A model in a time of ‘radical uncertainty’

Posted on April 29, 2020 by Pete Faulkner - Cairns, GDP, QLD, Townsville, Uncategorized

I spent part of this morning listening to Prof John Kay, author of Radical Uncertainty: Decision making for an unknowable future, talking to my friend Gene Tunny on his Economics Explained podcast. It’s a fascinating chat about a highly topical subject (although the book was written prior to the COVID-19 pandemic). I found the discussion on the usefulness of economic models especially interesting given that I’ve spent the last month or so developing just such a model to try and better understand the impacts of COVID-19 on our economies, and particularly our regional economies.

Prof Kay made a really important point in the interview (which I’m sure is expanded on in the book…although I am yet to read it. It was downloaded onto my Kindle this morning!) which I think is worth quoting…

“In order to talk about a probability in the world, you have to multiply the probability derived by the model by the probability that the model is true. But you don’t know what the probability that the model is true is, you just know that it’s really very low. It’s a misunderstanding of the way in which you use models, I’m still very much in favour of using models, but you use models not to make statements about the real world, but to get insights into the world.”

Prof Kay later talks about ensuring systems are ‘robust and resilient to things you are not going to be able to predict’ which put me in mind of the concept of anti-fragility (a step up from simple robustness) as described by Nassim Taleb in his excellent book Antifragile: Things That Gain From Disorder.

Bearing Prof Kay’s point about models providing insights rather than predictions, here are a series of ‘insights ‘ from various market participants about the possible impact of COVID-19 on the Australian economy this year.

	GDP 2020 %	Unemp June %
IMF	-6.7	na
UBS	-6.1	10.5
Westpac	-5.0	9.0
Treasury	-6.0	10.0
ANZ	-5.5	9.5
NAB	-4.4	11.7
JP Morgan	-6.9	11.4
AMP	na	9.5
Bloomberg	-6.0	na
HSBC	na	8.6
Average	-5.8	10.0

Our own model, see below, is somewhat more optimistic than the average on the scale of the GDP decline in 2020 (-5.0%) but rather more pessimistic when it comes to unemployment, we see it reaching 11.9% by mid-year. However, what is clear is that we are almost certainly about to witness an economic contraction unlike anything the country has seen for many decades.

The most important question yet to be answered with any degree of clarity is how quickly we come out of it. Much talk in economic circles these days is about a so-called Nike Swoosh recovery rather than the more conventional V, U or L shaped recoveries.

What is also apparent is that regional economies more reliant on those sectors which are being most heavily impacted by the pandemic and subsequent lock-downs (tourism, accommodation, food service, retail etc.) are set to fare worse than those more reliant on sectors which are less impacted (agriculture, health, utilities, mining etc). As a result our model shows regions such as Gold Coast, Cairns and the Sunshine Coast performing far worse (and for longer) than regions such as Toowoomba, Cassowary Coast and Ipswich.

We shall be updating our model as new economic, health and policy data becomes available…with Prof Kay’s words ringing in our ears…”use models not to make statements about the real world, but to get insights into the world.”!