Should Civil Servants be ‘ranked and yanked’?

In a speech to Parliament yesterday, Aung San Suu Kyi appealed for help to build Burma’s civil service. She said, “governments may come and go but a civil service will go on forever.” The irony of asking for such help from a politicians who never miss a chance to mock their own civil service will not be lost on Britain’s public servants.

The government published its Civil Service Reform Plan earlier this week. As it did with its employment law proposals, it trailed the reforms with some bonkers-sounding stuff that made the suggestions which eventually emerged look quite reasonable. I can’t decide whether this is a calculated ploy or just scattergun ineptitude.

In this case, the bonkers bit was even more bonkers than the Beecroft report. As the Telegraph reported:

Steve Hilton, the Prime Minister’s director of strategy who is leaving to take up an American university post this month, has privately told the Prime Minister that the civil service could function effectively with a 90 percent cut in staff.

He recently sent officials to “measure up” Somerset House, the main Government building in the 18th century from which Britain ruled the empire – compared to large areas of Whitehall, Victoria and Westminster today.

One Downing Street source said: “It turned out that 4,000 mandarins could run the whole empire, which rather puts today’s staffing into perspective.”

There is so much wrong with this it is difficult to know where to start. Britain was a simpler place at the end of the eighteenth century. Its population was less than a quarter of its current size and still predominantly rural. Its randomly acquired empire was administered by private companies, like the East India Company, and military governors. By the end of the 1850s, when a more industrialised society was emerging, the empire had been placed under formal government control and the modern Civil Service was formed, there were already some 40,000 Civil Servants. Comparing the administrative demands of the Eighteenth Century to the Nineteenth makes little sense. Drawing comparisons between that and the complex society of the Twenty-First Century is just plain silly.

Whatever else you might say about government ministers, they are not stupid. They know such comparisons are ridiculous. The only reason for releasing this story to the press, and claiming that 90 percent of the Civil Service could therefore be axed, was to wind people up. No doubt it’s part of that ‘creative chaos‘ that Coalition ministers banged on about when they were first elected. Stir things up a bit, scare people, disrupt stuff and see what happens.

It helps to reinforce the notion that the Civil Service is hugely overstaffed and full of time-serving jobsworths who don’t really do much and that, therefore, lots of them need to be sacked.

This, then, was the background to the decision to introduce forced rankings to the Civil Service appraisal process. Forced ranking, sometimes known as Rank And Yank, has been used in a number of companies over the last couple of decades, most famously General Electric under Jack Welch. Forced ranking, as the name implies, forces managers to rank members of their team according to a predefined distribution, so, for example, 20 percent are the top performers, the middle 70 percent are adequate and the bottom 10 percent are classed as poor performers. Poor performers are usually denied pay increases and, depending on the company’s policy, either put on probation or fired.

Despite the folklore around forced ranking, only a few firms actually sacked 10 percent of their workforce every year. Nevertheless, being classed as a poor performer and losing your pay rise is pretty unpleasant. Forced ranking is certainly one of the more brutal forms of performance management.

It has a certain logic behind it, especially in organisations where managers are reluctant to rate people poorly. Many of you will have seen those distributions where most people have been given threes and fours, even where entire departments are under performing. Forced ranking compels managers to have difficult conversations with at least 10 percent of their people. In an organisation like the Civil Service, where managers tend to avoid confrontation and performance management takes much longer than it does in most other places, might forced ranking force the pace a bit?

Perhaps it might but the system has its flaws too. I have seen organisations where almost everyone is performing well. Look around an office where there are lots of committed and energetic people. Could you really say that 10 percent of them should be judged as poor? On the other hand, I have also seen failing organisations where you’d be hard pushed to find 10 percent of staff, let alone 20, to class as high performers. To say that every organisation must have a 20/70/10 distribution might mean getting rid of some good people, or rewarding some bad ones, just to make up the numbers.

Furthermore, the system is no substitute for proper performance management skills. Poor managers will make a hash of a forced ranking system just as surely as any other sort. Cowardly managers (and, as I’ve said, I am one, so I know how this goes) are tempted to avoid giving poor ratings to the stroppy people and give them instead to the people who kick up the least fuss. At the root of many problems with performance management systems is managers’ conflict-avoidance. Forced ranking is not going to make them, of their staff, feel any more confident.

