Opening Pandora’s box: using city-scale risk models to quantify the global cost of catastrophes, and the value of resilience

posted in: Viewpoints | 1

In the Cambridge Centre for Risk Studies we have developed a framework to quantify the potential damage from a Pandora’s box of all ills – a comprehensive taxonomy of catastrophes. The methodology and outputs speak directly to the value of investing in resilience.

The analysis has been made possible by novel ‘Catastronomics’ techniques to analyse the impacts of catastrophes on economic output. These techniques provide a new metric, ‘GDP@Risk’: the loss to Gross Domestic Product that a catastrophe causes to an economic system.

We have applied GDP@Risk to understand the impact of over 20 threat types on 300 World Cities selected for their economic importance. The proposal, more generally, is that it is possible to estimate the cost to a business, city, region or the global economy, from all catastrophic shocks. And hence to understand the economic value of greater resilience (or lower vulnerability) of such systems to shocks. Holistic approaches like this are an antidote to risk management that reacts to threats taken from yesterday’s news headlines.

The outputs described below come from recent work at the Centre for Risk Studies that will be presented as the Lloyd’s City Risk Index, launch date 3 September 2015 at, and the Cambridge World Cities Risk Index at

Threat types and city characteristics

Our approach is to use a taxonomy of threats from the Cambridge Risk Framework and model the impacts of those threats in a simple, probabilistic way. This view addresses very large and very unlikely shocks: events whose scale makes them unpredictable in an operational sense. An assessment of a threat may account for its size, how it impacts on the system under study and the speed of recovery of that system, which we associate with resilience.

Although our analysis is unique with respect to the breadth of catastrophes considered, see below, it owes much to the principles of catastrophe modelling as reviewed in Gordon Woo’s essential text “Calculating Catastrophe”.

The Cambridge Risk Framework groups catastrophes into four broad threat classes: natural catastrophes; financial markets, trade and business; technology and space; and health and environment.

The 300 World Cities have been selected for their economic significance; they account for around 50 per cent of global GDP output today. Because of their different locations, cities have unique exposure profiles with varied likelihoods of experiencing a given type of threat. We characterise each of the cities by their

  • economic mix
  • population over time
  • quality of construction or the vulnerability of physical assets
  • an index of economic resilience.

Each catastrophe type is modelled probabilistically using three different threat scenarios with different likelihoods and magnitudes. For each city and threat type, the city characteristics and threat scenarios are combined to estimate the impact of a catastrophe in terms of GDP lost, on average, over a 5 year period following the event. We measure the total damage, GDP@Risk, by accumulating this loss per city per threat over all 300 cities and 23 threat types. GDP@Risk is an average of the economic losses caused by the different catastrophes that might occur in an average decade, baselined against economic performance between 2015 and 2025.

GDP@Risk of $5.4 trillion for World Cities

Our analysis estimates a global GDP@Risk of $5.4 trillion. This amounts to 1.4 per cent of the total GDP forecast to be generated by these cities in the next decade, some $370 trillion.

We found that just the top six perils together would generate more than 60% of GDP@Risk across the 300 World Cities. They are financial crisis, interstate war, human pandemic, and the natural catastrophe triad of windstorm, earthquake and flood.

Conversely, we ask what value might be gained globally by increasing economic resilience of all 300 World Cities. Our simple methodology suggests that between 10% and 25% of GDP@Risk could be recovered, in principle, by improving the resilience of cities.


The Cambridge World Cities Risk Index, based on GDP@Risk as an economic loss metric, is perhaps the only comprehensive assessment of perils that are inherent in our environment. The link between improving a city’s resilience and reducing its GDP@Risk is an important step towards putting a value on its resilience.

Daniel Ralph

Daniel Ralph

Danny Ralph is a Founder and Director of the Centre for Risk Studies, as well as Professor of Operations Research at Cambridge Judge Business School, and a Fellow of Churchill College. He is Editor-in-Chief of Mathematical Programming (Series B).

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.