- The Myth
The benefits of rewards are so entrenched in our society that the response to the title of this paper will most likely be a reaction of strong disbelief. Everyone “knows” that incentives are how we get things done. We are so immersed in ‘do-this-and-you’ll-get-that’ or the ‘carrot and stick’ philosophy that suggesting otherwise is heresy. From primary school to post-graduate studies and all through our work careers we are coaxed and ‘motivated’ by rewards and incentives. Scholarships and sales commissions are popularly applied incentives regardless of politics, race, sex or age. Our Western culture is totally and thoroughly committed to the use of rewards.
The thinking is that not using rewards is to not pay for performance, which is often viewed as rewarding incompetence, laziness and tantamount to giving a free lunch. (See the first Reference below, from which most of this information is derived.)
- The Reality
The reality is that rewards are used to manipulate the behavior of others. Approval is deliberately, speciously and conditionally given, as all or part of a calculated strategy to shape others’ behavior. Skipping aside the ethical question of this approach for the moment, let’s move directly onto the real results of reward systems based on research that has been repeated over and over again during the past 50 years. (The end of this paper refers to over 600 reference publications of research supporting these hypotheses.) Conclusions about rewards and incentives that fall out of this work are:
- Rewards work in the short term.
- They do not work in the long term.
- Incentive plans are used as a substitute for good management.
- The more rewards are used, the more you need to up the stakes.
- Rewards actually turn off people from the item for which the reward was designed, after the incentive program has been terminated.
A current case in point, at the time of this writing, might be the quarterfinals of National Hockey League playoffs in May 2003 where the 4 finalists are among the poorest incentive-based teams in the NHL – Anaheim, Minnesota, New Jersey and Ottawa. The payroll of the Ottawa Senators at $28 million, a team that finished first overall in the league can be compared to the New York Rangers with its payroll of $80 million, a team that did not even make it to one of the 16 teams of the playoffs. Invariably, people, who use rewards in a failing system, ask: “What can possibly be wrong?” I’ll tell them what’s wrong: incentives don’t work – they actually backfire.
- In Sum, What Really Works?
The essence of what really works is that it takes patience, effort, talent, time, and courage and thought to teach and manage people properly. It takes none of these qualities to use rewards; it is simpler to bribe. Section 13 below outlines the specific alternative ‘what to do’.
- Ethics of Rewards
Are rewards appropriate or inappropriate? Let’s start with conventional Western beliefs:
- People who set their minds to it can succeed.
- People who don’t succeed have only themselves to blame.
- Deserving people can and should be rewarded.
CCCC’s beliefs tend to counter some of the above. CCCC covers these ideas in many of its separate papers, books and manuals, available on request. Summarized, the CCCC beliefs, relevant to this paper, are:
a) Just because you want to succeed and are willing to work very hard to succeed, does not mean that you can or will.
b) Problems can only be solved when the ‘culpable’ person is involved in the solution.
c) Blaming someone for a situation works against being able to resolve that situation.
d) Excellence only occurs from the teamwork and cooperation and the resulting synergy. More comes out of a cooperative group than additive single inputs of each.
e) Equity in a two-person relationship is only possible with equality of influence. By extension this axiom applies to multi-person relationships. In most households parents don’t decide on how much food to give the children depending on whom contributed the most to the household that day. Rewards are a model of unequal status.
f) If you standardize the outcome, there is no need to standardize the means.
g) Problem solving helps develop a sense of responsibility.
h) Weaknesses of people or imbalance vis-à-vis others should not be the focus; instead the focus should be on a person’s strengths.
i) A safe environment is needed for problem solving – to admit mistakes without fear, to trust that there will be no witch-hunt.
j) Individuals need to feel they are owners, not tenants over their jobs or situations.
Rewards should be judged as effective if they create lasting change; their effects should persist when there are no more candies or gold stars to be handed out. By that standard, research suggests rewards fail miserably. Research in 1972 involved many disciplines including seat belt use studies, smoking studies, losing-weight studies. They all found that when the candies stopped, people went back to behaving as before.
