Last week the Institute of Risk Management North West regional group held a meeting on Adapting to the Global Risk Landscape.  The intention was to talk about some of the most serious long term risks and what we should be doing about them.  There’s a write up of the meeting appearing shortly on the IRM website, but this post is about something we didn’t talk about.

The first talk was by David Forster of Zurich Municipal and it led us through the World Economic Forum annual risk report for 2013.  As David was unable to attend it fell to me to present it.  The WEF document is a well-researched document, now in its eighth year, which seeks to provide the global risk profile.  Based on the picture it painted we had follow-up presentations on (a) what politicians, who stand in the front line in responding to these risks, should do, (b) what risk managers should do and (c) how continuity people can help, given that the WEF recommended response to the global risks is ‘resilience’.  The relationship between resilience and risk management is something we’ll return to another day.

The question for now is whether we should accept what is in the WEF risk report or not, whether we should follow its calls to action.  What evidence is there it it’s right?

The main technique used by the WEF is a poll of 1,000 experts from across the world who first of all are asked to rate 50 ‘risks’ which have been prepared earlier for likelihood and impact.

The first criticism is that the risks are not.  They are vague statements such as “severe income disparity – widening gaps between the richest and poorest citizens”, or “rising greenhouse gas emissions – governments, business and consumers fail to reduce greenhouse gas emissions and expand carbon sinks”.  In both cases these are issues which are with us.  Another is “global governance failure – weak or inadequate global institutions, agreements or networks, combined with competing national and political interests, impede attempts to cooperate on addressing global risks”, so it represents failure of a control measure for other risks.  Yet another, ranked low, is orbital debris.  The WEF points out that actually it is quite important as it can knock out the satellites we depend on for global commerce or monitoring climate change: again it’s a risk to a number of control measures.

Given this, it’s logically impossible to assess their likelihood (over the next 10 years) and impact.  The question doesn’t make sense.  And it particularly doesn’t make sense when you see that you’re asked to rate both quantities on a scale of 1 to 5 with no further guidance given on what these scales mean.  It’s a bit of a giveaway that the aggregated results (averages) show most of the risks between 3 and 4 on both scales so the presentation has a very restricted view of the 5×5 matrix.  What’s an even bigger giveaway is that the WEF publishes the distribution of the results for each individual risk.  What you can see is that they are distributed along the diagonal axis, ie if they have large likelihood they tend to have large impact.  This confirms what I have often seen: that people think of risk as high, medium and low, not in the 2-dimensional space of accepted wisdom.

I am just re-reading Doug Hubbard’s The Failure of Risk Management – review to come – and Hubbard does a fine job of assembling the evidence that 5×5 risk matrices, scoring schemes and the like are a very poor approach.  He’s not the first though, and the sooner these approaches are widely recognised as discredited, the better.

Another criticism of the WEF approach is that it has always been pretty volatile from year to year.  This is the first year the same list of risk is used, so it is straightforward to make the comparison.  The WEF provides the scatter charts and you can see that the world is getting to be a riskier place.  You also get the impression that the changes are driven by what has been in the news this year.  In that sense the WEF report is more a summary of Daily Mail scare stories (though for some reason falling house prices in the home counties does not feature) than a genuine forward look.

Of course the WEF recognises that a vanilla list of risks does not tell a very compelling story, so each year the Forum aims to extract a small number of narratives to discuss in depth.  The narratives are obtained by linking risks and highlighting different elements of the scatter chart.  For that reason they are known as constellations (nexuses in previous years).  The three constellations this year are again different to previous years, again undermining the whole concept of long term risks.  We seem to see them quite differently from one year to the next.

The way the WEF develops its constellations is to ask its 1,000 respondents to identify 3-10 links between the 50 risks.  It does not investigate what the nature of the links might be.  In fact the next stages are pretty non-transparent as there is a great deal of further analysis which goes on by the risk reporting team.  The report is produced through the collaboration of a number of prestigious organisations and I think it is this that is supposed to persuade us to accept its findings.  At least one of this year’s constellations actually feels pretty minor and there is a view that the constellations are chosen so as to be different from last year’s so that the narrative is more interesting.

Well prestige doesn’t do it for me, but I am willing to accept that a lot of thought has gone into the WEF report, so I don’t want to write it off.  But we should recognise it for what it is, a high class opinion survey with little basis in real risk concepts.  While I wouldn’t want to ignore the calls to action it makes, based on the information it creates, I would also want to sanity check them before committing too much time, effort or money.