another department within the same organisation.

Seven-step approach to benchmarking
The consulting firm Kaiser Associates proposes a seven-step approach to benchmarking as follows:

1. Determine which areas or functions to benchmark. It would probably not be feasible to benchmark all functions at one time, so it is necessary to choose those activities where benchmarking can bring the greatest benefits to the organisation. This may be based on which activities offer the greatest scope for cost savings, or which are ‘key service differentiators’.
2. Identify the performance indicators and performance drivers that will be measured during the benchmarking exercise.
3. Select the organisations that will be used as the benchmark.
4. Measure the performance of the benchmark using the measures identified in step two above.
5. Measure your own performance, and compare it to the benchmark to identify the gaps.
6. Specify actions and programmes to close the gap. This involves analysing how the benchmark achieves superior performance, and identifying similar practices that could be adopted.
7. Implement and monitor the actions and programmes. Monitoring should not be a one-off process, but should continue for a longer period after the benchmarking exercise.
   
   

Identifying the performance indicators
In public sector benchmarking, the performance indicators used tend to focus on cost and efficiency or differentiation. Cost variables might include items such as labour efficiency, or total costs for a particular function as a percentage of income.

Regarding differentiation, many of the metrics used would be of a qualitative nature, such as client satisfaction or quality of service. It is difficult to measure these directly due to their subjectivity. One approach is to use customer surveys for these.

In attempting to find relevant metrics, benchmarking exercise s carried out in the past by similar organisations can be a useful source. Much information about these is available in accounting and business journals, and online, or by contacting organisations that have already performed a benchmarking exercise. This is easier in the public sector, as the government – as overseer and beneficiary of the benchmarking – can often force other organisations to disclose information. This would not be the case in the private sector.

Selecting the benchmark
When choosing the benchmark, we can talk about different types of benchmarking:

• Internal benchmarking uses another organisation within the same organisation. For example, a comparison of the performance of the procurement department of one hospital with the procurement department of another hospital.
• External functional benchmarking is where a particular function is compared with that function for the organisation that performs that function best, regardless of which industry they are in.
• Competitive benchmarking is where a competitor is used as the benchmark. This may not be so common in the public sector.

When using external functional benchmarking, using a similar organisation in terms of objectives and size can make the process easier. It is not necessary then to take into account differences between the two organisations when comparing their performance, and it should be easier to adopt the practices of the benchmark if they are similar. For example, if a school uses a similar school in a different area as a comparator.

External functional benchmarking can also be performed successfully using out of category organisations – ie organisations that may be have totally different objectives and even different primary activities. Using such benchmarks will make the process more complex, but may provide the opportunity for an organisation to overtake, rather than simply to catch up with comparable organisations. For example, a public sector logistics department could use one of the private sector international courier companies as a benchmark for its logistics.

The most important factor when selecting the benchmark is to identify the ‘best in class’ for the activity or business process being benchmarked.

Measuring the performance of the benchmark
Prior to starting to perform measurement, the organisation will have identified what it wants to measure (in step two), so it should already be clear what information is required. The question now is how to obtain it.

Much information is already in the public domain. Many organisations publish information about best practice benchmarks for particular industries. Financial reports may provide information about cost efficiency. There may be newspaper reports or analysts’ reports available for larger listed companies. Publicly available information is a good place to start, but it is unlikely to provide all the information required for a successful benchmarking exercise.

Another source of information is ‘data sharing’ where other organisations are contacted, either directly and formally, or through professional conferences. This can be supplemented by interviewing of staff at the benchmark.

Factors that influence the effectiveness of benchmarking
Based on some empirical research, Sandra Tillema tried to identify what factors determine whether or not a benchmarking exercise actually leads to improved performance. A study of benchmarking carried by four Dutch water boards concluded that the performance of those water boards had not improved after the benchmarking.

Tillema claims that one reason for the lack of success of many public sector benchmarking exercises is that they focus only on measuring performance against the benchmark. They do not attempt to learn from, and adopt the practices of, the benchmark. Thus, the benchmarking is often a measuring exercise, not a learning exercise.

A second factor is the influence of stakeholders. A benchmarking exercise will only lead to improvement s if pressure is put on the organisation from its stakeholders to narrow the gap identified between the organisation and the benchmark. This pressure can come from internal stakeholders, such as supervisory boards, or external stakeholders, such as users of the service. In the case of the Dutch water boards, no pressure had been exerted on the managements of the board, which is why no improvement was experienced.

In some situations, benchmarking may lead to economic pressure on an organisation to improve. If poor benchmarking results are published, users of the service may switch to alternative providers. So benchmarking can lead to economic pressure on poorer organisations. In the case of state universities, for example, students have a choice which university to go to, and their choice may be influenced by published results of benchmarking.

Such economic pressure only works in situations where users can switch service provider. This is not always the case, and monopoly providers of services will not feel this economic pressure. Another issue is that users may not always understand the results of benchmarking, and their decisions may be based on factors other than the benchmarking exercise – for example, the student that chooses a university based on the better nightlife in the city where it is located, rather than the publicly available results of benchmarking.

Dysfunctional effects of benchmarking
Tillema notes that in common with all performance measurement, benchmarking may lead to dysfunctional behaviour. Management may take actions to improve their measured scores without improving underlying performance. A criticism by some of UK schools is that management focus on improving their performance in the government league tables, not on providing a good education for pupils. This is part of a general problem in performance management, which is the ‘what gets measured gets done’ concept.

Benchmarking can also be used to defend rather than improve poor performance. In such cases, management focus on explaining why their organisation performed poorly, citing factors that make their organisation different from its peers. In such cases, the benchmarking leads to little or no improvement in performance.

One factor that leads to greater levels of dysfunctional behaviour is where stakeholders misinterpret the results. For example, they do not take account of different operating environments, or different objectives of the organisation and benchmark, and this leads to unrealistic pressure to close the performance gaps.

Misinterpretation is less common where the results are only made available to expert stakeholders. Because the expert stakeholders have a better understanding of the performance, and the factors that may differentiate the organisation from its peer s, they will be more realistic in their assessment of the results. This reduces the incentive for managers to manipulate the results.

League tables
League tables are commonly used in the public sector to present the results of benchmarking. Under league tables, the various metrics are summarised into a weighted average overall score. A league table is then prepared, showing all organisations ranked according to their overall score – for example, UK schools are ranked by exam results.

The benefit of league tables is that many different areas of performance are summarised into one final score, showing how well the organisation has performed over all. League tables are also designed to improve competition among the organisations, giving an incentive to the poorer performers to improve so that they move up the table.

Common criticisms of league tables are that they apply arbitrary weighting to the various factors that are used in the calculation of the final score. It is also argued that they do not take into account differences between the organisations being measured. In schools, for example, one factor that affects the performance of schools quite strongly is the demographics of the area where the schools are located, with schools in poorer areas typically appearing towards the bottom of the table.

Conclusion
There has been a move towards making public sector bodies more efficient and effective, using more targets and benchmarking. It is not clear whether targets have improved the performance of public sector bodies or not. Benchmarking can lead to improved performance in some public organisations. However, it is most successful where stakeholders have the ability to apply pressure to the organisations to narrow the gap between their actual performance and that of the benchmark. If not, then the benchmarking may not lead to improved performance.