Running in Herds

In a nutshell: 
- People tend to place more value on others’ information than on their own, resulting in herd behavior.
- Herd behavior tends to result in worse outcomes than random choice and nearly never results in the best option.
- We don’t only retreat to others when we are uncertain, but also for problems that we can be sure to solve correctly.
- The effect is even worse, when the people deciding before us are in a higher hierarchical position.
- Conclusion: Use your information & know your influence!

My mum grew up on a farm and her favorite animals are pigs. Now, there may be an according bias in her stories, but it doesn’t really matter for the sake of this one. She once told me that she found it hilarious, how gullible cows can be. You put them in front of a locked gate, she says, and they will stand in front of it until you show them to go somewhere else. Even if there is an open gate a bit further down, they will not move from their favorite locked exit. Pigs on the other hand! Pigs will understand that life is not limited to one gate only, and they will find their way outside if that’s where they want to go. Such crafty animals.

Us humans can be quite cow-ish in our behavior. We too often follow the option that has been presented to us. In fact, we’re even likely to discard our own information in favor of someone else’s — even if we don’t know whether the quality of their information supersedes ours. We like to accept the closed gate and just assume the cow who lead us there knew what he/she was doing.

Banerjee (1992) calls this herd behavior. In essence, it means that if we are not among the first one or two people in line to voice our opinion, we make decisions not based on our own signals (information), but on the decisions others have made before us:

“If an agent has a signal, she follows that signal, unless someone before her has already followed someone else. Then she follows suit.” — Banerjee, 1992; emphasis added

Ergo, as soon as there is more than one other person who votes for an alternative option, we will discard our own choice and go with the flow.

This is due to two factors:

1. We assume that the other person holds more information than we do. If we have no information at all, this may be logical. However, if we hold information that clashes with the decision other people prefer, we believe that we have an information deficit and will join the mainstream decision.
2. Joining “the herd” seems beneficial for social reasons (less conflict, no solitude, etc.).

Now, of course it may be true that the (minimum) two people before you had better information, thus made the right choice and it was smart to follow. However, chances are minimal that this is the case. Herd behavior will almost certainly lead to an outcome that is neither optimal — nor, for that matter, better than the default!

If I, as a regular cow, see two others standing in front of the door, the default option of staying in the barn and using the other, open gate would be more beneficial than standing in front of the locked door. But silly me will stand with them, because, you know, peers rule! (I can already hear all the mums scolding: well, if all your friends were to jump off a bridge…!)

It’s not just peers, though. It gets much worse as soon as hierarchy is involved.

Lynne Zucker (1977), in an amazing and simple experiment, showed that people will follow almost everything their “superiors” will say. In a purely experimental setting in which strangers were given a fake hierarchy she showed that even the mere notion of authority made people question what they saw with their own eyes and parrot what their “boss” said they saw — even if the option presented by the boss was way off their own experience.

Lynne Zucker’s (1977) experiment:
The test subject was brought into a completely dark room with a tiny light on one side. They were supposed to tell the distance the light traveled. That was a trick question: the light didn’t move at all — but that’s really hard to tell in a completely dark chamber. Technically, the correct answer would have been 0 cm.
Round 1 (control group): The test subjects did the experiment alone and their numbers were reported. The numbers varied randomly.
Round 2: There was an actor in the chamber before the test subject went in, who told the test subject their estimate. The test subject then went in and estimated themselves. The numbers were again varying randomly. The influence of the stranger on the test subjects' answers was neglectable.
Round 3: The test subjects were told that they were in a hypothetical firm and the actor before them was their hypothetical boss. This "boss" then told the test subjects their estimate — which was quite off the charts. When the test subjects reported their numbers, they almost perfectly matched the numbers their "boss" had told them before — although the control group's numbers showed that under regular conditions, the numbers were quite different. Remember, this is a hypothetical firm, with no career consequences whatsoever. 

Hence, if the alpha-cow is voting for an option that is completely different from what your information is telling you to do, chances are that we would all just happily deny that we can clearly see the open gate a bit further along.

What does this tell us?

First, that we need to be more like pigs — at least more like the pigs my mum so admirably described: Curious.

