21st Century Target Setting

I wrote this article nearly six years ago - it still seems valid - particularly in the light current trends to move away from huge numbers of top down imposed targets. I hope the links still work.

Introduction

Many public service leaders and commentators are expressing increasing disquiet about the current approach to target setting for the public services.

David Batty writing in the Guardian (26/4/02) about the Victoria Climbié Enquiry entitled his piece ‘Performance targets compromise child protection’. Onora O’Neill speaking in the BBC Reith Lectures last year said ‘our revolution in accountability has not reduced attitudes of mistrust, but rather reinforced a culture of suspicion… we are galloping towards central planning by performance indicators reinforced by obsessions with blame and compensation’. A Telegraph headline of 19 December 2001 said ‘NHS patients were duped in waiting list fiddle’. Simon Caulkin in the Observer (5/8/01) declared that management ‘being tied to set goals is so often meaningless’ and can ‘lead to disaster’.

Dr Ian Bogle, retiring from five years as Chair of the BMA in June 2003 said 'I am absolutely appalled by the cheating going on and by the Government having put human beings in such a position that they feel that to preserve their jobs [they must do it]. The pressures are obscene and the Government should be ashamed of itself for the consequences.'[1]

And recently (and strikingly) James Strachan the new Chairman of the Audit Commission went on record to say to MPs: "The problem we have faced time and time again is the slavish devotion to targets, many of which have not been set very intelligently. It's a surefire way of not getting improvement in public services. People see targets set by government, monitored by them, and with responsibility for their validation. There is a real danger that people will not believe them," [2]

It seems therefore that people are becoming increasingly aware that the current performance management regime is not delivering the hoped for improvements in outcomes desired by the Government, the staff and managers who work in the public services and, indeed, the public themselves. This article is about another way – a way whereby the various levels of Government can still hold the public services to account for delivering results – as they most surely should – whilst tackling some of the difficulties with the current approach. As the voices for change become louder – it is vitally important that we do not ‘throw the baby out with the bathwater’ and reject target setting in all its forms. What we need is an evolution of thinking & practice – although that is not to say that some of the new ideas necessary should not be radical in their own right.

Targets and performance management are now so pervasive and influence at a very deep level how we conceptualise achievement in the public services – it is difficult to remember – what was the problem that performance management was meant to be solution for? This article will begin with an overview of the arguments against target setting and performance management in its current form. Further into the article, the way forward to a 21^st Century form of target setting is outlined.

It is hoped that this article will not only add to the clamour for a change in performance management practice but also offer some practical ideas for what is now required.

The key arguments 

To summarise briefly the key arguments against target setting and performance management in its current form:

• Targets are rarely set with anything other than a nod toward a statistical analysis of trends in performance of the system or service in question and there is little awareness of natural and special causes of variation in performance day-to-day, month-to-month, year on year. This can lead to reacting to changes that are merely part of an ordinary variation and sometimes not reacting when one is called for. Also a lack of dynamic statistical analyses can generate unrealistic targets.
  
• Outcomes are difficult to measure – by their very nature – so instead performance indicators (PI’s) are defined and targets set accordingly. These PI’s then become the objects for achievement rather than the outcomes themselves.
  
• Partly because targets are usually imposed (rather than developed in collaboration with the service providers and users) and partly by their very nature, targets can unfortunately foster ignoble or perverse attempts to hit the targets by whatever means are required (as past news stories around hospital waiting lists would appear to illustrate).
  
• Target setting and performance management only works when the system that is (or is not) delivering to target performance is understandable and understood such that interventions can be made that will change performance in the right direction. Without such understanding – requirements to improve performance can degenerate into ineffective exhortations to work harder, faster or better etc. 
  
• Target setting and performance management introduces a level of fear into work – a fear which can damage service delivery. Sometimes target setting is perceived as a form of bullying. When targets are not met – it is unfortunately too often the case that individuals are held culpable when in most cases it is the system itself, which has failed (in usually many places). Targets tend to foster ‘quick fixes’ rather than system (or ‘stay’) fixes.
  
• By focussing public servants and services upon targets and inspecting or appraising them against those targets – the targets become hugely important. When the targets do not reflect the local needs and wants of the public – target setting can distract people from giving the levels of service that they wish to give and the public want.
  
• Target setting and performance management can become a substitute for visionary and inspirational leadership. Management can too easily descend into ‘bean counting’ the PI’s and ‘double thinking’ leadership that pretends that all the targets really do matter – when in truth only a few key ones are of critical importance.
  
