The pursue of being wrong

In an interview with Capital Monitor, Mike Clark, founding director at responsible investment consultancy Ario Advisory said: “climate scenarios need improvements, such as more narrative context – that is, more qualitative information to accompany the quantitative data.” “Narratives eat modelling for breakfast,” he added.

To be clear, I don’t know or never met this person, and I couldn’t find any sign of a website about Mike’s investment company. But as a foresight facilitator, I can nothing but agree with his bold statement. Especially correlations of quantitative data to forecast the impact of climate change are particularly tricky if not impossible to produce. Correction: you can produce anything you want; it will always be precisely wrong. Why is that? Because humans prefer this over being kinda-right. We cling on to the comfortable feeling of exact science. Kinda-right leaves room for uncertainty so we prefer very precise forecasts even if they turn out to be completely wrong. We immediately forget and quickly jump to the next forecast. Nassim Nicholas Taleb described it in his book The Black Swan:

What is surprising is not the magnitude of our forecast errors, but our absence of awareness of it.

?So, global warming rising somewhere between 1,5 and 2,0 degrees Celsius in the next 40 years is not good enough. No, no, we want to know what will happen at 1,8°C. And what we need to do to lower it with 0,3°C. Nonsense of course. No model will be reliable and yes, qualitative methodologies offer alternative approaches to cope with complex and geo-political issues beyond any mathematical model’s capacity. And remember: sh.. in, sh.. out.

Narratives supporting future scenario building don’t have this burden of preciseness. The qualitative methodology originated from the Rand Corporation (read: Pentagon) and was rebooted by the Shell strategic planning group in the late sixties/early seventies. Pierre Wack and his team turned it into an art. Scenario building facilitators, like Esset Engage, prefer to emphasize on the strategic part of the trajectory, the ‘what if’ part which invites people to see possible consequences ‘if’ things will happen. Consequences which need to be reinforced, avoided, or mitigated depending of course on their impact. Narratives can be very significant at that stage. For the simple reason people understand them much better than statistics.

So, will I eat my daily breakfast of models tomorrow (in the statistical sense of course, no pun intended). No, the quantitative world became very powerful the last decades due to access of vast amounts of data and high speed algorithmic engines in the cloud. As in most good things, the answer lays in the middle, looking for synergies between both approaches. Something we, at Esset Engage, will never rule out.

?

Note: If you write down ‘no pun intended’, then for sure, pun was intended :-) Otherwise you would have chosen another phrase without pun. Apologies for the little big lie.