This may be common sense, but I think we can all use a reminder every now and then: Just because someone favors a different option, doesn’t mean your information isn’t valid. Also, it doesn’t mean the other person actually knows. I plea for open exchange of information and ideas, so that people can understand their own and others’ information’s quality and value.

Pigs — and after this fun fact I will stop with the farm references, I promise — actually get slightly annoyed, if you don’t let them roam and explore a bit. Why not do the same? It’s a known management trick to quickly present loads of “facts” that no-one can follow. People will shy away from asking clearing questions and as the person sounded like they knew what they were talking about, it must be at least sort of true. And questions are not always welcome (remember some teachers? *rolls with eyes*). However, I think it’s altogether rude to not welcome questions. It’s a stretch to assume others see the point you’re making as clearly as you are yourself — so whenever there are no questions encouraged, that makes me think: Have we all used our information? Or are we just dumb-struck by some overwhelming presentation?

We recently had a pivot in strategy in our own firm. Being a rather small flock ourselves, strategic decisions are of concern and upmost interest to all of us and are a collective governance decision. The pivot was initiated by two team members — and they presented their concerns and hopes in logical order. But just because their information seemed legit, we still took the time to reflect — for entire two weeks! So we went out and each of us thought about the proposal meticulously. Each of us did their own research, went to their own emotional “cores” and talked to other team members individually. No matter the result, the important part was that it was absolutely encouraged and welcome by the two who proposed the pivot. They, too, needed to know that the others made sure if they agreed and if not, to understand how to adjust.

Of course, two weeks are a long time, but it doesn’t always have to be that long. And I don’t want to advocate for cycles of proposals meetings and change meeting — this would probably waste everyone’s time. However, we know that time pressure clouds our decision making.

Another, quite simple ways of avoiding herd behavior is simply making sure everyone is herd heard. Woolley and colleagues (2010) found that collective intelligence relies on balanced participation. Together we are only better if we give each team member a platform to actually participate in a discussion. There are different ways for doing this:

• Do a round in which each member gets to give their opinion on a topic individually;
• have everyone write down their information before the discussion to avoid being sucked into the whirl of conformity — get creative!

NB: this also ups the chances you will actually benefit from the diversity in your team. Simply hiring a woman or a person of color will likely not help anyone, if their input is silenced.

If you’re reading this from a formal leadership position, think also how much trust you will gain through such encouragement. Many people in “power positions” like to take decisions on their own — heck, I sure do whenever I have the chance to make others do stuff the way I think is right. Bold of us to assume we know better. In fact, since all the people around you have made it this far, wouldn’t it be reckless to ignore their information base?

Secondly, we all know the office grapevine: someone always thinks they know better — but they didn’t say so. Well, if they had the chance, then the “villain attribute” is on them if they knowingly chose to withhold their information.

But — here come’s point number two: We all need to understand our power and leadership potential, if formally attested or not.

This is two-fold and I don’t weigh the two parts at all, we just need to know them both:

• If you voice your opinion and others simply trott along, chances are you’ve just created a herd and should double-check why they’re following you and if they have used their information.
• If you want to make a change — understand that you really only need to convince one other person to start a movement! And then do. We hear it time and time again, but as long as we’re discussing options, the decision has not been made. As soon as we do, others can follow. (But that, we can discuss the “just do it” part of business another time. I think you’ve got the point.)

Honestly, it’s hard on both ends sometimes and herd behavior is a nice little shortcut our brain developed to have an easy way out. If you’re trying to convince someone and don’t sincerely inquire other points of view, it’ll be easier and faster — but potentially wrong. If you’re presented with an idea, even if it contradicts with your information, going along will be the path of less resistance — but potentially wrong. No matter which side you’re on, you can, however, with a bit more effort, make the better decision, just be like the pigs*!

• Sorry, had to use it one more time

Want to discuss more?

My name is Mary-Jane and I’m a partner at 1789 Innovations. We believe that having a true impact in times of uncertainty happens through people. Our goal is to empower each other to become creators of change to go beyond the revolution — together. To do so, we partner up with firms to create a novel way of doing business and share everything we learn online and on events. For more information about 1789 visit our website 1789Innovations.com.