Reasonable Worst-Case Scenarios

Reasonable Worst-Case Scenarios

We’re hearing a lot from Government about “reasonable worst-case scenarios” at the moment. At first sight it seems sensible. We need to think beyond the “base case” or “central case”, but no-one wants to bother with the most extreme case. But what is the methodology behind it? Or is it just a subjective assessment?

According to Government documents referring to Operation Yellowhammer, the Brexit planning process, a reasonable worst-case scenario (RWCS) is “plausible” – as opposed to the extreme worst-case scenario. A planning assumption is “The reasonable worst-case scenario you are anticipating, including the likely severity and duration. A reasonable worst-case scenario is the worst plausible, not most likely, manifestation of the risk in question.”  Reasonable worst case planning assumptions “should provide a common, stretching, scenario for stakeholders to plan against and for which, if plans are in place, a reasonable level of preparedness can be expected for most manifestations of the risk. It is not a likely scenario or a prediction but an outline of what could feasibly happen in a reasonable worst case. The assumptions are deliberately stretching and challenging in order to facilitate effective contingency planning.”

Yellowhammer “has always been a reasonable worst case scenario and never a ‘base’ or ‘central’ scenario.”

So by definition, we’re in the realm of qualitative judgements as well as quantitative analysis. The National Risk Assessment is classified, so it’s quite difficult to get at the groundwork. From our experience, one takes facts than can be quantified (number of nuclear power stations) and then adds qualitative judgements (likelihood of any nuclear power station failing catastrophically in any 100 year period) and then tries to place a quantitative value of the qualitative judgement. So there is always a degree of subjectivity involved in the process, even if it’s disguised by the fact that this always comes out in tables and charts and numbers.

The National Risk Register – which is the unclassified version of the NRA – has a methodology section (2017 version, section 4, p69ff) which is gloriously light. But there is some detail. For instance:

  1. Each of the risks in the NRA is described as a ‘reasonable worst case scenario’. For a risk to be included in the NRA, they must (note that we have taken the wording exactly from the National Risk Register):
fulfil the definition of a civil emergency, as described on page 6. Note that long-term trends (such as climate change) increase the chance or severity of civil emergencies (e.g. floods), but do not constitute civil emergencies in themselves; legal definition
have at least a 1 in 20,000 chance of occurring in the UK in the next five years, or in the case of malicious attacks; have a plausibility score of “1” or more (see section on plausibility below); plausibility, therefore subjective
have an expected impact that reaches a minimum threshold (typically significant damage to human welfare in the UK). minimum threshold – qualitative AND quantitative
  1. For civil emergencies:
experts assign likelihood scores to each risk on a scale of one to five. likelihood score – qualitative AND quantitative
For each step on this scale, the probability of an event happening in the next five years increases roughly tenfold. For some risks,
data such as historical analysis and numeric modelling can be used to inform estimates of likelihood (especially for naturally and accidentally occurring hazards).
Scientific expertise is also sought to inform the development and review of risks. qualitative
Where possible, a combination of this analysis and expert judgement is used to estimate the approximate likelihood of an event occurring. qualitative AND quantitative
  1. The plausibility of terrorist attacks or other malicious incidents is assessed slightly differently.
The willingness of individuals or groups to carry out attacks is balanced against an objective assessment of their capability – now and, as far as possible, over the next five years – and the vulnerability of their potential targets. The two scales are not directly comparable with one another; however, for the purposes of planning, a hazard or threat in the top right quadrant of either matrix would be given the same priority. qualitative AND quantitative

You’ll see that we have tried to assess what type of inputs go into the assessment. And it’s clear that much of it, despite the numbers and percentages that come out of the end, are qualitative judgements transferred into, apparently definitive, figures.

The introduction of the word “plausible” in the definition of the RWCS is ultimately the point here. Even if all of the data was factual, and based entirely on provable numbers, the question whether it is “plausible” by definition introduces a level of subjectivity. Fukushima, for instance, was possible, but not plausible. Plausibility does not necessarily include wild cards. But it’s still not clear whether “plausible” implies 20%, 10% or 5% likelihood (even if that could be estimated).

We would also make the point that the RWCS is not a “scenario” at all in the futurist’s sense. It seems to be much more a “moderate downside” case, compared with a “base case” or “central case”; it has the feel of being a one-dimensional view of the world. Useful scenarios are much more multi-facetted, and allow for the interaction of several different disruptive (or serendipitous) events – pandemic AND “no-deal” Brexit AND an extreme weather event.

Quite possibly in reality the RWCS is a proper scenario, but we suspect what is happening here is that the expression itself is used as a convenient, flexible shorthand to communicate the idea of the base case to the general public.  What matters then is the quality of the contingency planning, and preparedness, which itself can be coloured by the cost of being prepared.

You’d expect us to promote the idea of proper, communicated, scenarios. We believe these have real value – in this case, across the whole of government and with the wider stakeholder group. Particularly, it is our experience that understanding the parameters of a scenario in detail substantially aid its usefulness. If people understand what went into it, they are more capable of making their own judgements on how to use the scenario.

As we enter the new year, with Covid-19 still in place, and Brexit finally fully biting, reasonable worst-cases are going to be critical throughout the public and private sectors. Understanding where they come from, what they are, and how to use them, will be key.

Written by Huw Williams, SAMI Principal and Jonathan Blanchard Smith, SAMI Fellow and Director

The views expressed are those of the author(s) and not necessarily of SAMI Consulting.

SAMI Consulting was founded in 1989 by Shell and St Andrews University. They have undertaken scenario planning projects for a wide range of UK and international organisations. Their core skill is providing the link between futures research and strategy.

If you enjoyed this blog from SAMI Consulting, the home of scenario planning, please sign up for our monthly newsletter at and/or browse our website at

Image by Gerd Altmann from Pixabay

One Comment
  1. Well done and thank you both guys for unpacking that …. It’s shows very clearly how we the human race are disappearing at a quantum rate of knots up our own bureaucratic arse 🤣

Leave a reply

Your email address will not be published. Required fields are marked *