A 1961 University of Kentucky study with 72 boys showed the reward group performance was clearly inferior. A 1962 New York University study involving 128 undergraduates showed that the rewarded group took 50% longer to solve problems. Two studies at the University of Texas in the early 1970’s illustrated that the rewarded group got fewer responses right. Four other studies in the same year yielded the same results. Even more sophisticated studies in the 1980s came up with the same conclusions. In the 1986 Jenkins assessed 28 reward experiments – surprise: same results. A 1992 study showed that rewards undermine performance – artists do less creative work when work is commissioned. Other investigative reports showed children receiving rewards had lower standards on achievement tests than those without.
Similar lackluster results fell out of 47 studies of employee participation in decision-making. Rewards in law firm studies came to the same conclusion. Surveys at federal agencies in Canada and US showed that incentives failed to produce the expected results. More recently the highly rewarded executives of Enron, Adelphia, Global Crossing et al. failed to deliver the goods.
- The Effectiveness of Rewards
Rewards induce immediate compliance to your goals. Based on that they appear to be effective. However, they affect what we do, not what we feel. What they do not do is alter attitudes, emotions, senses, personal expectations, wills or commitments.
Rewards are ineffective in the short term when (a) the task is very interesting or (b) when the solution is not obvious, thus, forcing creativity.
Even if we have no reservations about manipulating people, the plain truth is that rewards backfire.
People who are offered rewards tend to choose easier tasks, are less likely to use information available, are answer oriented, illogical, work harder and produce more activity but activity of lower quality with more errors and less creativity.
Not a single study has ever found rewards provide long-term improvement in the quality of work.
- Why are Rewards Ineffective?
a) Punishment is a control mechanism and rarely causes long-term enthusiasm. Punishment is built into rewards because (i) rewards are as controlling as punishment and (ii) losers of a reward competition are demoralized. People, who are punished, although they may respond to avoid more punishment, are demoralized.
b) Competitive rewards fracture relationships stressing the imbalance between two people. Competition adds conflict, jealousies, lack of cooperation, playing favorites, unequal treatment, setting people against each other. “Everyone else is an obstacle to my success.” Competitive rewards create anxieties that affect performance. Those who figure they have no chance, simply give up. In such situations people attribute success to factors beyond their control (luck) accepting less responsibility for their performance. Persons sitting in judgment of you, warp your relationship with them. If people do not feel safe, or if trust is lacking, people will not ask for help when needed; thus, performance suffers. Rewards instill a feeling that one is being evaluated rather than supported.
c) Rewards do not look at reasons why things developed a certain way in the first place. “Why ask why? Let’s fix it with rewards and incentives.” Thus, such practitioners don’t really know what is going on. Rewards are gimmicks and shortcuts that mask problems and ignore reasons. Rewards displace careful management.
d) Rewards discourage risk-taking and narrow the focus. A person working for a reward will do what is necessary but no more. There is no need for the person to challenge the boundaries. Unreflective expedience encourages repetition of what has worked in the past. The goal is not to succeed at the task but to succeed at the reward. The task stands between you and the reward. Rewards motivate people to get rewards, not to do the task.
e) Rewards undermine intrinsic motivation (i.e. self motivation). Rewards tend to kill off interest in the very thing we are bribing people to do. Studies show that rewards actually smother enthusiasm for things that people would otherwise enjoy. Rewarded children will avoid a toy later whereas un-rewarded children will continue to enjoy that toy. We are less likely to engage in a rewarded activity on our own later. The reasons postulated for this turn of events are that rewards devalue the item; if it has to be rewarded it is not worth doing for its own sake. Working for a reward is losing control over our own destiny. Reward interferes with self-destiny, i.e. intrinsic motivation. People feel like Pawns; they feel like tenants rather than owners. De-motivators are: threats, watching, evaluating, forced-to-time pressures, being ordered around and competing where there is only one winner.
f) Pay is ranked #5 or #6 on almost all employee job satisfaction surveys. (Besides, no sum ever suffices.) Self-satisfaction or intrinsic motivation is ranked #1. To assume money is the driver indicates a lack of understanding of human motivation. Incentive plans are based on the model of money being the driver; thus they fail to succeed.
g) There is always a high cost to administer a reward system, to ensure consistency, fairness of measure, accuracy and adherence to rules. Individual records must be meticulously maintained. The cost for the care and feeding of any award system is immense. Ask the airline points administrators. My own frustration with airline points as a business owner was finding staff choosing airlines, not by cost but by the reward – the frequent flyer points.