• Inspecting services against targets is too late, expensive and ineffective – it is in effect allowing services to deliver ineffective services rather than encouraging them to create systems for continuous improvement. Targets can foster ‘downstream’ measures of quality whereas it is much more effective to focus ‘upstream’ and assess where in the chain of actions things go right or wrong.
  
• Targets are by their very nature time focussed – often on a very short timescale. Most of the systems and issues wrestled with by the public services take years – if not decades – to come about. Improvement needs to work on that same timescale. In other words it’s a bit like having a speedometer that tells you what speed you were doing half an hour ago.
  
• The resources that go into setting targets and then responding to being held to account for the achievement of those targets are all resources that could be spent on delivering services and achieving outcomes. Balancing quite how many resources should go into setting and assessing against targets is a very hard task.
  
• Target setting and performance management often pitch one organisation against another, one individual against another when partnership and collaboration would be a much more effective strategy. Many local partnerships – endeavouring to make ‘joined up government’ work on the ground – find that the targets set by different parts of Whitehall make their task harder rather than easier.
  
• People – the public, the public service professionals, the politicians - know all these flaws above and as a consequence – because of the overarching importance attached to targets in their current form – become dissociated from their work and often lose their inherent ability to be passionate, creative and committed to their work and what they are trying to achieve. 
  

A way forward

This article is not just be about what is wrong –set out below are some practical proposals about what needs to happen to deliver sustainable social results, continuous improvement in those results, value for money and accountability. These proposals have been assembled into a 20 point plan for change:

1. There must a more widespread use of statistical process control[3] and general statistical methods to really understand how public service systems operate. (At the very least – matrices of performance data with more than nine cells should be banned in favour of performance data being presented graphically.)
   
2. The aim in monitoring performance should be to understand and seek to control variation in performance – as far as is possible in the complex systems that contribute to social outcomes. It is vital to avoid responding to variations which trend analysis shows are merely ‘ordinary’ variation and to ensure there is a response to ‘special’ causes of variation.
   
3. When considering what may be causing the variations in social results, practices need to be boosted that affirm the need to search truly creatively and rigorously – without (moral, historical, political etc) assumptions. Coupled to this is must be stressed that aiming for ‘what works’ is not a utilitarian recipe for focussing only on ends and forgetting the means to get there – often the public remember the means far more than the ends!
   
4. There needs to be far more education for the practitioners, managers, politicians and media in understanding variation and system improvement.
   
5. Government and management at all levels need to involve practitioners, users and other key stakeholders far more, in the search for PI’s that come closer to assessing progress towards desired for outcomes
   
6. We need processes / events / conferences / meetings / written communication / website bulletin boards to achieve much greater understanding and commitment to the outcomes being sought
   
7. Key leaders need to assert every opportunity to emphasise that PI’s are indicators – nothing more nothing less!
   
8. The balance between ‘top down’ and ‘bottom up’ target setting needs to shift more towards trusting and allowing the front line organisations – who are delivering the outcomes – to set their own targets. They must then be held to account for meeting their own targets and being committed to continuous improvement, contributing to outcomes and putting in place leadership and systems that deliver ever better results for their users and other stakeholders.
   
9. More research is needed to help each system to understand itself – informed by statistical understanding of variations in performance, the wit & experience of all those involved in providing the service and comparisons to other organisations. Under the title of evidence based practice – more is needed to find out just what are the actual mechanisms or interventions that actually deliver results. What are the differences in practice that really deliver different (and better) results?
   
10. When anyone – as a politician, manager or practitioner – intervenes in system, it is vital to understand what are the aims and evaluate the result against this. Knowing whether the result was achieved or not is also not enough – one has to understand why it was or was not achieved – reflective practice must be the order of the day. This is as much a proposal for government ministers as it is for doctors as it is for waste recycling managers as it is for crown court judges. Everyone needs to know (and have measures to validate self knowledge) – ‘is what I am doing working and indeed – am I doing it better than last year’?
    
11. It must never be forgotten that probably most of the systems that deliver social results are extraordinarily complex and it is rarely the case that one individual (the senior manager or director or middle manager) has all the knowledge required to make the best possible decision. The ‘manager’s right to manage’ should be interpreted as the manager’s right to ensure the system is well managed – by all those involved. In our increasingly complex world, participative management is not a ‘nice to do’ but a ‘need to do’. Whole system approaches need to be more widely used.
    
