It's been a long-held assumption that human performance fits a normal distribution — a bell curve in which only a very small number of people are outliers.
When applied to work performance, it basically means that most performers (68%) are in the middle and are average performers while 16% are high performers and naturally the other 16% are low performers.
While easy to understand, this model does not accurately reflect the way people perform. Research conducted in 2012 by O’Boyle and Aguinis with 633,263 people in a total of 198 samples found that performance in 94% of these groups did not follow a normal distribution. Rather they fell into what is called a "Power Law" distribution. Most performance was generated by a small group of “superstars”.
This graph of employee performance looks like a ski slope (red) where the under-performers are at the upper most starting point of the slope and high performers are at the tip of the end. The typical/average performer would fall below the mean or average result. Again, in this model, the typical performer is “below average” (speaking from a statistical perspective).
There would be approximately 80% below average, 10% around the middle, and 10% exceeding; meaning that the assumption that the typical performer is average is a myth of the normal distribution assumption. The results lend themselves to the “80 / 20 rule”, which says that 80% of the work is done by 20% of the people.
To quote Josh Bersin:
“The really big difference between the “bell curve” and the “power curve” is that the power curve reflects the fact that a small number of people deliver an inordinate amount of contribution. This means that "most people" are below the mean. It does NOT imply that most people are lower performers, only the fact that the variability of performance is high and that the curve should not be equal above and below the mean.”
Sounds cold doesn't it? People believe the bell curve is "fair." It’s nice and neat. There are an equal number of people above and below the average. And fairness is very important. But fairness does not mean "equality" or "equivalent rewards for all." High performing companies have very wide variations in compensation, reflecting the fact that some people really do drive far more value than others.
Given the research findings, doesn’t this indicate that compensation is inefficiently distributed? If you believe the results, then at the very least we need to examine the design and focus of our programs. Perhaps it’s time to rethink them - and find ways to create a higher variability in pay. Think about paying people based on the value they deliver.
Maybe we need to shift focus away from the majority of the current workforce and to the superstars that have the greatest company impact. The revenue producers. Those with critical skills that can “make or break” the company.
Maybe when you find someone who's truly outstanding, the normal pay rules shouldn't apply.
What do you think?
Jacque Vilet, President of Vilet International, has over 20 years’ experience in Global Human Resources with major multinationals such as Intel, National Semiconductor and Seagate Technology. She has managed both local/ in-country national and expatriate programs and has been an expat twice during her career. Her true love is working with local national issues. Jacque has the following certifications: CCP, GPHR, HCS and SWP as well as a B.S. and M.S in Psychology and an MBA. She belongs to SHRM, Human Capital Institute and World at Work. Jacque been a speaker in the U.S., Asia and Europe, and is a regular contributor to various HR and talent management publications.
Bell curves with symmetrical distributions only appear in robust collections of random samples or universal populations. A workforce of carefully selected employees does NOT fit those criteria. High-end outliers are treasured and coddled while low-enders are lopped off rather than retained. I've always described the distribution pattern as like a baseball cap with a long bill for the right-side tail.
While this may be news to accountants and engineers, it should be elementary Comp 201 in our profession.
Posted by: E. James (Jim) Brennan | 04/08/2014 at 01:42 PM
I like your baseball cap analogy. You say it should be elementary Comp 201 for our profession: Well most companies out there use the bell curve. So guess not many of us have taken Comp 201.
Posted by: [email protected] | 04/08/2014 at 05:07 PM
HR and comp people should know the facts because they see first-hand how employers value key performers while eagerly shedding the opposite type. Most of our work deals with those two opposite outlier group. It tends to be top management (generally sales or accounting types) who impose destructive policies based on unrealistic bell curve assumptions.
Or, I could be wrong...
Posted by: E. James (Jim) Brennan | 04/08/2014 at 06:57 PM
Good article.
Glad to have people realize that the Bell Curve only applies to REALLY big data. Even then it is suspect. But, I think we need to be careful when, in the name of science, a new paradigm is defined that is supposed to represent the "right way".
I have seen companies where performance is carried by 80% of the employees and companies where it has been carried by 5%. I have seen almost everything in between.
It should be noted that the Power Law (Theory, Principal, Hypothesis) almost perfectly reflect wealth disparity in the US where the to 10% hold about 80% of the wealth (or something like that). But, in other very successful countries (like those in Scandinavia) the distribution is far flatter, yet total results fairly comparable.
The Power Law may be another example of how BIG data, when looked at from a specific angle, provides a "law" that is unsupported by specific data looked at from the perspective of reality.
The days where I believe a simply rule can apply to all (or even most) companies are essentially gone.
I have seen "lumpy" performance contributions and incredibly flat contributions.
Most companies are far too small to apply any type of generic rule. This is why strategy and culture are such big components of compensation, and many other types of, success.