Praise if inappropriately applied, falls into the same category as other forms of reward; namely, it can be dangerous to real achievement. Praise can provide two types of feedback, one good and one bad.
The first is a straightforward description of information on how well one has done a task. It results in encouragement that leaves the recipient feeling in charge, self-determining, and intrinsically motivated.
The second is controlling by making someone dependent on approval.
In either case the question to be asked is: Does the praise benefit the giver or receiver? If it benefits the giver, it is manipulative, power seeking or controlling. It is more of: ‘do- this-and-you’ll-get-that’.
Praise is positive when its purpose is towards (i) improving performance (ii) encouraging appropriate behavior or values and (iii) helping individual feel good about themselves.
Studies confirm that most praise does not correlate with achievement gains. Praise, like other rewards, has been shown to even be a disincentive to do as well as was happening before. Praise drags down performance. Why? Among the reasons are, praise, if too freely given (i) leads to a sense of low expectations and thus decreased effort (ii) can increase the pressure to live up to praise which makes the person more self-conscious (see 7e above) (iii) sees risk decrease (see 7d above); a counter suggestion by the one doling out praise will have the person dependent on praise, withdraw their original suggestion (iv) undermines intrinsic motivation (see 7e above).
Praise bolsters the power of the person giving praise. Telling someone their work is good is every much a value judgment as saying it is bad. The problem with a positive judgment is that it is a judgment. It can make a person dependent on us – on our approval. Praising children for the way they behave gives them no reason to continue acting responsibly. They respond only to the bait.
- Making Praise Work
The first test for making praise work is: Is the praise helping the individual develop a sense of personal control? Are we encouraging the person towards self-judgments about what makes good performance?
A second test centers on intrinsic motivation: Are we creating condition for the person to become more deeply engaged in the activity? Or are they doing it to win our approval?
A third test is to evaluate our own motives. Are we controlling someone’s behavior for our own convenience or to get that person to like us? Both are manipulative.
A fourth test is to ask ourselves: How do our comments sound to the other person? Does our encouragement sound like we are limiting the other person’s autonomy? (Simply ask the recipient of the effect of the praise and note if there is resistance or indifference.)
To praise effectively, provide specific information of how well someone has done and why it is good. “Joshua, thank you very much. Your completing that assignment on Wednesday means I can deliver my report by Thursday.”
- Deeper Reason
Deeper reasons for rewards’ ineffectiveness begin with an understanding that an attack produces a defensive maneuver, which is driven by our basic instinct for survival. A survival response is human and quite normal. In fact it is a necessity. Punishment, a form of attack, produces resentment rather than responsibility. It results in defiance, defensiveness and sometimes, rage. Punishment is a means to control. Thus any attempt to control produces resentment – and a corresponding backlash. People respond to punishment (attack) in kind rather than with a new determination. “I’ll take the reward and deplete you of your bank account just to show you, but damned if I’ll lift my finger one iota more.” Since reward and punishment are the flip sides of the same coin, then rewards work against real results.
What drives people? People left to their own devices search for and overcome challenges; they try to master skills; they attempt to attain competence; on achieving competence, people seek new levels of complexity in what they do. People in all walks of life, in all ages, follow this pattern.
A rewards system is a model of a human relationship based on one person controlling another. It is a model of unequal status (in violation of 4e above). It is used to pressure people to do what they wouldn’t freely do. “Do this and you’ll get that – whatever the giver deems appropriate to the situation. Thus rewards emphasize the greater power of one person over the other.
Rewards work against solving problems together (in violation of 4d above).
Rewards benefit the person giving them but they are frequently justified as being in the interest of the persons receiving them.
