It is still a bit surprising that this blog is so critical of the record of the socialist governments of Cameron, Osborne, Clegg and successors from 2010 to 2024. But while they hacked away at the machinery of public administration quite mindlessly they did not seem to do substantially worse on civil service efficiency than other previous governments. It is not a simple problem because "efficiency" in government administration is a difficult concept as it has rather different goals from business.
I came to the same conclusions and noted how there is plenty of evidence that performance related pay works, if there is a clear link from performance to pay and the incentives are large enough - something the civil service has never done. Also worth noting that this completely useless performance management mish-mash is presided over by the Cabinet Office which, from publicly available data, has around a thousand staff working on HR matters, spending c£60m a year.
My conclusions were: "As I was writing this blog, there was another news story about planned government cuts to the civil service, This time a 15% reduction in running costs by 2030. I hope that part of the reduction will come through a thorough review of what the government no longer needs to do or can do with fewer people. But it would be nice to think that the majority of any cuts would fall on the weakest performers. A well implemented system of performance pay would also be welcome, as it should motivate the best staff and help attract the good quality, risk-taking people that ministers say they want in the civil service. But given the civil service’s chronic inability to manage the performance of its staff, these particular outcome sound highly unlikely."
«I came to the same conclusions and noted how there is plenty of evidence that performance related pay works, if there is a clear link from performance to pay and the incentives are large enough - something the civil service has never done.»
The difficult is not just how to make "a clear link from performance to pay" but how to measure "performance" in the civil service where profit is not the measure of success but faithful and fair execution of the law.
According to Cory Doctorow, the reason that the Civil Service performs badly is not primarily because of 'inefficient' actors but because there is lack of viewpoint diversity. A good civil service has members with ideas and beliefs that disagree and contradict each other and reflect the diverse opinions of the public they serve. When it comes to making regulations, they hash out all of problems and the objections and produce something which reflects a deep understanding of the problem matter. A bad civil service is a monoculture. Sometimes this is because the civil service has been captured by members of the industry they propose to regulate. Sometimes this is because people of one persuasion say 'the science is settled!' and then refuse to hire people who disagree with them. The result is a group that tries to manage and rule the people who disagree with them, rather than ever admit that they might have a point now and then,
The Home Office would be improved by an annual ‘Hunger Games’ style event, designed to ‘redeploy’ 25% of the workforce. We could barricade off the whole of Marsham St / Horseferry Road and let them go for it. By selling the live TV rights HMG would also make a few quid, too.
The Home Office would be improved by an annual ‘Hunger Games’ style event, designed to ‘redeploy’ 25% of the workforce. We could barricade off the whole of Marsham St / Horseferry Road and let them go for it. By selling the live TV rights HMG would also make a few quid, too.
I have never seen performance management done well in any large organisation, I've been in 4 corporates, 3 schools and a consultant. Ratings are usually given following some pseudo-objective process involving collecting evidence which wastes everybody's time. In some settings, e.g. a sales environment, financial bonuses can work, but in most settings effective collaboration is more important than individual brilliance, so performance-related pay can be decisive and is usually counterproductive. It's a lazy, unsophisticated way to motivate people.
«I have never seen performance management done well in any large organisation, I've been in 4 corporates, 3 schools and a consultant.»
The difficulty is defining "performance" as there are "cost centre" and "profit centre" roles and "performance" means something very different.
For "profit centre" ones it is "profit" which is not entirely easy to define itself but at least in principle it has some objectivity.
For "cost centre" roles one could fantasize that "performance" is a ratio between cost of inputs and quality and quantity of outputs but assessing quantity and especially quality of outputs is very difficult indeed in many "cost centre" roles. Therefore the popularity of bullshit "KPIs". Nearly all roles in the civil service are "cost centre" (for the civil service, not necessarily for the state or its population).
One approach would be to accept that "cost centre" role performance is an entirely subjective evaluation of quantity and quality of output and make it depend on the vote of a 3-way commission (manager, manager's manager, someone from a completely different branch of the organization).
«so performance-related pay can be decisive and is usually counterproductive. It's a lazy, unsophisticated way to motivate people.»
As a rule the real purpose of performance evaluations is not to influence performance but to remind employees of the power of managers.
