Business Today


Is it Time to Bid Adieu to the Bell Curve?
     Print Edition: August 30, 2015
Debate: Is it Time to Bid Adieu to the Bell Curve?
Photo: Reuters


Bakul Dholakia
Director General, International Management Institute, New Delhi

For times immemorial, organisations have endeavoured to get the best out of the people working with them. But now, performance appraisal is all about performance management. It has become one of the most difficult and tricky terrains for managers, worldwide. Systems have evolved as a holistic developmental tool for managing the performance of people in an organisation. The use of quantitative tools to get an objective view of the performance of an individual also came into the picture.

Then, in 1994, Herrn-stein and Murray came up with their seminal work: The Bell Curve: Intelligence and Class Structure in American Life. They argued that intelligence is measurable across racial, language and national boundaries, and is one of the most important factors related to the economic, social and overall success of a country. They said it was not possible to manipulate IQ and, therefore, future success manipulations were unlikely.

Soon, it was adopted across the globe. The bell curve demanded organisations to classify employees under three categories: high, average and low performers. While there were significant advantages of its use, there were disadvantages, too. The model suggested that high performers will be further motivated and continue to improve if they were given incentives such as stock options, performance bonus, paid holidays, etc. This, in turn, would make the average performers aspire to be part of the high performer category and with the help of training modules, improvement programmes, etc., it will increase the overall productivity and ensure accountability across the organisation. The bell curve was appropriate for an organisation with a large workforce, but could not be effectively implemented on a handful of people. The doubt about the fairness of the classification also loomed large.

The context in which it was introduced was very different from what it is now. Today, generations that were subjected to this system are very different in terms of aspirations and ambitions, psychological make-up and mind-set, and value system. The times when employees were considered as a factor of production have also evolved to a time when employees are acquired and retained to help them realise their potential for growth and performance. Earlier, there were sufficient examples where employees would be categorised in a lower category despite very good performance. If this happened once, it could be a one-off incident, but if it happened repeatedly then it would result in enhanced levels of anxiety and frustration, and would become counterproductive for the organisation.

Some organisations realised the negative impact of the system and did away with it. Like all systems, implementation was key to the success or failure of the bell curve. Paucity of time on the part of the manager and little understanding about the staff lead to failed implementations because everyone wants to get over with the process. Sincerity and developmental assessment is what is required. This would ensure any such system of classification works. It is important to have classification, but that should not mean we place people in them to make it look complete. It seems the time has come when you need to have more categories instead of the existing three: high, average and low performers, and a real-time approach to putting each employees performance on record as and when its happens. Any performance management system has to have both rewards and penalties, but should ascertain and achieve a productive outcome, as against a mix of productive and counterproductive, or under productive, outcomes. The underlying principle has to be about leveraging the potential that employees bring to the organisation and making the best use of it. That will come only with a positive productive view to performance assessment.


Adil Malia
Group President (Human Resources), Essar Group

It is high time that the corporate Rip van Winkle wakes up to write an obituary to the bell curve. And those who successfully do it could receive the much-coveted corporate No-bell Prize!

This debate is the outcome of what exactly happens when a statistical tool introduced to check the robustness of a process is maniacally pursued by its protagonists - beyond a point the process and its purpose gets forgotten and the tool pretends to be the original process. And, as a result, this annual ritual sacrifices many deserving executives.

If there is any one process that has alienated and mispositioned HR managers vis-a-vis their business colleagues and rest of the organisation, it is the bell curve process. The misguided arguments of the protagonists in favour of the process would leverage around the following logic: Companies cannot afford to adopt absolute performance rating systems when the markets they operate in have a relative performance system; In absence of a viable alternate system there is a need to continue with the old devil; Bell-curve is the best and the fairest system to distribute rewards and compensation; and, every organisation needs a logical tool to explain away their reward distribution philosophy and the bell-curve has a statistical normal distribution logic supporting it.

For the uninitiated, statisticians call bell-curve, a 'normal distribution' curve. The model assumes that we have an equivalent number of people above and below the average and that there will be a very small number of people - two standard deviations above and below the average. In corporate application, it effectively means that if the final performance ratings of all employees were plotted on a graph and if that graph threw up a bell like parabola shape, the logical conclusion would be that overall a fair, normal and robust performance rating exercise was undertaken. But that is where the Achilles heel pains. Managers have, over a period of time, allowed the bell curve test to masquerade itself as the sole purpose of the performance management process.

Strong prosecution arguments, however, back the plea for capital punishment to this rogue tool.

First, we are in the knowledge era. The bell curve performance is more suitable for classical industry workers whose operational performances could be accurately measured by outputs. Knowledge performers are high-quality input drivers.

Second, we live in VUCA (volatility, uncertainty, complexity and ambiguity) times where nothing is normal. Goals shift, strategies evolve, new challenges emerge and employees move from projects in quick succession to cope with dynamic market challenges. How can a performance rating distribution pattern follow a normal curve, which is otherwise suitable for a sub-normal business climate?

Third, employee performance has to be decoupled from its compensation or rewards distribution agenda. Performance review in VUCA times has to focus on development, coaching and performance enhancement of employees.

The forced ranking process results in capricious rankings, power struggles among managers and unhealthy competition among colleagues. It not only diminishes the value of top performers, but also pushes away mid-level performers into the bottom level, making them look like 'losers'.

Mathematical constructs have limited application to human behaviour, more so, when it has a developmental agenda with an emotional anchorage.

The bell curve applied to people performance creates more problems than what it strives to solve, including motivation and unfairness. The new world has no space for a rusted tool of the past. Let us bury it and on its tombstone let us write an epitaph: "B.Curve, Rest in Peace. Forever. We are preparing for the normals of the new world!"

  • Print

A    A   A