Saltar para o conteúdo principal

FCCNFCT Digital Services

Access the national information on scientific and technological funding in Portugal

Activity Funded

Impulse response functions using local projections with instrumental variables with applications to macroeconomics

Funções de resposta ao impulso utilizando projeções locais com variáveis instrumentais, com aplicações à macroeconomia.

Reference
2024.17080.PEX
Project Start Date
2026-02-01
Project End Date
2027-07-31
Principal Investigator
Scientific Area
Social sciences
Funding Program
Concurso para Projetos de Investigação de caráter Exploratório em Todos os Domínios Científicos 2024

Abstract

Our project comprises three distinct parts focused on impulse response functions (IRFs) estimated through local projections with instrumental variables (LP-IV). IRFs are crucial for understanding the effects of monetary policy on the economy. Monetary policy significantly impacts macroeconomic variables (Bernanke& Blinder, 1992) and financial markets (Bernanke & Kuttner, 2005; Madeira & Madeira, 2019), making research in this area highly relevant for policy decisions. As Samuelson & Nordhaus (1985) highlight (in  Economics —a leading textbook for decades), monetary policy is "the most powerful and useful tool that macroeconomic policymakers have." IRFs also help assess the effects of fiscal policy (Blanchard & Perotti, 2002). Local projections (LP) have become a key tool for estimating IRFs, offering aflexible alternative to vector autoregressions. Since Jordà (2005), LPs have gained popularity in empirical macroeconomics and finance due to their ability to handle nonlinearities, structural breaks, and heterogeneous dynamics without restrictive assumptions. LP-IV (Local Projections with Instrumental Variables) enhances causal inference by addressing endogeneity, making it essential for macroeconomic research on exogenous shocks. Despite their advantages, conventional LP-IV methods suffer from autocorrelation and finite-sample bias in both covariance stationary and persistent data settings. The first part of our project develops a novel Modified Indirect Least Squares (MILS) estimator that corrects both issues, extending Lusompa’s (2023) autocorrelation correction to LP-IV while accommodating more shocks than observed variables. Leveraging the ILS interpretation of LP-IV (Plagborg-Møller& Wolf, 2021), we establish theoretical efficiency gains, reduce bias, and improve inference robustness. We further extend MILS to persistent data and multiple-instrument settings, linking it to Generalized Method of Moments (GMM) for broader empirical applications. Our findings provide a unified inference framework, eliminating the need for unit root pre-testing, which can distort inference (Mikusheva, 2007). Estimating the dynamic effects of aggregate uncertainty on individual outcomes is a central challenge in applied macroeconomics, yet inference in panel LPs remains underdeveloped. Almuzara& Sancibrian (2024) derive LP variance under stationarity and mean-independent errors but do not address autocorrelation, a critical issue in empirical work. The second part of our project extends MILS to panel LP and LP-IV, addressing autocorrelation and model misspecification while allowing for non-stationary data and more general error dependence. To enhance inference robustness, we develop a valid bootstrap method, improving IRF estimation reliability. Finally, in our third part of the project , we apply our autocorrelation and bias correction methods to three key macroeconomic applications. First, we reassess the impact of US monetary policy shocks on output, inflation, and credit, constructing a new instrument for the unobserved shock using an established approach with newly acquired data. Second, we provide the first empirical study of ECB monetary policy effects on Eurozone economies using a tailored shock identification method. Third, we extend our panel MILS method to a SUR (Seemingly Unrelated Regression) framework to analyze the effects of monetary policy on PCEPI (Personal Consumption Expenditures Price Index) subcomponents, offering fresh insights, particularly for Europe. By estimating IRFs with and without correction, our analysis enhances understanding of monetary policy transmission across economies and sectors, producing valuable insights for researchers, policymakers, and central bankers. Local projections bridge applied macroeconomics and applied microeconomics, providing a flexible, policy-relevant framework. The methodological advancements from this project could open new research avenues, benefiting not only macroeconomists but also microeconomists studying firm, household, and financial market responses to aggregate shocks.

Institutions

Main Institutions

  • ISEG Lisbon School of Economics and Management (ISEG/ULisboa)

Other Institutions

  • Tilburg University (TU)
  • ISCTE-Instituto Universitário de Lisboa (ISCTE-IUL)
  • School of Social Sciences - University of Manchester (SSS - UM)

Funding 59.967,30 €

Fundação para a Ciência e a Tecnologia (FCT) - Portugal

59.967,30 €