«On looking at the sickness data in this article, my first instinct was, I wonder how that compares with the private sector?»
Trivial comparisons between private and public sectors workers are meaningless because the workforces have very different profiles; in particular after many rounds of outsourcing of lower-paid roles public sector roles tend to be more white-collar and middle-class than private sector ones.
Also in the private sector there are many employers with "3 absences and you are out" policies and consequently many employees with "work through the pain" attitudes. Perhaps many taxpayers (especially retired ones I guess) would love for those policies and attitudes to be more common in the civil service too.
That's a fair point, although the CIPD survey is not as representative as ONS data, which shows the public sector is still significantly worse than the private sector on days lost through absences - see link below. Also, as Neil mentioned, the disparities between departments are impossible to explain and suggest that there is plenty of scope for improvement among many of them.
What you describe sounds like any large organisation anywhere - public or private. The fetish of efficiency you demonstrate - as if there is a perfect organisation with a perfect structure and perfect staff that just needs to be carefully sculpted out of the imperfect stuff of humanity (by you, of course) is fantasy. This is humanity. It is imperfect, not always focused, not necessarily expert, often mistaken, sometimes confused, rarely completely committed. People have lives and flaws and strengths and weaknesses, and stuff happens - all the time. I can absolutely guarantee that there is some dude in Singapore making pretty much the same complaints right now.
People who go on about this are people with very little self awareness, because they always imagine they are a manifestation of the sort of excellence they crave. But...they aren't. They are just someone with some anecdotes, a lot of opinions, and a hustle. What most organisations don't need is someone coming along every five minutes reorganising them according to their latest pet theory. Organisations work because of the people in them. They also fail because of that. But, it is not either or. It's a continual mixture of both. The trick is to try to make sure the first outweighs the second. That is it.
The original post did not contain any anecdotes, but did have a lot of data, all of which suggest some serious problems - useless performance management systems, unsuitable people getting promoted, almost impossible to fire people, very probably bogus claims under the Equalities Act. Maybe you can address those things, which I'd guess most readers would find dispiriting and in need of addressing, not by reorganisations, but by effectively managing people and reducing civil service numbers.
Data… Ah yes, the carefully selected bits that support the conclusion. The point being, that all these things will happen somewhere at some time in any organisation. Even in Singapore. Taking incidents and turning them into a pervasisve pattern, then ascribing that pattern to your pet theory - there is a whole ocean of data to fish in with that goal. Provided you dangle long enough, you’ll catch something. Most organisations are in need of improvement. The real question is, how to improve things without screwing something else up, and if these management consultant types are so clever, how come they are still in business? Shouldn’t they have perfected the design by now?
This is a grift pitch. Look, your organisation is messed up - I can fix it. repeat ad nauseam.
I don't know who you think is grifting, but I don't think your comment is helping anyone. If you don't like data at all, how would you try and run any organisation? If you think Neil's data is flawed in some way, please tell us how.
I respect data. I used to be an information analyst. But this is not data. It is selectively presented statistics. I am acutely aware how statistics can be tweaked to support almost any case. This whole article is a text book example. This is a grift to push his failed and deeply unpopular political party. It resonates with all those people politically motivated to want it to be true. It’s a transparent attempt to project the blame for the calamitous government that he was part of onto others, which wouldn’t fool a child.
I am gratified to see that the US doesn’t suffer from this phenomenon alone. The nature of the US government system “can” provide a certain level of “flexibility “ however it is locality and level dependent. At the federal and state level the vast percentage of employment is considered a social program. Competency or employment preparedness is not generally considered just demographic factors. As there is to some degree a need to get things done, a small percentage of the opportunities are competent. The primary connection between performance and management happen at the locality level, and that is situation dependent. In a practical sense, performance management is a social process not a competency one.
From a legal rather than procedural perspective, is the civil service subject to different employment law than the private sector? Ie) Does it require ""just"" cultural, or in fact also legal changes to change outcomes?
It is still a bit surprising that this blog is so critical of the record of the socialist governments of Cameron, Osborne, Clegg and successors from 2010 to 2024. But while they hacked away at the machinery of public administration quite mindlessly they did not seem to do substantially worse on civil service efficiency than other previous governments. It is not a simple problem because "efficiency" in government administration is a difficult concept as it has rather different goals from business.
