Scenario synthesis links economic narratives to risk distributions
A Federal Reserve working paper presents Scenario Synthesis, a statistical framework that connects narrative scenario analysis with predictive risk distributions. Authors Tobias Adrian, Domenico Giannone, Matteo Luciani, and Mike West apply the method to historical Tealbook projections.
Concordance across macroeconomic models
The framework bridges narrative scenario analysis with predictive densities by treating scenarios as conditional models and fitting them to unconditional reference distributions.
Using an expected misclassification rate concordance criterion regularized by a Dirichlet prior, the method assigns statistical weights to individual scenarios.
Testing the December 2007 Tealbook reveals that staff scenarios underrepresented left-tail risk, allocating a 0.24 weight to the credit crunch scenario against the New York Fed Blackbook density.
In the December 2018 Tealbook, the scenario set successfully spanned the Outlook-at-Risk reference distribution, assigning a 0.48 weight to the baseline and 0.17 to foreign slowdown.
From fan charts to structural mechanisms
Central bank risk communication has divided between structural scenario storytelling and probabilistic fan charts.
Following Ben Bernanke's 2024 review of the Bank of England, several monetary authorities shifted toward narrative paths to replace opaque distributions.
However, standard macroeconomic models such as FRB/US rely on linear approximations that exclude financial accelerators and state-dependent Phillips curves, leaving scenarios vulnerable to missing nonlinear tail events.
Disciplining narrative guesswork
The synthesis provides a vital mathematical discipline for subjective scenario selection.
However, the framework remains hostage to the chosen reference density, which can still miss unseen systemic risks.
Weighting alternative paths clarifies policy disagreement but cannot fix omitted structural mechanisms.