I investigate the properties of output gaps in New Keynesian dynamic stochastic general equilibrium models and study the relationship between theory-based quantities and the estimates obtained with standard approaches. Theoretical gaps display low-frequency variations, have similar frequency domain representations as potentials and are generally correlated with them. Potentials have important business cycle variability. Existing statistical approaches fail to recognise these features and generate distorted estimates. Gaps are best estimated with a polynomial filter. Explanations for the outcomes are given. I propose a statistical procedure reducing estimation biases.
Canova, Fabio & Pappa, Evi (2024)
The macroeconomic effects of EU regional structural funds
This reflective article presents a case study of journalists in the media reporting from the trial against the former president of the International Biathlon Union. The main corruption charges against the former president were concerned with a car made available to him free of charge by a sponsor of biathlon, exclusive hunting trips paid for by a marketing firm, valuable watches given to him by biathlon organizers, and prostitutes offered to him by Russian biathlon officials. In terms of monetary value, the most valuable bribe was the car, and the least valuable bribe was sex work. Yet findings of this case study showed that the sport journalists preferred to write numerous articles about the alleged access to prostitutes. Fifty-nine major news reports reviewed in the presented research form the basis for this conclusion. Potentially long prison sentences, money involved, and famous defendants are some of the characteristics when selecting and reporting from white-collar crime cases. The findings of this case study support this claim
Canova, Fabio (2024)
Should we trust cross sectional multiplier estimates?
I examine the properties of cross-sectional estimators of multipliers, elasticities, or pass-throughs when a conventional spatial macroeconomic specification generates the data. A number of important biases plague standard estimates; the most relevant one occurs when the units display heterogeneous dynamics. Methods that work well in this situation are suggested. An experimental setting shows the magnitude of the biases cross-sectional estimators display. Average estimates of local fiscal multipliers in the US states are compared and contrasted.
Dynamic equilibrium models are specified to track persistent time series. Thus, unit roots are typically introduced as exogenous driving forces and the optimality conditions adjusted to produce a stationary solution. This adjustment step requires tedious algebra and often leads to algebraic mistakes, especially in complicated models. We propose an algorithm employing differentiation rules that simplifies the step of rendering non-stationary models stationary. It is easy to implement and works when trends are stochastic or deterministic, exogenous or endogenously determined. Three examples illustrate the mechanics and the properties of the approach. A comparison with existing methods is provided (97 words).
Canova, Fabio & Ferroni, Filippo (2022)
Mind the gap: Stylized dynamic facts and structural models
We study what happens to identified shocks and to dynamic responses
when the data generating process features q disturbances but q 1 < q
variables are used in an empirical model. Identified shocks are linear
combinations of current and past values of all structural disturbances
and do not necessarily combine disturbances of the same
type. Theory-
based restrictions may be insufficient to obtain structural
dynamics. We revisit the evidence regarding the transmission
of house price and of uncertainty shocks. We provide suggestions on
how to compare the dynamics of larger scale DSGEs models with
smaller scale VARs. (JEL E12, E13, E23, E31, E43, R31)
Canova, Fabio & Matthes, Christian (2021)
A composite likelihood approach for dynamic structural models
We explain how to use the composite likelihood function to ameliorate estimation, computational and inferential problems in dynamic stochastic general equilibrium models. We combine the information present in different models or data sets to estimate the parameters common across models. We provide intuition for why the methodology works and alternative interpretations of the estimators we construct and of the statistics we employ. We present a number of situations where the methodology has the potential to resolve well-known problems and to provide a justification for existing practices that pool different estimates. In each case, we provide an example to illustrate how the approach works and its properties in practice.
Canova, Fabio & Matthes, Christian (2021)
Dealing with misspecification in structural macroeconometric models
We consider a set of potentially misspecified structural models, geometrically combine their likelihood functions, and estimate the parameters using composite methods. In a Monte Carlo study, composite estimators dominate likelihood‐based estimators in mean squared error and composite models are superior to individual models in the Kullback–Leibler sense. We describe Bayesian quasi‐posterior computations and compare our approach to Bayesian model averaging, finite mixture, and robust control procedures. We robustify inference using the composite posterior distribution of the parameters and the pool of models. We provide estimates of the marginal propensity to consume and evaluate the role of technology shocks for output fluctuations.
Canova, Fabio; Ferroni, Filippo & Matthes, Christian (2020)
DETECTING AND ANALYZING THE EFFECTS OF TIME‐VARYING PARAMETERS IN DSGE MODELS
We study how structural parameter variations affect the decision rules and economic inference. We provide diagnostics to detect parameter variations and to ascertain whether they are exogenous or endogenous. A constant parameter model poorly approximates a time‐varying data generating process (DGP), except in a handful of relevant cases. Linear approximations do not produce time‐varying decision rules; higher‐order approximations can do this only if parameter disturbances are treated as decision rule coefficients. Structural responses are time invariant regardless of order of approximation. Adding endogenous variations to the parameter controlling leverage in Gertler and Karadi's model substantially improves the fit of the model.
We analyse the pass-through of monetary policy measures to lending rates to households and firms in the euro area using novel bank-level datasets. Banks’ characteristics such as the capital ratio, exposure to domestic sovereign debt, percentage of non-performing loans and stability of funding structure are responsible for the heterogeneity in the pass-through of conventional monetary policy changes. The location of a bank is irrelevant. Non-standard measures reduce lending rate heterogeneities. Banks located in financially stressed countries and with weak balance sheets are most affected. Banks’ lending margins have fallen considerably.
Canova, Fabio & Sahneh, Mehdi Hamidi (2018)
Are Small-Scale SVARs Useful for Business Cycle Analysis? Revisiting Nonfundamentalness