A very good post. I posted on this topic last year, focussing on performance-related pay - https://freeblogger.substack.com/p/managing-staff-performance-in-the
I came to the same conclusions and noted how there is plenty of evidence that performance related pay works, if there is a clear link from performance to pay and the incentives are large enough - something the civil service has never done. Also worth noting that this completely useless performance management mish-mash is presided over by the Cabinet Office which, from publicly available data, has around a thousand staff working on HR matters, spending c£60m a year.
My conclusions were: "As I was writing this blog, there was another news story about planned government cuts to the civil service, This time a 15% reduction in running costs by 2030. I hope that part of the reduction will come through a thorough review of what the government no longer needs to do or can do with fewer people. But it would be nice to think that the majority of any cuts would fall on the weakest performers. A well implemented system of performance pay would also be welcome, as it should motivate the best staff and help attract the good quality, risk-taking people that ministers say they want in the civil service. But given the civil service’s chronic inability to manage the performance of its staff, these particular outcome sound highly unlikely."
«I came to the same conclusions and noted how there is plenty of evidence that performance related pay works, if there is a clear link from performance to pay and the incentives are large enough - something the civil service has never done.»
The difficult is not just how to make "a clear link from performance to pay" but how to measure "performance" in the civil service where profit is not the measure of success but faithful and fair execution of the law.
According to Cory Doctorow, the reason that the Civil Service performs badly is not primarily because of 'inefficient' actors but because there is lack of viewpoint diversity. A good civil service has members with ideas and beliefs that disagree and contradict each other and reflect the diverse opinions of the public they serve. When it comes to making regulations, they hash out all of problems and the objections and produce something which reflects a deep understanding of the problem matter. A bad civil service is a monoculture. Sometimes this is because the civil service has been captured by members of the industry they propose to regulate. Sometimes this is because people of one persuasion say 'the science is settled!' and then refuse to hire people who disagree with them. The result is a group that tries to manage and rule the people who disagree with them, rather than ever admit that they might have a point now and then,
The Home Office would be improved by an annual ‘Hunger Games’ style event, designed to ‘redeploy’ 25% of the workforce. We could barricade off the whole of Marsham St / Horseferry Road and let them go for it. By selling the live TV rights HMG would also make a few quid, too.
The Home Office would be improved by an annual ‘Hunger Games’ style event, designed to ‘redeploy’ 25% of the workforce. We could barricade off the whole of Marsham St / Horseferry Road and let them go for it. By selling the live TV rights HMG would also make a few quid, too.
I have never seen performance management done well in any large organisation, I've been in 4 corporates, 3 schools and a consultant. Ratings are usually given following some pseudo-objective process involving collecting evidence which wastes everybody's time. In some settings, e.g. a sales environment, financial bonuses can work, but in most settings effective collaboration is more important than individual brilliance, so performance-related pay can be decisive and is usually counterproductive. It's a lazy, unsophisticated way to motivate people.
«I have never seen performance management done well in any large organisation, I've been in 4 corporates, 3 schools and a consultant.»
The difficulty is defining "performance" as there are "cost centre" and "profit centre" roles and "performance" means something very different.
For "profit centre" ones it is "profit" which is not entirely easy to define itself but at least in principle it has some objectivity.
For "cost centre" roles one could fantasize that "performance" is a ratio between cost of inputs and quality and quantity of outputs but assessing quantity and especially quality of outputs is very difficult indeed in many "cost centre" roles. Therefore the popularity of bullshit "KPIs". Nearly all roles in the civil service are "cost centre" (for the civil service, not necessarily for the state or its population).
One approach would be to accept that "cost centre" role performance is an entirely subjective evaluation of quantity and quality of output and make it depend on the vote of a 3-way commission (manager, manager's manager, someone from a completely different branch of the organization).
«so performance-related pay can be decisive and is usually counterproductive. It's a lazy, unsophisticated way to motivate people.»
As a rule the real purpose of performance evaluations is not to influence performance but to remind employees of the power of managers.
On looking at the sickness data in this article, my first instinct was, I wonder how that compares with the private sector? I found that the UK national average is 9.4 days per year. So on that basis, the civil service is doing better than the private sector. https://www.cipd.org/uk/about/press-releases/workplace-absence-soars-nearly-two-working-weeks-each-year/?hl=en-GB
«On looking at the sickness data in this article, my first instinct was, I wonder how that compares with the private sector?»