The vulnerability of employees to unfair and arbitrary judgements under forced ranking systems has led to successful legal challenges even under America’s light-touch employment laws. For this reason, too, there has not been much take up for the pure sack-the-bottom-ten-percent version in the UK. Our employment tribunals would almost certainly find against employers in many of the resulting dismissal cases. British organisations that use forced ranking tend to put the lower performers on performance plans and deprive them of pay rises. Which is pretty much what the government seems to be proposing for the Civil Service.

But perhaps the major flaw in forced ranking is that it individualises the problem of corporate underperformance and downplays any contributing systemic factors. As Andrew McLennan says:

Behavioural scientists have known for years that we consistently underestimate contextual influences on individual performance and correspondingly give individuals too much credit or blame for their successes and failures22. Forced ranking systems do not acknowledge organisational performance constraints and arguably even discourage diagnosis of them. Why would anyone tackle the challenge of problem diagnosis believing that fighting a ‘war for talent’ is the key to success?

There are any number of reasons why organisations and individuals under-perform. Assuming that everything can be solved by the annual 10 percent clear-out can blind you to what else might be going on.

Now I should point out that I have never used a forced ranking system. A company I worked for introduced forced ranking just as I left (pure coincidence, I assure you) but canned it after a couple of years. The aggro and morale problems just weren’t worth it.

Other companies are coming to similar conclusions. Opinion is divided on the effectiveness of forced ranking. The evidence so far is inconclusive but managers seem to be voting with their feet. In the US, there has been a move away from using forced rankings in recent years. Many of its former advocates like Microsoft, Goodyear, Ford and even GE have now stopped using it. Some have even suggested that Rank and Yank contributed to the cronyism and groupthink which led to the downfall of Enron.

So is the government getting into something just as it is going out of fashion? Is it about to impose a system on the Civil Service just as that system is being abandoned by many of the people who once advocated it? Frances Maude seems to think it is necessary to adopt such a system in order to weed out the poor performers but there is nothing to suggest that it will be any more effective than any other system at tackling the Civil Service’s poor performance management. Just as bad performance management is not an employment law problem, it’s not a problem with procedures either.

I know I’ve said this before but I still stand by my 4 point plan for improving performance management:

  1. Give managers a good understanding of the law and the organisation’s procedures. This makes the law less intimidating and managers learn to see the disciplinary procedure as a management tool, not an obstacle. De-mystifying laws and procedures reduces the fear of them.
  2. Coach managers in the art of difficult conversations. Most of us don’t like tackling performance problems because it is uncomfortable. Helping people to deal with their own discomfort (for it will never go completely) and have the conversations anyway gives them the confidence to act.
  3. Provide political backing. In the public sector this may mean ‘Big P’ as well as ‘small p’ political. Managers need to be certain that, if they stick their necks out and tackle performance issues, their bosses will not hang them out to dry as soon as things get tricky. Senior executives must back their managers all the way, even if they make the odd cock-up.
  4. Stop colluding with each other. All too often there is an unwritten truce at senior level. I won’t ask questions about performance management in your area if you don’t ask questions about it in mine. If, as managers, we let each other off the hook about performance problems, avoiding such issues becomes normal.

If you have these things in place you can make a system designed on the back of a fag-packet work. If you don’t, even a state-of-the-art one won’t. Rank and Yank is no different. Implementing it is high risk; doing so without managers having the will or the skill to use it properly is likely to be disastrous.

Anything which claims to make it easier to sack large numbers of poorly performing Civil Servants is bound to be well received in some quarters. There is nothing to say that forced ranking will achieve this though. The experience of other companies suggests that it won’t. In any case, there is nothing stopping the Civil Service from sacking poor performers now. The reason it doesn’t has little to do with the appraisal systems in use so changing the system is unlikely to make much difference.

Forced ranking is one of the bluntest instruments in the performance management toolbox. Having seen its limitations, many companies are rejecting it. But, just as everyone else is abandoning it, the government has made forced ranking a key element of its Civil Service reform. That’s a bit like buying the last pair of loon pants in 1977.

This entry was posted in Uncategorized. Bookmark the permalink.

