Lake Wobegon is a fictional town in Garrison Keillor’s long-running radio show A Prairie Home Companion where it is jokingly said “all the children are above average.” While I am conversationally-dangerous enough in statistics theory to explain how this is possible – hint: average of what? – the point of the joke is insightful. Wherever there is a metric that people feel is important – e.g., income, height – it is basic human nature for people to want to think of themselves as “better than average” on that metric. That phenomenom is so typical there is even a name for it – “The Lake Wobegon Effect.” 

One place where the desire for everyone to be “above average” can have negative consequences for organizations is in performance appraisal systems. If everyone is “above average”, what does that say about your average expectations? How does that affect your overall expenditure and distribution of annual bonuses if they are based on performance ratings?

Let’s imagine an example rating system that is typical of many organizations with 5 levels of ratings:

(1) the lowest rating for those who consistently performed below the expectations for their role,

(2) the second lowest rating for those who were inconsistent – a mix of meeting and not meeting expectations,

(3) the rating for those who consistently met the expectations as they were hired, and are paid, to do,

(4) the second highest rating for those who get the job done and sometimes go significantly above expectations, and

(5) the highest rating for those rare folks who consistently performed above expectations (and some may suspect are superheroes in disguise.) 

Which of those rating categories will be the one that has the highest share of all people rated in it? Common sense would suggest the middle of the road rating of 3 – consistently meeting expectations. After all, if your organization met its overall expectations – e.g., its corporate earnings forecast – for the year, you would think that on, on average, the organization’s employees met their individual expectations to get them there.

More typically, however, the Lake Wobegon Effect takes over and the box in the middle – “performing as expected” – is not the most populous (or in statistics-speak, the “mode”. The one right above the middle one – the “performing as expected and then some” is typically the most populous box.

Why?

Reason number one is the Lake Wobegon effect – on average, people don’t want to think of themselves as average. Unless you figure out someway to change human nature, you aren’t going to change that.

Reason number two is something you can have some control over, however. All things being equal, most managers would prefer not to have difficult conversations with the people they give ratings to about those ratings. If they gave people who are meeting expectations that “average” rating, they worry they will get pushback from their employee.

How can you get managers to have those difficult, yet appropriate, conversations? Here are five suggested strategies, from least direct to most direct.

1) Educate – Ignorance can be one reason for inflated ratings. You can train your managers how to appropriately rate people by sharing baseline examples of what an appropriate rating is for individuals. You can create baseline examples for each rating category by identifying example performance appraisals that nicely describe what a person performing at that role looks like. You can use real life examples (protecting for privacy as appropriate) or you can create a made up one from scratch or that is a composite of real ones.

2) Calibrate – Get managers at the same level together to discuss the ratings they are proposing for their own teams and review those of others. By doing so you can get valuable discussions that compare people and how they are rated. “Pat rated Gale as a 5 out of 5 while Jamie rated Taylor as a 4 out of 5. They have similar jobs. Does that make sense to people who have worked with them?” If moderated well, those conversations can be quite enlightening to help managers see how they are applying the ratings versus others.

3) Motivate – Even if managers are educated and calibrated, they may still want to avoid difficult conversations that might ensue if they rated people as they should be rated, not as they want to be rated. One way to motivate them to rate their people appropriately is to make it clear to them that delivering accurate ratings of their people is a major component of their job that they are going to be rated on. Maybe they are lucky enough to manage a team of people that are all above expectations. If that is the case, that team should collectively be producing above expectations. If they are not exceeding team expectations, does that mean their manager is holding them back? Recasting the performance ratings in that light gives managers a check on their desire to take the easy way out, especially if you make it a big enough part of their overall rating, they will take notice.

4) Incorporate – If you do all the above and you find you still have ratings inflation, you can adjust it on the back end by incorporating it into your overall performance management program. If you use ratings to determine an annual bonus payment or salary increase, you can set the overall budget impact of those costs as fixed and let the distribution determine how it is doled out to individuals. If ratings are generally skewed high, that means there are more people than expected in the high ratings category. If you set a fixed bonus or pay increase for individuals in the higher categories, then you will end up paying more in bonuses and pay increases than you budgeted for. If you set the budget for those as fixed and then set up a system to allocate it based on the actual ratings distribution, you will be able to live within your budget. Doing so will also create an interesting dynamic, where highly rated performers will not be getting bonuses and pay increases as big as they might have expected. If your organization wants a “pay for performance” culture, it will find itself in tension with ratings inflation.

5) Allocate – If you do all the above, you will find that there is an incentive for some parts of your organization to be high raters to push any pain off to other parts of the organization. If you want to make sure incentives are aligned within each division, you can fix the bonus and salary increase pools by division. That way, if one division ranks everyone at the highest level, they will not be able to pay as high of a bonus or salary increase as other divisions that do not have everyone in the highest group. The benefit is that market forces would probably take hold at some point, and the best performers would figure out it is more profitable to work for other divisions. That tension between divisions could quickly turn from productive to disruptive if you want to maintain a common organizational culture of “pay for performance.”

Performance management is one of the most important things you do as an organization. It sets the tone for your overall organizational culture. If you tie performance management to compensation, you are doubling down on the importantce of your performance management system. Money talks, and how you decide to allocate that based on performance says a lot about your organizational values. Whatever performance management system you have, make sure you have thought through how it will end up in practice, not just on paper.