Research
PUBLICATIONS
Modelling pandemic risks for policy analysis and forecasting (joint with Elena Angelini, Milan Damjanović and Matthieu Darracq Pariès), Economic Modelling, Volume 120, 2023, 106162, ISSN 0264-9993,
What were the economic effects of vaccination campaign, agents’ adaptation to containment measures, and fiscal and monetary measures during the COVID-19 pandemic? To quantify these effects, we augment the ECB’s macro-econometric model with an endogenous epidemiological block, providing one of the first institutional policy models with epidemiological elements. The model tracks closely actual pandemic developments by relying on detailed epidemiological data and by nesting a wide range of features, including vaccination campaigns, emergence of new variants and learning effects. Importantly, it includes an endogenous policy reaction function for containment measures, allowing counterfactual simulations. Our simulation results show that: (i) agents’ adaptation to containment measures over time was key in reducing negative macroeconomic consequences of lockdowns, (ii) the vaccination campaign slowed down infection rates and hospitalizations, enabling a relaxation of containment measures and avoiding a sharp double dip in economic activity and (iii) complementary fiscal and monetary policy interventions provided support to the economy.
Interest rates and foreign spillovers (joint with Roberto De Santis), European Economic Review, Volume 144, 2022, 104043, ISSN 0014-2921
We show that medium-term interest rates in the euro area, Japan, UK and US are affected by domestic and foreign shocks. We find that US rates are the main source of spillovers globally and are less exposed to foreign shocks. Foreign spillovers to European rates were negligible only during the sovereign debt crisis and the introduction of more aggressive monetary policies by the ECB. We identify causal relations among asset prices through structural vector autoregressions (SVAR) and magnitude restrictions. We use preliminary regressions on event days to estimate key parameters employed to constrain the structural parameter space of the SVAR.
Sources of borrowing and fiscal multipliers (joint with Romanos Priftis), The Economic Journal, Volume 131, Issue 633, January 2021, Pages 498–519
This paper finds that debt-financed fiscal multipliers vary depending on the location of the debt buyer. In a sample of 33 countries fiscal multipliers are larger when government purchases are financed by issuing debt to foreign investors (non- residents), compared to when they are financed by issuing debt to home investors (residents). In a theoretical model, the location of the government creditor produces these differential responses through the extent that private investment is crowded out. International capital mobility of the resident private sector decreases the difference between the two types of financing both in the model and in the data.
Animal Spirits, Fundamental Factors and Business Cycle Fluctuations (joint with Stephane Dées), Journal of Macroeconomics, Volume 61 (2019), 103-123
We explore empirically the role of noisy information in cyclical developments to separate fluctuations due to genuine changes in fundamentals from those driven by temporary animal spirits (or noise shocks). Exploiting the fact that the econometrician has a richer dataset in some dimensions than the consumers, we use a novel identification scheme in a structural VAR framework and show that noise shocks are important drivers of business cycle fluctuations. In particular, noise shocks play a large role in consumption expenditures showing how false perceptions about future fundamentals influence consumer behaviors.
Spillovers during Euro Area Sovereign Debt Crisis: Network Analysis with Absolute Magnitude Restrictions (joint with Roberto De Santis), Journal of Applied Econometrics 33.5 (2018): 727-747.
This paper studies spillovers among US and European sovereign yields. We employ absolute magnitude restrictions on the impact matrix to identify the countries that were the main sources of spillovers. Despite the large size of shocks from euro area stressed countries, connectedness among sovereign yields declined between 2008 and 2012 due to financial fragmentation, particularly between countries with more divergent business and fiscal cycles. We show that none of the sovereign yields were insulated from foreign shocks and that shocks to the Greek bond market in 2010 explained 20–30% of the variance of sovereign yields in stressed countries, while in 2011–2012 Italy (not Spain) was the source of systemic risk.
WORKING PAPERS
Introducing ECB-BASE: The blueprint of the new ECB semi-structural model for the euro area (joint with Elena Angelini, Nikola Bokan, Kai Christoffel and Matteo Ciccarelli)
This paper presents the blueprint of a new ECB multi-country model. The version documented in the following pages is estimated on euro area data. As a prelude to the countrymodels, this version is meant to enhance the understanding of the main model mechanisms,enlarge the suite of area wide tools, and provide a tool for a top down approach betweeneuro area and country modelling. The model converges to a well-defined steady state and itsproperties are in line with macroeconomic theory and standard empirical benchmarks. Thedesign is aligned to its role as workhorse model in the context of the forecasting and policysimulation exercises at the ECB.
Keywords: Semi-structural model, euro area, simulations, forecasting, monetary policy
JEL Classification: C3, C5, E1, E2, E5
ECB working paper series
Can the Term Spread Reveal the Effect of News Shocks?
The paper shows how the term spread reacts to an unanticipated (surprise) and anticipated (news) Total Factor Productivity (TFP) shock. In a standard DSGE model, the surprise TFP shock decreases the term spread, while the term spread increases following news shock. The theoretical predictions serve to rationalize a sign restriction identification scheme to obtain news shocks in the data. The proposed method enables the identification of news shocks without restricting artificially constructed TFP data. I find that the economy expands following a positive news shock, but the bulk of the effect is delayed. News shocks explain around 30 percent of business cycle fluctuations, while they explain a higher share of variations in forward-looking variables, such as stock prices and the term spread.
Keywords: Technology shocks, News shocks, Term spread, Business Cycles, Identification, Structural Vector Autoregression, Dynamic Factor models
JEL Classification: C32, E32, E43
WORK IN PROGRESS
The news shock: evidence from oilfield discoveries (joint with Kirill Shakhnov)
We find that oil discoveries produce large anticipation effects on macroeconomic variables. Investment, imports and consumption increase immediately after the discovery of oil, while output and exports increase in the second period after the discovery. This is not driven by the oil production, because the production of oil only increases two periods after the discovery of oil. A standard RBC model is not able to replicate the observed patterns. We propose a modification of the standard RBC model needed to match the empirical results.
Keywords: News shocks, Anticipation Effects, Oil Discoveries, Panel Estimation, Local Projections
JEL Classification: C33, E32, Q32
TFP or IST Driven Business Cycles? - an Application of Importance Restrictions
I propose a novel identification scheme of structural shocks in Vector Autoregression models (SVAR) model that relies on assumptions about the relative importance of shocks in explaining the Forecast Error Variance (FEV) and Historical Variance (HV) of the endogenous variables. The methodology performs well in Monte Carlo exercises using data generated by a standard real business cycles model (RBC). I apply the methodology to investigate the importance of neutral technology shocks (TFP) and investment specific technology shocks (IST) as causes of business cycle fluctuations in US data. I find that their effects are in line with predictions of a standard neoclassical model and that TFP shocks are more important for inducing cyclical fluctuations.
Keywords: Identification, Structural Vector Autoregression, Forecast Error Variance, Historical Variance Decomposition, Technology shocks, Investment specifi.c technology shocks, Business Cycles
JEL Classification: C32, E32