Macroeconomic Forecasting from Input-Output Tables Alone: A Darwinian Agent-Based Approach with FIGARO Data

TL;DR

This paper introduces a novel macroeconomic forecasting method using a Darwinian agent-based model driven solely by input-output tables, achieving accuracy comparable to professional forecasts without relying on time series or parameter estimation.

Contribution

The paper presents a new approach that uses input-output tables alone to generate out-of-sample GDP forecasts through an agent-based simulation, demonstrating high accuracy and cross-country portability.

Findings

Achieved MAE of 1.22 pp over 9 years, comparable to professional forecasts.

Confirmed cross-country portability across 33 countries with no parameter tuning.

Demonstrated the model's ability to simulate shock propagation and match firm size distributions.

Abstract

How much macroeconomic information is contained in a single input-output table? We feed FIGARO 64-sector symmetric tables into DEPLOYERS, a Darwinian agent-based simulator, producing genuine out-of-sample GDP forecasts. For each year, the model reads one FIGARO table for year N, self-organizes an artificial economy through evolutionary natural selection, then runs 12 months of autonomous free-market dynamics whose emergent growth rate predicts year N+1. The I-O table is the only input: no time series, no estimated parameters, no expectations formation, no external forecasts. We present five results. First, a 9-year Austrian panel (2010-2018) using 12-seed ensembles produces MAE of 1.22 pp overall; for five non-crisis years, MAE falls to 0.42 pp -- comparable to the best professional forecaster (WIFO: 0.48 pp). A Swedish 9-year panel independently confirms this accuracy (normal-years MAE 0.80 pp). Second, cross-country portability is demonstrated across 33 of 37 tested FIGARO countries with zero parameter changes. Third, a German 9-year panel reveals systematic +3.7 pp positive bias from export dependency -- an informative negative result pointing to multi-country network simulation as the natural extension. Fourth, a COVID-19 simulation demonstrates the I-O structure as a shock propagation mechanism: a 19-month timeline produces Year 1 GDP -4.62% vs empirical -6.6%. Fifth, emergent firm size distributions match European Commission data without micro-target calibration. These results establish the I-O table as serving a dual purpose: structural baseline engine and dynamic shock propagation mechanism. Since FIGARO covers 46 countries, the approach is immediately portable without retuning parameters.