Last Updated Jan 3, 2011 12:21 PM EST
Rather than set up an experiment to observe peoples' actions under certain conditions, Herold instead used game theory, a branch of applied mathematics that relies on mathematical proofs to predict how individuals will behave under certain scenarios. Herold's proof assumes that no matter what the situation is, individuals act in a way that is in their best interests. Herold also assumes the individuals have good information about how other people are acting (in other words, in his model, there is no deception and no sabotage).
Herold says his proof shows that the most efficient strategy is to start off by rewarding those who do heed the management's request, and then, once a critical mass of workers are cooperating, punish those who are still refusing. In other words, start with the carrot, then add the stick. As Herold says, "Reward may help you get to a situation where you can establish punishment."
In Praise of Praise
Herold looked at how two groups of people--the "employees" receiving the commands, and the "enforcers" or managers issuing the orders-- responded in three hypothetical situations. In the first situation, the enforcers could use only rewards; in the second they could use only punishment; and in the third, they could use either.
Scenario One: Reward Only. People cooperate if they are rewarded, but this is onerous for managers. Not surprisingly, most people cooperate if they believe they'll receive a reward. But this turns out to be expensive, since the enforcers have to pay for a reward for every single person.
Scenario Two: Punishment Only. People may cooperate if punished, or they may rebel with little consequence. At the beginning of this set-up, all the enforcers punish everyone who doesn't comply. After that, things can go one of two ways. The first variation is that everyone realizes the consequences of not cooperating, so everyone cooperates and very little actual punishment takes place. Sometimes, however, there's a second outcome: Everyone flat-out refuses to cooperate, and the enforcers give up.
Scenario Three: Reward, then Punishment. Everyone cooperates, but with less cost to their managers. In this situation, the enforcers begin by offering rewards. That creates a climate of cooperation without having to reward every single person. Once that's established, the enforcers can stop offering rewards and instead say they're going to punish those who don't get into line.
Herold admits that his findings don't leave a lot of room for executives who can't stand to mete out punishment. Says Herold, "People who are always nice have a hard time in my model."
Do Herold's findings sound feasible--or pie-in-the-sky?
Image courtesy flickr user opensourceway, CC 2.0