Adapt Research Ltd has put global catastrophic risk back on the agenda with a review that says resilience research and policy planning still leave major gaps. The paper starts with 3,200 candidate studies and ends with 152, but the bigger problem is the evidence those papers do not yet provide.
That matters because policymakers in Wellington, Reykjavík and Montevideo are still being asked to judge exposure and choose resilience measures without strong guidance on which options actually hold up. The review says the field has grown fast over the last decade, yet the evidence base still falls short of the decisions governments need to make.
OpenAlex to 152 papers
The review began with a systematic OpenAlex scan that returned 3,200 candidate papers. It was screened down to 551, then to 147 read in full. The authors added 77 expert-identified papers to fill the gaps, producing a final corpus of 152 papers.
That sorting process is part of the story. The field is large enough to generate volume, but not yet clean enough to answer every policy question with the same level of confidence. The review organizes the evidence around three harm pathways rather than a list of hazards, which lets it compare how different threats collapse into a small number of system failures.
Cropping and dark grids
Nuclear winter and a Tambora-scale eruption both end in a crop-yield problem. Cyberattack, HEMP and a severe geomagnetic storm all end in a dark grid. The review uses that structure to show that the resilience problem is less about the label on the hazard and more about the failure mode it triggers.
During Covid-19, islands recorded 64.8 excess deaths per 100,000 compared with 194.3 for non-islands. Jurisdictions running explicit exclusion and elimination strategies recorded -2.1 excess deaths per 100,000, while others recorded 166.5. Those figures give the paper its clearest practical edge: some resilience choices tracked with better outcomes, but not every favorable pattern came from the same kind of place or policy.
Folke and Walker in policy
Folke and Walker in the review frame the theoretical point: isolation, self-sufficiency, governance quality and decentralisation show up often as resilience factors. Democracy and lower inequality also predicted better health and sometimes macroeconomic outcomes. Citizens’ assemblies are presented as one mechanism for making resilience policy legitimate rather than merely technical.
But the authors stop short of turning those signals into a single score. The resilience factors are not necessarily additive, and the evidence is not strong enough to build a composite GCR resilience ranking that would settle trade-offs among them. A country trying to prepare for the next shock has to weigh whether more isolation beats more governance quality, or whether decentralisation matters more than self-sufficiency, one choice at a time.
That leaves the immediate task for policymakers in Wellington, Reykjavík and Montevideo clear enough. They do not need another abstract hazard list; they need a way to see the evidence, deliberate over the trade-offs and choose the mix of resilience investments that fits their own exposure.