Posted by: Dan Walter | 04/09/2014 at 05:06 AM
I am not surprised by this as it mirrors my experience that, even in large multi-national companies, the people who can, and hopefully do, "put the bread on the table" are relatively few, rare and thus in high demand. Most people, due a number of factors including the opportunities open to them from the organisational design and processes, and their own level of motivation and skills, do a perfectly good, at times great, job...but are not "key talent".
I guess the ultimate example could be a movie...the list of people on the credits who all had to work as well as they can to MAKE the movie goes on for minutes...the people who actually make the difference between whether the movie is any good and is profitable, is very small.
Posted by: Michael Cope | 04/10/2014 at 11:24 AM
If we apply the concept of performance and who does all the work to our compensation structures, then we wouldn't limit the size of increase a top performer that makes more than the mid-range (I understand that not everyone employs this concept). We would throw the "max of the range" concept out the window and pay the top performer appropriately based on his/her contributions. If it happens to be above market, so what! If the 20% leave for better opportunities (compensation, benefits, recognition, etc.) then we either have to replace them with other 20%ers ......who would probably cost more than the 20%ers that left...... or count on some of the 80%ers stepping up.
Posted by: Sherlyn Keiling | 04/10/2014 at 02:06 PM
Thanks Dan and Michael.
Dan -- one could argue that the 10% who hold most of the wealth are not necessarily top performers --- should be but maybe not!
Michael --- yeah this is the tough part. When a company has solid "Joes" and "Janes" that are really great performers they may still not be the people with key skills. And it goes against the grain not to treat them as special as you do the ones that "bring home the bacon".
Sherlyn --- Great! I personally don't like to argue about "nits" on not giving key/critical people increases because they are at the top of the salary range. I'm afraid that when Comp pros make this argument it shows they don't understand the realities of business.
Posted by: [email protected] | 04/10/2014 at 02:24 PM
This reminds me of a post I wrote a while back “Rewarding ‘Hyper-Performers’” and the need to rethink typical compensation rules.
http://integrated-rewards.com/2014/02/rewarding-hyper-performers/
Posted by: Trevor Norcross | 04/10/2014 at 03:02 PM
Like the article Trevor! Unfortunately we still have managers and comp professionals that only think inside the box and have difficulty working with blurred lines or no lines. The more we design our comp programs to support business needs and recognize the performers that deliver, instead of spreading the peanut butter, the more effective our programs will be to drive better performance both at the company and individual level.
Posted by: Sherlyn Keiling | 04/10/2014 at 04:10 PM
Great article Trevor!!
Posted by: [email protected] | 04/10/2014 at 04:43 PM
The Power Law does not improve on the fundamental flaw of the Bell Curve, rather it simply takes it further along. The issue remains that both theories confuse Talent with Performance. Yes it is true that in a group there will be a few outstanding individuals and a larger number within the same talent spectrum, however this is merely a predictor of performance and does not always translate to performance which is simply delivering against set objectives. The reason is that work within such groups becomes distributed in such a way that gives more responsibility and accountability to the outstanding few which then translates to the 80/20 rule, but everyone does perform as your less talented majority still do what's expected of them (sometimes a little more) and that's where most performance management systems fall flat!
However, this is where great managers and great companies differ.....everyone gets an equal shot with emphasis on team work and not the individual. Analogies are abound in team sports where a group of less talented players can defeat a team of immensely talented players by sticking together to make the whole greater than the sum of its parts. That's the holy grail of management, that's what HR has to help deliver.
I think the Bell Curve if used correctly can help to motivate for high group performance, as it is fundamentally a reward tool, a means to distribute limited reward spend in a way that ensures your best performers get enough to keep them engaged. It should be dynamic i.e. multiple variations used to define what performance at different levels looks like and how rewards will be distributed. This should be defined at the start to motivate for higher group performance and achievement of aggregate objectives while providing a mechanism for identifying top performers and rewarding them accordingly.
Outstanding talent in critical roles could be handled within a process that provides a "talent premium" if the retention of such talent is fundamental to achievement of business objectives. This should however be aligned to a Pay for Performance philosophy i.e. outstanding talent should actually deliver outstanding performance.
Posted by: Oluyonmi Okunowo | 04/11/2014 at 01:41 AM
You make some very good points Oluyonmi. Thanks for your input!
Posted by: [email protected] | 04/11/2014 at 10:13 PM
Agree with Oluyonmi driving motivation and Jacque with the premise that the bell curve may not always be most appropriate. In all cases, the business environment should be evaluated to determine best approach for distributing limited funds. I would propose shifting the curve as needed ad it is not a one size fits all. Or, if the power curve is more appropriate for the business, then use it. Whatever model is chosen can be used in budget modeling to most appropriately distribute the funds. That is the value that the compensation professional brings to the table in a business strategy decision.
Posted by: Karen D | 04/12/2014 at 01:05 PM