This post is based on the key messages in relation to collective decision-making in the TAI research paper on “Better decision-making: a toolkit”.
Institutional investing is increasingly a team activity. We make decisions together, not alone. For example, in 2010, more than 70% of all US domestic equity mutual funds were managed by teams of portfolio managers compared to only 30% in 1992[1]. In the asset owner space, investment committees are responsible for pretty much all the highest-impact decisions of a fund.
Against this backdrop, the concept of collective, instead of individual, decision-making is more interesting and more relevant to our industry. What makes some groups more effective in making decisions than others?
It turns out that this is an area where both management scientists and practitioners largely agree. A wise team tends to have these four traits:
-
Members of the team have a wide range of perspectives and opinions inputting to decision-making
-
The team is very effective in sharing and processing the relevant information
-
Independence of individual judgement is preserved during the discussion
-
There is an effective means of combining individual opinions.
Like any other skills, we can learn to be more effective in making collective decisions. And that starts with stronger practice in each of these four areas.
Diverse opinions
The fundamental source of diverse opinions is cognitive diversity, a subject our industry is increasingly paying attention to. In this post, I will explore another challenge. Assuming that your team already enjoys a healthy level of cognitive diversity, does it mean you will automatically get diverse opinions? That answer is not always.
For example, your team might make decisions based on the HiPPO principle – where decisions ultimately rest with the highest paid person’s opinions. This is an environment where diverse opinions tend to be suppressed. Why risk disagreeing with your boss when you know that, at the end of the day, your opinions don’t really matter?
The key here is to create a psychological safe space that allows diverse opinions to emerge freely. Creating the right team culture is the first and most important step. Leaders need to set the right tone and lead by examples – ie, encouraging rather than suppressing views to challenge the HiPPOs. There are specific techniques such as pre-mortem or devil’s advocate to deliberately empower the voice of disagreement.
Information sharing
A key challenge in relation to information sharing is that it is an interactive process: the information you share will influence and be influenced by the information shared by others. Sometimes these influences are unhelpful. For example you might not feel completely comfortable sharing a piece of information that challenges the consensus, or worse, a point your boss just raised.
The solution is a surprisingly simple one. Why not share the information before the discussion? Use emails, use on-line tools. Share the information you want to share before it gets messy when the social interaction begins. And don’t just create a deck of slides with data and charts. Narratives are scientifically proven to be more effective in making an argument that resonates with your readers.
Independence
Committees like consensus-building. We prefer to get everyone on board for collective decisions. Granted, if the consensus is authentic, it creates strong collective buy-in to the decisions and follow-up actions.
The problem is that, in practice, consensus is a pretty high hurdle. Often, in order to achieve it, either the group itself is too homogenous to start with or opposing views are suppressed. Independence of individual judgements is therefore in jeopardy.
Again, there is a rather simple solution. Let’s vote. An informal show of hands, however, doesn’t really count because it can still create the similar social tension as consensus building. Modern technology makes anonymous voting very straightforward. So just get on with it.
For really important decisions, a case can even be made to have two rounds of voting: one before the deliberation process and one afterwards. Pre-voting helps collect individual judgements, before the helpful and unhelpful influence of social interaction. Second round gives team members an opportunity to update their opinions.
Combining individual views
So your team took a vote: three voted yes and two voted no. Where does it leave you in terms of the final decision? In practice there is a rich level of nuances in this final step of decision-making.
The team can simply take the result of voting at its face value. There are more yes than no so the team collectively supports the motion.
But maybe the two members of team who voted no happened to be the subject matter experts for this specific decision and their votes were pre-agreed to be worth twice as much as the other votes. The collective decision is therefore now against the motion. We are into the realm of reliability-weighted voting here, a theoretically sound concept that individual opinions should be weighted in accordance with their reliability. It is a very difficult idea to implement in practice though.
Or maybe your members were also asked to express their confidence level alongside their votes. If the minority votes carry very high confidence levels individually, they might flip the collective decision.
Another common practice is that the voting result not the final decision but instead is used as an input, albeit a very important one, to the team leader to ultimately decide. This might sound HiPPO like but in practice there are legitimate arguments to support this practice, in some circumstances. It promotes clearer accountability. It is less complex and easier to communicate, especially externally. It also addresses the issue of uninformed votes.
This approach would require checks and balances on the team leader, of course. For example, one group I have worked with gives the team leader discretion to make the final decision, unless there is a supermajority (eg, 75%). Others favour comply-or-explain: the team leader remains accountable, but is subject to a formal justification process if a supermajority is vetoed.
There isn’t really a single best practice in this area – what works best for you is highly context dependent. Make sure you have good reasons to support your choice of practice. And trying to force a consensus where one does not naturally exist is a bad practice you should always avoid.
How a group thinks and decides is an emergent property of individual thought processes, communication patterns, dependencies, relationships and other aspects of interactions among group members. Effective decision-making emerges when the whole is greater than the sum of the parts. It is also a skill a team can learn and develop.