David Maister - Professional Business, Professional Life
David’s ResourcesAbout David
NEW! Browse my materials by topic of interest:StrategyManagingClient RelationsCareersGeneral

Passion, People and Principles

Stop Paying for Performance

post # 43 — April 3, 2006 — a Managing post

The disadvantage of pay-for-performance compensation systems is that they provide a wonderful excuse not to manage. If someone’s performance is down, instead of management seeing that there is an obligation to go help that person, they have a wonderful cop-out. They say – “We cut his pay. We’ve done our job. The rest is up to him or her.”

Contrast this with what happens in a system where everyone gets a relatively fixed salary or share of profits, which changes as you get more seniority – if you survive.

In such a system (often called a lockstep system because people move in lockstep up the pay scale), if someone underperforms, you have only two choices. You either (a) work with that person and help them improve to deserve the same income as their peers or (b) if you cannot restore them to full share, you have to ask them to leave. Notice that having lockstep without the guts to deal with performance issues is clearly a disaster! A fixed-share or fixed salary system FORCES YOU to manage, ie to be intolerant of underperformance.

In other words, by not paying for performance, you end up with higher performance by tackling performance issues. By paying for performance, you get less performance because the system allows you to accommodate underperformance.

As Gilbert and Sullivan said in The Mikado – A most ingenious paradox!

There are further problems with pay-for performance systems. If you have lived for the last 20 years in a performance-based system, which is almost always based upon individual pay, then by definition you will have no institutional loyalty.

The system has said to you for the last 20 years – the heck with the firm, you will be paid on what you do. We neither require of you to have any loyalty nor do we give you any loyalty, because if your numbers are down, we cut your pay.

So, the individual pay system is designed to minimize firm cohesion and thereby create the sort of “jumping ship” syndrome that we have seen in every profession.

If, however, you have lived for the last 20 years in a system where everyone is in the same boat together, and have all risen and fallen together, based upon sharing in firm-wide results, the probability that the people will stay together through hard times is measurably higher.

For the last 20 years firms have gone out of their way to say, “You are on your own. Never forget that.” That’s the message that every partner in the majority of professional firms gets from their firms.

There has been a very unfortunate history with all this. Many countries (for example, the UK and the Netherlands) had two ancient traditions, which proved to be unfortunate in combination. One was a tradition of lock-step compensation. They also had a tradition of being “tolerant”. Collegiality in Europe meant – you leave me alone about my flaws and I will leave you alone about yours.

The combination of these two approaches were disastrous. If everyone gets a fixed share regardless of how they perform, and trhere is no tackling of performance, you get the famous “free rider” problem.

About twenty years ago, firms began to realize that they had to either start managing their people, or they had to start dropping the equal salary or equal share system, introducing pay for performance. The bad news, of course, is that instead of doing the sensible thing, which is going toward an intolerant, fixed share or salary system – in other words, start MANAGING – they decided to go the other way and remained tolerant (unmanaged)and switch to pay for performance to accommodate wide ranges of contribution. It has been disastrous.

Compensation as an incentive system is a cop-out or a crutch, just an excuse for not providing proper management.

There are two ways to assess the effectiveness of a compensation scheme. First, is the money going to the right people and, second, what is the system doing to create desired behaviors and prevent bad behaviors?

A compensation system is and must always pay those that have done well. Most systems end up paying the right people. However, the design of the system also influences how people behave, i.e. what aspects of performance they pay attention to.

In assessing a pay scheme, you have to ask whether the system is getting people to do the right things. Most are not. Any time you try to use compensation as an incentive scheme, it will backfire.

The minute you tell a bright person what three things will be rewarded, you give them permission to ignore the ninety-seven other things they should be doing to make the firm work. People do like specificity. Every year people will come to you and ask what three things they have to do to get a good bonus. And every year, you must say: “Gee! I don’t know, and if I knew I certainly wouldn’t tell you.”

You must pay people as well as you can. You must be generous and fair, but you must never reduce the compensation system to a formula based upon a limited number of things. With a formula, you give people the permission to be an ugly human being and no one can do anything about it.

The problem is that firms are trying to use compensation as a means of raising performance, which it is not well-equipped to do. Most compensation schemes in professional firms are basically saying, “If you succeed we will pay you.” This is a wonderful system for someone who is already a natural superstar and knows how to win, but in a sense unneeded for the superstar. It doesn’t add anything to the success of the firm because the superstar is going to win anyway.

For someone who is not a superstar – and doesn’t know how to win, doesn’t know how to develop business, doesn’t know how to delegate, doesn’t know how to get organized – then just saying, I will pay you if you do it, is not exactly a tactic that is going to work. For the person who doesn’t know how to raise his or her performance, changing the incentive won’t help. And what about the person who underperforms? Is cutting their pay really the right way to energize them to perform better?

If you really want to raise firm profits by increasing performance, what you must do is manage people, and that involves one-on-one interactions. Talk to people and help them raise their game, by encouraging them, exciting them, lift their enthusiasm, and help them deal with their issues.


Ian Welsh said:

An interesting article.

“A compensation system is and must always pay those that have done well. Most systems end up paying the right people. ”

I think this is more true in professional/partnership firms (which, of course, is what you’re discussing) than it is in general corporations where it seems that senior executive and CEO recompense is very much detatched from performance. Hewlett Packard and Delphi are high profile recent examples, but in general CEO salaries don’t seem to have strong correlations with firm performance, however you want to measure that performance. The same appears to be true of senior executive comensation and bonusses.

