Tag Archives: game theory

Why Collaborate?

Given a choice, most of us would choose to work somewhere where our colleagues offered support whenever they could and actively shared information they thought would help us. Perhaps some of us prefer the idea of our colleagues helping us but not having to help in return. Our choice is to collaborate or compete. It seems like corporate capitalism favours ruthless competition and short-term self-interest. But psychology suggests competing or trying to cheat the collaboration system does not pay off in the long run.

When I think about my own working environments over the years, there has been a wide variation in how collaborative they have been. Project teams are by their nature temporary. Like most project managers, I have had the interesting experience of becoming a part of a great many temporary teams in various organisations.  I realise, looking back, that I take my cues about how much I collaborate from the group I am joining. This approach is explained by the psychological theory around collaboration. Teams seem to find a level of collaboration. This is part of what is referred to as a Nash Equilibrium (after him out of A Beautiful Mind). We make decisions about how open and giving to be, not in isolation, but keenly aware of how giving our colleagues are likely to be. In competitive environments, team members act as if someone will end up on top and the others will be the losers. This relates to the zero sum game in game theory.

Psychologists and game theorists sometimes talk about collaboration as reciprocal altruism and a famous game in this area is The Prisoner’s Dilemma (Flood and Dresher 1950). The premise is that two prisoners are being questioned separately by police. If they both stay quiet they both get off on lesser charges. If prisoner A gives evidence against prisoner B, prisoner A gets off scot free and prisoner B gets a heavy sentence. If they both give evidence on each other, they both get the heavy sentence.

Analysis of The Prisoner’s Dilemma reveals a lot about collaboration. Mathematically, we are best off over time if everyone collaborates with each other (Axelrod and Hamilton 1981 and Worker and Reader 2004). Collaboration here means both prisoners collaborate with each other (i.e. both stay quiet) not that they collaborate with the police. This dynamic reminds me of being on a see-saw. If we jump off without warning the other person, we get to decide and control our descent but the other person gets to come down with a rude bump. It takes coordination for both of you to come down slowly enough to be comfortable (and nobody be the loser with a sore rump). This is by far the least stressful way to descend from a see-saw, as I’m sure you will agree.

So it seems that true collaboration, not cheating the system, is in our long term interest and is least stressful in the short term. I’ll talk below about how this relates to my experience of project management specifically. In general, it is a phrase frequently encountered on the web but if you Google “project management collaboration”, you get a bunch of software tools rather than theory examples. I guess the tools are easier to monetise than the theory.

Perhaps the benefits of collaboration in project management are so universally accepted as to be invisible. Examples come to mind easily. In a programme setting, open and proactive communication makes it easier to manage dependencies and risks. Projects benefit from sharing lessons learnt and exchanging intelligence on common risks and issues. Examples of a lack of collaboration in projects are similarly easy to call to mind. In competitive teams, members take on as little as possible and avoid supporting others, while maximising the appearance of their contribution. Likewise, if team members are competing then sharing intelligence about risks and issues lessens the chance of their project looking good relative to others.

Considering how something like the Nash Equilibrium might relate to my experience of project teams, I’d say the collaborative dynamic is surprisingly binary and sensitive to competition. As soon as one team member competes to do the minimum work for the maximum recognition, everyone else has two choices: either continue to collaborate and feel a mug, or compete.

The logical conclusion from my experiences of project teams chimes with the game theory concepts above. Collaborative teams are the most efficient and pleasant kind to work in, and if you want to work in a collaborative team, everyone needs to collaborate.

Of course, people don’t choose to either collaborate or compete based purely on reason. There is a nice story about a turtle and a scorpion that illustrates non-cooperation in the face of self-interest.

The references above to The Prisoner’s Dilemma come from the evolutionary psychology section of my Open University textbook. There are new books being written about this topic all the time and it will be interesting to see how much of this filters through to the workplace.

The above has been a brief look at collaboration in project teams. I have found it particularly hard to cover such a large topic in this short post. I have discussed choice of approach (collaborate or not) but I have not talked about the detail of how we collaborate. I touched on this in my post on how project teams talk. Mercer’s exploratory talk is a good example of trying to add light to a debate, rather than just the heat added by disputational talk.