5.5 Disagreeing must be done efficiently.
Working oneself through disagreements can be time-consuming, so you can imaging how an idea meritocracy—where disagreement is not just tolerated but encouraged—could become dysfunctional if it’s not managed well. Imagine how inefficient it would be if a teacher ran a large class by asking each of the students individually what they thought, and then debated with all of them, instead of conveying their own views first and taking questions later.
People who want to disagree must keep this in mind and follow the tools and protocols for disagreeing well.
a. Know when to stop debating and move on to agreeing about what should be done.
I have seen people who agree on the major issues waste hours arguing over details. It’s more important to do big things well than to do the small things perfectly. But when people disagree on the importance of debating something, it probably should be debated. Operating otherwise would essentially give someone (typically the boss) a de facto veto.
b. Use believability weighting as a tool rather than a substitute for decision making by Responsible Parties.
Believability-weighted decision making is a way of supplementing and challenging the decisions of Responsible Parties, not overruling them. As Bridgewater‘s system currently exists, everyone is allowed to give input, but their believability is weighted based on the evidence (their track records, test results, and other data). Responsible Parties can overrule believability-weighted voting but only at their peril. When a decision maker chooses to bet on his own opinion over the consensus of believable others, he is making a bold statement that will be proven right or wrong by the results.
c. Since you don’t have the time to throughly examine everyone’s thinking yourself, choose your believable people wisely.
Generally speaking, it’s best to choose three believable people who care a lot about achieving the best outcome and who are willing to openly disagree with each other and have their reasoning probed. Of course the number three isn’t set in stone; the group could be larger or smaller. Its ideal size depends on the amount of time available, how important the decision is, how objectively you can assess your own and others’ decision-making abilities, and how important it is to have a lot of people understand the reasoning behind the decision.
d. When you’re responsible for a decision, compare the believability-weighted decision making of the crowd to what you believe.
When they’re at odds, you should work hard to resolve the disagreement.
If you are about to make a decision that the believability-weighted consensus thinks is wrong, think very carefully before you proceed. It’s likely that you’re wrong, but even if you’re right, there’s a good chance that you’ll lose respect by overruling the process. You should try hard to get in sync, and if you still can’t do that, you should be able to put your finger on exactly what it is you disagree with, understand the risks of being wrong, and clearly explain your reasons and logic to others. If you can’t do those things, you probably should suspend your own judgment and go with the believablity-weighted vote.
* Source: Principles by Ray Dalio