1. The Frame War of March 2023
When Silicon Valley Bank failed on 10 March 2023, what followed was not merely a financial event — it was a linguistic battle. Within hours, competing conceptual frames were deployed by regulators, executives, and commentators, each carrying profoundly different implications for what was happening, who was responsible, and what should be done.
Framing theory, rooted in Goffman's original formulation and subsequently developed through cognitive linguistics by Lakoff, Fillmore, and others, holds that the frames used to describe events shape the cognitive schemas through which audiences interpret them. In a financial crisis, where public behaviour is itself a market variable, the stakes of framing are not merely communicative — they are economic.
2. Frame Taxonomy: Six Dominant Frames
Our analysis identified six dominant frames active across the corpus:
CONTAGION positions the crisis as a pathogen spreading through a susceptible system. It activates fear, urgency, and systemic vulnerability. CONTAINMENT positions institutional actors as effective quarantine agents. These two frames often appeared in direct competition within minutes of each other across different institutional channels.
PANIC attributes instability to irrational human behaviour rather than structural factors. PRUDENCE counters by positioning regulatory action as the natural, rational response of responsible stewards.
SYSTEMIC FAILURE invokes deep structural dysfunction, triggering calls for fundamental reform. ISOLATED INCIDENT minimises contagion risk while risking credibility if events prove otherwise.
3. Temporal Frame Dynamics
One of the most significant findings of this study concerns the temporal dynamics of frame adoption. Rather than frames being independently generated by each institution, our data shows strong evidence of frame contagion — the spread of communicative choices across institutional actors.
When one institution deployed CONTAGION framing, the probability of adjacent institutions adopting the same frame within 4 hours was 73%. This represents a form of communicative herding, where institutions — under conditions of high uncertainty and time pressure — anchor to the frames they observe others using, creating self-reinforcing frame cycles.
By contrast, institutions that had pre-prepared frame response protocols showed significantly more frame resilience, maintaining their chosen frames under pressure and showing better downstream credibility metrics.
4. Trust Correlates and Market Implications
Correlating frame choices with YouGov daily trust tracker data revealed clear patterns. Institutions that deployed CONTAINMENT framing within the first six hours of the crisis retained 22% more public trust over the subsequent 30-day period than those that delayed or used CONTAGION framing.
Surprisingly, the use of abstract technical language showed a bifurcated effect: it increased credibility ratings among financially sophisticated audiences while simultaneously decreasing comprehension among general publics. This suggests institutional communicators face a genuine dilemma between specialist credibility and public accessibility — one that our findings suggest can be partially resolved through careful frame layering.
5. Implications for Crisis Communication Protocol
These findings have immediate practical implications for institutional crisis communication planning. First, the speed of frame dynamics in a digital environment means that 'wait and see' communication strategies are no longer viable — institutions must have pre-defined frame protocols ready for activation.
Second, the contagion effect in framing means that early, confident frame deployment by a credible institution can stabilise the broader communicative environment. Central banks, given their perceived authority, have an outsized responsibility for frame setting in financial crises.
Third, the distinction between short-term trust preservation and long-term credibility requires explicit attention. Frames that stabilise confidence in the immediate term may damage credibility if they are later perceived as minimising real risks.