13 Responses to Should Civil Servants be ‘ranked and yanked’?

  1. Pingback: Should Civil Servants be ‘ranked and yanked’? - Rick - Member Blogs - HR Blogs - HR Space from Personnel Today and Xpert HR

  2. Rick, as ever your point is well made.

    But can I add another aspect, too? Forced rankings and 10% dismissal do I believe (still) happen in some major UK companies. e.g. in top consulting…. and then they wonder why so many of their aspiring female senior managers just disappear.

    Truth is I’d suggest, most women – including very talented top-level ones – just can’t be bothered to cope with the angst of a load of testosterone-fuelled alpha males who routinely juggle people’s lives as though they were bits of flotsam. These women, having trained up in such companies, very sensibly leave of their own volition to run their own businesses, before the testosterone drowns them out.

    And the chaps are so convinced of their own invincibility that
    (a) they believe they will never be in that unhappy 10% (oh, eventually, most of them will) of ‘counsel move on’, and
    (b) they think this extraordinary company behaviour, based as you say on completely false logic, is OK.

    So yes, forced rankings do indeed lead to aggressive, risk-taking group think (the sensible ones who dislike huge personal and professional risk have left).

    And this same forced ranking business also results in quite evident gender discrimination. I challenge anyone to look at the middle to senior management progression stats for relevant companies and then see it otherwise.

  3. I suspect, aside from wanting to put a cat amongst the pigeons, Steve Hilton alluding to a 90% cut in the public sector is designed to make a 20% cut seem positively generous.

    Agree with your analysis of the Rank and Yank system. Ultimately a rigid application of 20, 70, 10 won’t work for all organisations, the success of any system will depend on how it is implemented.

  4. B.O. Locks says:

    When Francis Maude was interviewed on radio about his plans to rid the Civil Service of poor performers he finally conceded to a fairly dogged interviewer that poor performers were very rare in the Civil Service. He has since further backtracked by praising the Civil Service in Parliament.. As with most issues involving Maude, it is all spin and no trousers and one has to suspect his agenda. On the one hand he is proposing a radical shake up to rid the Civil Service of poor performers but on the other hand he concedes that the alleged poor performers are very rare. Is this just a ruse to reduce headcount through dismissal via alleged incompetency so as to avoid redundancy and other contractual expenditure? Or he does he do it to provide an impression that the government is doing something so as to allay the electorate’s ill-defined, gutteral grievances? Is Maude a master scape-goater?.I think he is and I am tempted at this point to breach Goodwin’s (?) Law but I won’t.

    Re the “rank and yank” scheme, is this not also another business fad sold to us by yet another business “guru”? In business, it always helps if the label rhymes, doesn’t it?.Workers and Shirkers. Sack me or back me. I recently heard of an employee management scheme christened as “cream and park”. The cream is kept and everyone else is parked (or dumped). Rather charming, don’t you think?. It doesn’t rhyme though, so hopefully it won’t catch on although I fear it already has amongst our big-hearted, socially responsible. corporate friends. A bit of sarcasm there.

    I am just pleased I am self-employed. Markets beat hierarchies hands down.

  5. A distinction should be made between ranking and rank-and-yank. Setting an arbitrary figure for a cohort to be thrown to the wolves is obviously nonsense, but ranking can improve performance management if deployed intelligently.

    The real target of a ranking exercise is not the rankees but the rankers. In other words, “forcing” line managers to collectively assess team member performance can help them face up to performance issues that they would otherwise brush under the carpet.

    They key is to oblige them to rank as a group (which specifically addresses your point 4), and use an honest broker to ensure uncomfortable questions aren’t elided. This is obviously more applicable in an organisation where teams are fluid and there is a lot of interaction between them.

    It also helps if collective management ranking is augmented by peer ranking in a 360-degree review. This allows for the introduction of more subtle assessments into the conversation. The lowest performer in the team, in objective teams, may well contribute more than anyone else to the team’s overall morale and well-being.

    At the other end of the scale, it helps surface the real talent (not always who you might expect) and promote consensus on development and succession planning.

    If ranking promotes discussion and obliges management to be answerable for its own judgement, then it’s a good thing. If it becomes merely a mechanical cull, then it’s a bad thing.

  6. ThinkPurpose says:

    Just ditch them! Not the 10% of staff, but 100% of appraisal schemes.