Adopting the pay for performance in general firms has been even more destructive than it has been in professional firms.

posted on April 4, 2006

Doug Fletcher said:

David: This article really got me to thinking…and I see your point. I’ve tried both approaches with my consulting firm- the pay for performance is harder to implement and manage than a simple profit sharing program. However, I can remember when I was an employee – that I felt profit sharing systems under compensated the true achievers – and over compensated the dead wood in the firm (which there is always some around however well it is managed). This tended to de-motivate me when I saw that I was getting paid the same as a mediocre performer. I know with a lot of jobs its hard to know who is really doing the heavy lifting versus just going through the motions. Some positions, however, like sales positions – it is easy to measure one’s success over a long period of time. How do you feel about pay for performance for salespeople? Thanks.

posted on April 5, 2006

David (Maister) said:

Ian, thanks for your coments on the corporate world, which I know less well. I’m intrigued that you say that pay for performance has been even more destructive than in professional firms. I’d love for you to elaborate, if you’re willing.

Doug, I still oppose pay for performance schemes for salespeople. They always end up selling what’s incented by the commission or pay scheme, rather than (a) what’s profitable to the company (b) strategic© in the clients’ best interests. You always end up with transactions not relationships.

Don’t get me wrong – I’m all for making sure that those who contribute more do (over time) earn more. I just want it to be a “smoothed-average-over-time” judgment scheme, not a finite measurement scheme based on a limited number of factors.

As Alfie Kohn pointed out in his book PUNISHED BY REWARDS, its not that measure and reward systems are not powerful, it’s that they are TOO powerful – they encourage people to forget everything not included in the measurements.

posted on April 6, 2006

Greg Magnus said:

David, thanks for the post. You make several excellent points, but I respectfully disagree with a few of your conclusions as does a large number of the best managers in the country according to Gallop’s research (see “First, Break All the Rules: What the World’s Greatest Managers Do Differently”

by Marcus Buckingham and Curt Coffman; and “Now, Discover Your Strengths” by Marcus Buckingham, Donald O. Clifton).

For example, you state, “The minute you tell a bright person what three things will be rewarded, you give them permission to ignore the ninety-seven other things they should be doing to make the firm work”. If this person’s talents lie within those three things and their performance is outstanding in this area, eliminate the other 97 things from their schedule. Give someone else those tasks if they are critical in meeting the firm’s business goals. I agree managers are paid to manage, but I also believe the majority of their time should be spent finding people’s true talents and putting these people in positions that use those talents. Management isn’t about fixing weaknesses or trying to “encourage and excite” people to do things they don’t naturally do well.

“For someone who is not a superstar — and doesn’t know how to win, doesn’t know how to develop business, doesn’t know how to delegate, doesn’t know how to get organized…” this person shouldn’t be tasked with any of the above since they don’t have the right talents for the position. All the one-on-one time in the world will only raise their performance a meager amount and the return on your “management time” investment is limited. If the position requires this person to be good at these things, you have the wrong person in the wrong position. It has nothing to do with incentives and pay.

Concerning your response to Doug, “I still oppose pay for performance schemes for salespeople. They always end up selling what’s incented by the commission or pay scheme, rather than (a) what’s profitable to the company (b) strategic© in the clients’ best interests. You always end up with transactions not relationships.” In this scenario, the company set up the wrong pay scheme if the salesperson isn’t paid the best commission on the most profitable sales. If the goals of the company include building relationships, pay the salesperson based on the number of repeat sales and survey feedback from clients. It isn’t the salesperson’s fault he’s good at selling based on what he’s paid most to sell – that’s exactly what we want!

posted on April 7, 2006

David (Maister) said:

Greg, I like the Marcus Buckingham stuff, too, and think what they have to say needs to be taken seriously. However, I’d continue to keep the debate going on your excellent points by arguing that there are limits to the ability to fit each person into a role where they only have to play to their strengths.

Accepting your point fully – that people should be put in roles that play to their strengths – there still will be the need to help people perform. I don’t accept that it’s ONLY about getting people into the right role.

All I’m arguing is that I’d rather help people perform by talking to them rather than just by trying to create incentive schemes.

And, for what it’s worth, I have less faith than you do that the incentive scheme can be designed to incorporate all the relevant strategic considerations. Sales people are smart enough to game any system, so my opinion – yes, opinion, no more – is that I’d vote to put in place a judgment system, not a formula-based incentive system.

posted on April 7, 2006

Gene Turner said:

As a practising lawyer who is not a partner, I’ve always been interested in the concept of pay for performance as it applies to non-partners, and think it would be a good thing if designed and implemented correctly.

It strikes me that as opportunities for partnership have decreased and as more people choose not to pursue partnership anyway, there are reduced (potentially no) incentives for lawyers to exceed expectations (whether in terms of chargeable hours or other valuable investment activities, such as precedents, or other knowledge management). Many firms provide a fixed base salary, with no performance bonus at all. Even if management systems mean it is not possible to free-ride, there is little incentive to do anything more than meet the expectations implicit in the salary.

It seems to me that if a firm were to set a reasonable financial budget for which a reasonable salary would be paid, and were to then offer a meaningful share in any surplus above that (using at least the same proportion as salary is to fee budget) then incentives would surely be higher than in a case where 100% of any surplus accrues to the partnership.

To counter individual incentives leading to undesirable behaviour, the scheme could be applied on a team basis (hopefully the teams would be big enough to avoid inter-team competitiveness issues) yet small enough to ensure individuals felt that they had some impact on team performance.

As well as including financial measures in budgets, I would have thought budgets should include a component for “investment” (client development/precedents/other knowledge management/training – whatever is most relevant to the individual and the team from year to year) that should be met every year. In addition, there could be some flexibility to enable employees to credit up to a certain amount of additional investment time against financial budgets (which would improve incentives to do something productive with downtime that is out of their control).

I’m sure this would not be perfect, and that there would be some anomalies along the way, but would it not be better than having nothing at all?

Are there firms out there that currently do anything like this, and does it work?

posted on April 11, 2007