Through our experience in analysing and forecasting emerging risks, we have recognized a need for a clear and accurate methodology. To address this, we are introducing our current forecasting framework, the Global Risk Tracker. This model builds on the works of Marko Papic (Geopolitical Alpha, 2020), Peter Turchin (Historical Dynamics, 2003), and other leading scholars in the field of geopolitical risk. Our approach focuses on analysing policymakers' actions within a historical context, validating the validity and diagnosticity of available data, and considering the broader context in which policy decisions occur.
The framework is supported by three key pillars, each guided by specific questions:
By asking ourselves these questions, we can more accurately predict the likelihood of events that may impact on our clients’ people, assets or reputation.
Examples highlighting the importance of these pillars when considering developing real-world scenarios:
I: Future Actions are Influenced by Past Behaviours
The first pillar is key to determining a forecast because past actions are good indicators of future events. For instance, Ukrainians often reference 2014 as to when Russian troops annexed Crimea and intervened in the Donbas, as the true beginning of their conflict with Russia. Ignoring these events would have led to incorrect predictions about Russian aggression in 2022.
II: Facing Facts
Our emphasis on the broader context helps narrow down potential outcomes by recognizing constraints and deterrents that policymakers cannot dismiss or ignore. Despite recent rhetoric about de-dollarization from emerging nations, the reality is that most of these countries still hold significant US dollar reserves, and their leaders maintain substantial funds in offshore accounts.
III: Quality, Not Quantity of Information
Lastly, we scrutinise available data to measure the potential knock-on effects of a policy action being taken and its real-world implications. Not all data is created equal, and more information does not always produce better forecasting results. While the White House and Number 10 were preparing to invade Iraq in 2003, analysis of military stock and their potential consequences was plentiful, but flawed which led to significant consequences.
A Mathematical Approach
Our approach to assessing the likelihood of major decisions and scenarios incorporates these three pillars.
P(X,Y) represents the probability that a decision-maker (r) chooses to take action X, leading to an associated outcome (Y).
The formula incorporates:
pr: The expected likelihood that a decision-maker (r) will take action ( X ) based on their desire for outcome ( Y ).
z(r,i): The impact of a given deterrent on the decision-maker's choice.
u(r,i) : The impact of a given constraint on the decision-maker's ability to achieve outcome ( Y ) by taking action ( X ).
Whenever we assign a numerical value to the likelihood of a policymaker’s decision or an outcome’s being realised, we have derived the value using the this formula.