    As you refer to, contextual factors outweigh individual, don’t concentrate on the visible (the staff cluttering up the office) and ignore the much less visible but more powerful system that produces the work. “Good” staff will be beaten by a poor system every time.

    A good 95% of performance is inherent in the system, by which I mean the processes, the structures, the policies official and unofficial, the things that are paid attention to and rewarded, the performance measures, essentially everything that defines “the way things are done around here”.

    People come and people go, but the system remains. And nobody ever interviews, hires or fires a system. It arrives like an unwanted house guest as the inevitable result of management thinking.

    Ranking of staff ignores basic maths too. Things like regression to the mean, that the poorest performers are more likely to NOT be the poorest performers at the next annual appraisal, due to poor dumb random variation that is everywhere.

  7. Alex says:

    Surely civil servants having good job security is a positive thing. It should be hard to fire them, r do we want to see a return of the kind of political patronage that happened long ago?

  8. Indy Neogy says:

    Great piece. Worth noting a couple of extra points:

    1) Much of GE’s success in the Welch era now looks context dependent on financial engineering that stopped working when the boom ended. Pity Welch’s successor who has to deal with this. At the same time, stop and consider whether that contextual success was any kind of advert at all for any of their HR policies.

    2) It needs to be noted that if you Rank and Yank, but aren’t a better place to work than your competitors (could be better pay, or conditions, or promotion opportunities, or pension, etc.) then in the process of Yanking you will slowly degrade the quality of your workforce. Largely because it’s impossible to rank accurately all the time, partly because of politics, partly because of random variation and partly because if we’re honest, a lot of knowledge jobs are hard to rank.

    3) Enough data – anecdote time – 8 out of every 10 entrepreneurial/innovative people I ever met who worked at GE were in the middle of a plan to leave GE and work for themselves. If you institutionalise risk into the employment contract, you encourage people to ask why they are working for your big clunky organisation in the first place.

    • B.O. Locks says:

      Yes indeed. The sooner the concept of employment is dissolved and replaced by contracts between contractors and self-employed, self-governing and autonomous, human beings the better. Organisations are islands of autocracy that separate employees from markets and the fruits of their labour.

      Markets, not hierarchies!

    • Rick says:

      Thanks Indy, great story. Perhaps if you create such a high-pressure environment a lot of people will yank themselves before they are ranked.

  9. Tyler Murphy says:

    Great post – it’s completely ridiculous to allow this kind of story to surface when all it does, as you say, is rile up the public and feed into common stereotyping.

  10. Sue Smith says:

    A job in a Rank and Yank company might be worth having a go at if you are an ambitious youngster who fancies chancing it in order to see if you can get lucky with those million pound banking bonuses we all hear so much about. But this policy is not just aimed at well paid Whitehall mandarins. It is being ruthlessly applied to all civil servants including thousands who earn less than the national average wage. The low paid have not chosen to gamble on insecure employment and highly driven work environments in the hope of winning big salaries and bonuses. They depend on their jobs to pay their mortgage or rent and feed their children. If the work is no longer there for them to do then they might reasonably hope to be paid some redundancy and to leave with a clean work record, not a work reference that says “dismissed on grounds of capability”. Low paid civil servants have had decades of ever increasing pressure, targets and efficiencies from all governments but government special advisors argue, not on the basis of fact or evidence, but on the basis of political ideology and theory that only the market is efficient so, because they are “public sector” they are, a-priori, inefficient. The idea that the only impact these measures will have is to make some lazy workers a bit less lazy is largely a myth. Thousands are being set up to fail as managers compete to set ever more impossible targets in order to guarantee that at very least 10% cannot meet the targets whatever they do. Otherwise what pretext would they have for declaring the designated percentage of staff as inefficient? It would be a matter of embarrassment for the government if managers failed to deliver their 10% quota. Francis Maude boasts that he can get rid of civil servants and that this is his formula for doing it. He is working hard to save the casino economy but those who try to survive by working hard for a living are being cleaned out to pay for the gambling loses. There seems to have been a change in what counts as “Conservative” values, for example, morality and prudence seem to have got lost somewhere along the way.

  11. Pingback: PIP or Drive for performance improvement? – Naomi Stanford

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s