12. To repeat and emphasise one of Deming’s (see the acknowledgements below for a reference) fourteen points – drive out fear! In a workplace – fear can lead towards unforeseen and negative consequences.
    
13. Everyone knows that quick fixes do not work – just as it is widely known a rush DIY job will eventually (if not sooner than that) come round to haunt the DIYer. Leaders must do all that they can to encourage the use intelligence, resources and skills and so engender a culture of practice that is increasingly and robustly working towards stay fix approaches to improvement being the usual approach.
    
14. There needs to be more recognition that often positive social outcomes cannot be measured in only quantitative ways – a person, a community, a town or country might just feel or look different. At a subjective level – it might be recognised – but at an objective level it cannot be bottled or counted. Simply people saying – ‘its better round here now’ – might be the most profound result that could have been achieved.  In other words our indicators of performance and achievement must be a balance of the qualitative and quantitative.
    
15. We need far more hands on inspirational leadership from every echelon of the public service – leadership that co-creates a vision of the results we need and the commitment to achieve these – working together across the divides between national and local government, different agencies, the statutory and voluntary sectors and (most importantly) between the public themselves and the public service practitioners and managers.
    
16. Often far too many resources are put into making small improvements in existing service delivery systems where what is required a radical overhaul and replacement – where quality and a profound commitment is built in from the start – rather than bolted on at the end. We need more courage to stop investing in ‘tweaking’ and inspecting existing organisations and more investing in designing new and imaginative forms of service delivery – co-designed with the public, underpinned by electronic means of service delivery and run on the energy of public service commitment.
    
17. While the argument about whether targets work or not will rage on well past this article, it is now well established that short term targets do not work very well at all. The society that we have now is the product of generations and historical trends – there are no short term methods for the sustainable reduction in gun crime (for example). Obviously Governments are driven to improve matters – and are held to account for delivering those improvements – but the promises made need to be visionary and improvements in crime, education, health, the environment etc – all take time. The media in particular needs to understand this.
    
18. Whatever efforts we put into system and service improvement, we must measure the benefits of those efforts. The resources consumed by preparing for and responding to inspections need to be measured. The time it takes to measure progress against targets needs to be measured. The amount of resources used to analyse the system and deliver stay fix improvements also needs to be audited. But above all – public service organisations need to have far more robust systems for assessing the costs of not investing in improvement – as these costs are often overlooked and skew the appraisal of whether it is worth investing in a new form of service delivery or system improvement.
    
19. Problems ensue when different parts of Whitehall impose targets that do not add up on the ground. As many frontline practitioners know – delivering genuinely joined up services on the ground is seriously hampered by the apparent lack of coordination at the most senior levels of Government. Efforts are being made (the Social Exclusion Unit stands proud in this respect as mentioned above) – but efforts need to be redoubled again and again.
    
20. Finally there needs to be ways to move beyond ‘bean counting’ and beyond the presumption that targets can whip the public services into shape. All dimensions of the public services need to pursue 

• measurement that helps services to improve - demonstrably
  
• leadership that inspires such improvement
  
• an end to parochialism
  
• supporting the public services – with all the energy and commitment of those who work in them
  
• really, really listening to the public and acting on what their informed judgments are telling us about what they want improved and how to go about it
  

Conclusion

This piece has not been written as an academic exploration of the pros and cons of target setting and its associated performance management protocols, referenced to books and journals. It has been written in somewhat polemic terms to carry on and ignite more the debate about targets and their value to the public.

Old fashioned accountability around annual reports was never enough. Target setting of the kinds we are appear to be suffering from now are not working. We now need a third way – as outlined in the 20 point plan above. 

Jon Harvey

30 June 2003

Acknowledgements

Much of this analysis above is not new and can be traced to the work of W. Edwards Deming. Deming was one the key progenitors of the modern quality movement and remains hugely influential to this day. An internet search on his name will yield numerous websites dedicated to communicating his philosophy as set out in his is famous ’14 Points’ (see Out of the Crisis, 1986).

---

[1] (Full report http://politics.guardian.co.uk/publicservices/story/0,11032,987365,00.html) 29/6/03

[2] (Full report http://politics.guardian.co.uk/publicservices/story/0,11032,871947,00.html) 10/1/03

[3] See ‘SPC in the Office’ by Mal Owen & John Morgan. (June 2000) Greenfield Publishing; ISBN: 0952332841) for an accessible explanation of SPC and how to apply it to performance.