Trivial comparisons between private and public sectors workers are meaningless because the workforces have very different profiles; in particular after many rounds of outsourcing of lower-paid roles public sector roles tend to be more white-collar and middle-class than private sector ones.
Also in the private sector there are many employers with "3 absences and you are out" policies and consequently many employees with "work through the pain" attitudes. Perhaps many taxpayers (especially retired ones I guess) would love for those policies and attitudes to be more common in the civil service too.
That's a fair point, although the CIPD survey is not as representative as ONS data, which shows the public sector is still significantly worse than the private sector on days lost through absences - see link below. Also, as Neil mentioned, the disparities between departments are impossible to explain and suggest that there is plenty of scope for improvement among many of them.
https://www.theemploymentlawsolicitors.co.uk/news/2025/06/11/sickness-absence-2/#:~:text=After%20peaking%20at%202.6%25%20in,has%20fallen%20by%20just%200.2%25.
What you describe sounds like any large organisation anywhere - public or private. The fetish of efficiency you demonstrate - as if there is a perfect organisation with a perfect structure and perfect staff that just needs to be carefully sculpted out of the imperfect stuff of humanity (by you, of course) is fantasy. This is humanity. It is imperfect, not always focused, not necessarily expert, often mistaken, sometimes confused, rarely completely committed. People have lives and flaws and strengths and weaknesses, and stuff happens - all the time. I can absolutely guarantee that there is some dude in Singapore making pretty much the same complaints right now.
People who go on about this are people with very little self awareness, because they always imagine they are a manifestation of the sort of excellence they crave. But...they aren't. They are just someone with some anecdotes, a lot of opinions, and a hustle. What most organisations don't need is someone coming along every five minutes reorganising them according to their latest pet theory. Organisations work because of the people in them. They also fail because of that. But, it is not either or. It's a continual mixture of both. The trick is to try to make sure the first outweighs the second. That is it.
The original post did not contain any anecdotes, but did have a lot of data, all of which suggest some serious problems - useless performance management systems, unsuitable people getting promoted, almost impossible to fire people, very probably bogus claims under the Equalities Act. Maybe you can address those things, which I'd guess most readers would find dispiriting and in need of addressing, not by reorganisations, but by effectively managing people and reducing civil service numbers.
Data… Ah yes, the carefully selected bits that support the conclusion. The point being, that all these things will happen somewhere at some time in any organisation. Even in Singapore. Taking incidents and turning them into a pervasisve pattern, then ascribing that pattern to your pet theory - there is a whole ocean of data to fish in with that goal. Provided you dangle long enough, you’ll catch something. Most organisations are in need of improvement. The real question is, how to improve things without screwing something else up, and if these management consultant types are so clever, how come they are still in business? Shouldn’t they have perfected the design by now?
This is a grift pitch. Look, your organisation is messed up - I can fix it. repeat ad nauseam.
I don't know who you think is grifting, but I don't think your comment is helping anyone. If you don't like data at all, how would you try and run any organisation? If you think Neil's data is flawed in some way, please tell us how.
I respect data. I used to be an information analyst. But this is not data. It is selectively presented statistics. I am acutely aware how statistics can be tweaked to support almost any case. This whole article is a text book example. This is a grift to push his failed and deeply unpopular political party. It resonates with all those people politically motivated to want it to be true. It’s a transparent attempt to project the blame for the calamitous government that he was part of onto others, which wouldn’t fool a child.
I am gratified to see that the US doesn’t suffer from this phenomenon alone. The nature of the US government system “can” provide a certain level of “flexibility “ however it is locality and level dependent. At the federal and state level the vast percentage of employment is considered a social program. Competency or employment preparedness is not generally considered just demographic factors. As there is to some degree a need to get things done, a small percentage of the opportunities are competent. The primary connection between performance and management happen at the locality level, and that is situation dependent. In a practical sense, performance management is a social process not a competency one.
From a legal rather than procedural perspective, is the civil service subject to different employment law than the private sector? Ie) Does it require ""just"" cultural, or in fact also legal changes to change outcomes?