- What not to Do
Do not use rewards as a means of control. Ask yourself some questions first:
- Is the intention of you as a rewarder to teach a skill, promote a value, boost self-esteem of the recipient or is it to get what you want?
- Does the rewardee feel manipulated?
- How important is the reward that is given? How big is it? How closely is it tied to quality of performance?
- Does the award avoid driving the actions of rewardee to satisfy your own needs?
We can’t avoid some control, e.g.: controlling children from running out onto a heavy traffic street. The key is to keep it to a minimum, to maintain a balance between necessity and manipulation.
Do not apply collective punishment. It is viewed as unfair. Collective reward is equally unfair.
Don’t offer positive comments that primarily benefit what you want.
On a personal side, avoid the tendency to offer praise to disadvantaged (abused) people that is conditional on love. They need unconditional love – no strings attached – namely support and approval.
Rewarding for quality is a fools’ errand.
Don’t leave airline points in the hands of employees (your air travel expenses will increase).
- What Will Happen
With a reward system in place you can expect any of the following:
- Hidden mechanisms to get job done
- Hidden production or cached items
- Phony, manipulated records
- Hostility towards those who administer the reward plan
- Cynicism regarding managements fairness
- Lack of cooperation with other parts of organization
- Collusion to defeat the incentive system
- Inhibition to real quality
- Inhibition to real productivity and output
- Short term thinking and behavior
- Destructive approach towards long term thinking or planning
- Bitter losers
- Assignment of responsibility without authority – for factors beyond their control
- Lower return to shareholders (see above, Section 5, last paragraph).
- What to Do
Work with the person to get results without resorting to rewards. Get to the managing style of employee-by-employee, one-by-one. This approach requires less control, but more work. Thus standardize the outcome, so you don’t have to standardize the means.
If there are rewards:
- Provide unexpected rewards, surprise, after the fact.
- Reduce their conspicuousness.
- Avoid contests – make the award available to many people.
- Make reward similar to the task. If you feel compelled to give a person a reward for making a better loudspeaker, make the reward a loudspeaker. (It means they are doing it for the love of the loudspeaker, not for a distracting reward.)
- Have potential recipients participate in the ‘what, how and to whom’ of the rewards.
If there are public awards:
- Avoid competitive comparisons.
- Make them fun for everyone. Awards must not put down losers.
- If public awards are for your own selfish motives, state these at the outset.
- State the awarder’s subjectivity.
- Award for things not really likely to change a person’s behavior patterns, e.g. award for person who traveled the furthest distance to get there is not likely to encourage others to travel farther.
If there is praise:
- Don’t praise people; praise what they do.
- Make praise specific. Avoid phony praise (observe your tone of voice – high pitch, singing voice, usually means phony).
- Avoid praise comparative to others.
- Offer positive comments that benefit the recipient.
- Offer positive comments in private – not publicly.
- Give information that describes how well someone has done. Keep to the facts.
- Give negative feedback in a factual way – not to kill interest. Describe failure in terms of the problem to be solved.
- Stop keeping compensation secrets. People will assume, and usually assume incorrectly upwards, especially with executive compensation.
- Treat people well: provide useful feedback, meeting needs for self-determination; provide social support.
- Have employees participate in setting their own goals.
- Remove constraints to motivation, i.e. the de-motivators.
- Decouple compensation reviews from performance reviews.
- Remove unequal ‘incentives’ for career paths between specialization and management.
- Consider replacing commissions with straight salary.
- Do performance reviews frequently, at least quarterly.
- Focus reviews on future, not past.
- Find the thrills for each employee via a “discovery” program.
- Focus on outcome, not means to achieve it for each individual.
- Match responsibility with authority. Decentralize authority.
- Listen to obvious signs of discontent: frequent absences, bitter or over-serious attitudes.
- Get all employees involved on solutions that affect them.
Rewards do not work. In fact they are counter-productive. The evidence is all around us; just examine the past effectiveness of your own reward programs. Observe the companies asking: “Why can’t we get this to work?” They fine-tune the incentive program instead of looking elsewhere. For your own business, we suggest you focus more on managing people than on manipulating people with rewards.