Econometrics for Macroeconomic Policy Evaluation

Informations projet

POLICYMETRICS

N° de convention de subvention: 101041145

DOI

10.3030/101041145

Projet clôturé

Date de signature de la CE 31 Mars 2022

Date de début 1 Septembre 2022

Date de fin 31 Août 2025

Financé au titre de

• European Research Council (ERC)

Coût total

€ 614 900,00

Contribution de l’UE

€ 614 900,00

Coordonné par

UNIVERSIDAD POMPEU FABRA
Spain

Periodic Reporting for period 2 - POLICYMETRICS (Econometrics for Macroeconomic Policy Evaluation)

Période du rapport: 2024-09-01 au 2025-08-31

Résumé du contexte et des objectifs généraux du projet

Modern monetary and fiscal policy decisions have far-reaching consequences for inflation, employment, public debt and financial stability. Yet evaluating whether policy makers have acted optimally remains difficult. Traditional approaches rely on fully specified structural macroeconomic models, which are often complex, controversial and sensitive to assumptions. This creates uncertainty about whether policy evaluations are robust and credible, especially when policies themselves change the economic environment.
The project “Econometrics for Macroeconomic Policy Evaluation” (POLICYMETRICS) addresses this challenge by developing new econometric tools to evaluate macroeconomic policy without requiring a complete structural model of the economy. The central idea is that two measurable objects—forecasts of policy objectives and impulse responses of these objectives to policy shocks—can serve as sufficient statistics for policy evaluation. By focusing on these directly estimable quantities, the project aims to provide transparent, model-robust, and time-consistent methods to assess policy performance.
The expected impact is significant. Reliable evaluation tools can improve accountability of policy institutions, inform public debate, and support better decision-making in central banks and fiscal authorities. The project also integrates social sciences by combining econometric methodology with economic theory and institutional analysis, thereby strengthening the link between quantitative methods and real-world policy design.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

The project delivered five core research contributions and several supporting methodological advances.
First, it demonstrated that policy optimality can be evaluated using only forecasts and impulse responses. These statistics are sufficient to detect and quantify deviations from optimal policy. The approach was applied to U.S. monetary policy from 1990 to 2022, showing that while deviations from optimality were typically small, some notable episodes—such as the early phase of the Great Recession—featured measurable departures from optimal decisions.
Second, the methodology was extended to allow historical comparisons of policy institutions over long time horizons. Applying the framework to 150 years of U.S. monetary policy revealed substantial improvements in policy conduct over the past three decades and provided a systematic comparison between different monetary regimes.
Third, the framework was adapted to measure policy makers’ preferences. A revealed-preference approach was developed to quantify “fiscal discipline” in the euro area, measuring how policy makers balance compliance with fiscal rules against macroeconomic stabilisation objectives.
Fourth, a comprehensive methodological review consolidated the sufficient-statistics approach and demonstrated its application to the evaluation of European Central Bank performance.
Fifth, the project introduced a novel econometric method—innovation-powered instrumental variables—which improves the precision of causal estimates when credible identification strategies suffer from low statistical power. This method reduces uncertainty in estimated policy effects by combining narrative identification with parametric information in a robust way.
In addition, the project produced advances in inference for dynamic causal effects and structural vector autoregressions, strengthening the econometric foundations of policy analysis.
Overall, the action resulted in high-impact journal publications, advanced working papers under revision at leading journals, new econometric methods, and the development of policy-oriented analytical tools.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

The project advances the state of the art in macroeconomic policy evaluation in several ways.
Conceptually, it shows that policy evaluation can be conducted using sufficient macroeconomic statistics alone, without specifying and estimating a full structural model. This represents a major methodological shift. Importantly, the approach is robust to the Lucas critique and can be implemented in a time-consistent manner—two longstanding challenges in macroeconomic analysis.
Methodologically, the innovation-powered IV estimator improves the trade-off between credibility and statistical precision in causal inference. Empirical applications show substantial reductions in confidence intervals for monetary and fiscal policy effects, increasing the reliability of policy conclusions.
Practically, the results provide tools that can be implemented by policy institutions using standard econometric techniques. Software development is underway to make the evaluation statistics accessible to non-specialists.
For further uptake and impact, continued work is needed in three areas:
1. Further research to refine identification of dynamic causal effects, especially for “news shocks” where expectations about future policies matter.
2. Development of user-friendly software interfaces to facilitate adoption by central banks and fiscal authorities.
3. Broader dissemination and training to ensure that practitioners can effectively apply the methods.
In summary, POLICYMETRICS delivers a new, robust framework for evaluating macroeconomic policy based on sufficient statistics, introduces methodological innovations that improve causal inference, and provides empirical evidence on monetary and fiscal policy performance. The results strengthen the scientific foundations of policy evaluation and offer practical tools for improving economic decision-making.