ABSTRACT. This paper identifies a new common risk factor in stock returns related to the fear of future jumps. The factor can be added to standard asset-pricing models leading to a five-factor model which is directed at capturing the size, value, profitability, momentum and fear in stock returns. The model outperforms the four-factor model of Carhart (1997).

ABSTRACT. This paper derives the predictive probability density function of a GARCH(1,1) process, under Gaussian or Student t innovations. The analytic form is novel, and replaces current methods based on approximations and simulations.

ABSTRACT. We consider bootstrap-based inference in non-causal autoregressive models with heavy tail innovations. We present asymptotic theory for the bootstrap considered and include detailed simulations to demonstrate the finite sample superiority when compared to asymptotic inference. The results are used for testing for no bubble type dynamics in a non-causal model for log-stock prices.

ABSTRACT. Volatility in financial markets alternates persistent turmoil and quiet periods. Modelling realized volatility time series requires a specification in which these subperiods are adequately represented. Changes in regimes is a solution, but the question of whether transition between periods is abrupt or smooth remains open. We provide a new class of models with a set of parameters subject to abrupt changes in regime (MS) and another set subject to smooth transition changes (ST). These models capture the possibility that regimes may overlap with one another ({\it fuzzy}). The empirical application is carried out on the volatility of four US indices and shows that the flexibility of the new model allows for a better overall performance over either MS or ST, and provides some interesting interpretation of the fuzzy regimes as related to transitions between low and high volatility.

ABSTRACT. In this study, we analyze the impact of formal early childcare on a number of non-cognitive child outcomes. Using a newly available dataset for Northern Italy, we condition our analysis on several predetermined socio demographic characteristics of the household and the child and we use different econometric techniques to deal with endogeneity of the choice of childcare. Looking at children at the end of first grade, we find that attendance of formal early chilcare has beneficial effects on generosity and prosocial behavior of the child, reduces her hyperactivity problems and improves her attitude towards schooling. The results of this research are likely to have important policy implications, since increasing availability of formal childcare can be an effective policy for improving subsequent child outcomes and reducing inequality among children from different backgrounds, in addition to facilitating maternal labor market participation.

ABSTRACT. The paper addresses the relationship between carnivals and economic performance. More specifically, it focuses on the Venice Carnivals in the period 2004-2014. The economic impact on local economy is measured in terms of tourism, passenger traffic, unemployment and spending of tourists. Following the literature, univariate time series econometric methods (ARIMA, intervention analysis) are used. The study provides evidence of Venice Carnival specific effects and confirms and quantifies its positive impacts on tourism in Venice as approximately 50.000 additional visitor arrivals and approximately 170.000 additional overnight stays. The chapter also quantifies the financial impact on the city economy and estimate the effects on short-term employment. As existing quantitative research in the field of cultural economics does not include carnival art issues, while also seldom surpasses the topic of ex-ante analysis, when tackling the issue of economic impacts, the chapter is a contribution in methodological as well as tourism and cultural economic sense.

ABSTRACT. This paper studies what happens when we move from a short regression to a long regression in a setting where both regressions are subject to misspecification. In this setup, the least-squares estimator in the long regression may have larger inconsistency than the least-squares estimator in the short regression. We provide a simple interpretation for the comparison of the inconsistencies and study under which conditions the additional regressors in the long regression represent a `balanced' addition to the short regression. We illustrate our results with a Monte Carlo experiment whose design is based on the empirical relationship between old-age health status and early-life conditions using retrospective data on the elderly European population from the Survey of Health, Ageing and Retirement in Europe (SHARE).

ABSTRACT. Recent theoretical advances identify unobservable cognitive parameters from the choices made by boundedly rational agents. However such models are often impractical for applications, as they assume that a large number of observations are available, sometimes even from all subsets of a universal set of alternatives. In this paper we focus on attention as a cognitive variable and show how to identify, in the theoretical model, the unknown attention distribution in a population of agents from a single dataset recording aggregate product shares. We also formulate an empirical version of the model, in which taste heterogeneity is captured as well as cognitive heterogeneity, and run a simple application of our methodology with actual data.

ABSTRACT. The condition of Risk Aversion implies that the Utility Function must be concave. Taking into account the dependence of the Utility Function on the wealth that in turn depends on the return, we consider a return with any type of two-parameter distribution. It is possible to define Risk and Return as a generic functions of these two parameters. This paper determines the Differential Conditions for the definitions of Risk and Return that maintain the Risk Aversion property in the 3D space of the Risk, Return and Expected Utility Function. As a particular case, in the paper we discuss these conditions in the case of the CRRA Utility Function and the Truncated Normal distribution.

ABSTRACT. Uncertainty about the choice of identifying assumptions in causal studies has been ignored until now. We consider uncertainty over a class of models that impose different sets of identifying assumptions, which, in general, leads to a mix of point- and set-identified models. We propose a method for performing inference in the presence of this type of uncertainty by generalizing Bayesian model averaging to allow for set-identified models. Our proposal is to consider ambiguous belief (multiple posteriors) for the set-identified models, and to combine them with a single posterior for models that are either point-identified or that impose non-dogmatic identifying assumptions in the form of a Bayesian prior. The output is a set of posteriors (post-averaging ambiguous belief ) that are mixtures of the single posterior and any element of the class of multiple posteriors, with mixture weights the posterior probabilities of the models. We establish conditions under which the data are informative about model probabilities, which occurs when the models are falsifiable and/or specify different priors for reduced-form parameters, and examine the asymptotic behavior of the posterior model probabilities. We propose to summarize the post-averaging ambiguous belief by reporting the range of posterior means and the associated credible regions, and offer a simple algorithm to compute these quantities. The method is general and allows for dogmatic and non-dogmatic identifying assumptions, multiple point-identified models, multiple set-identified models, and nested or non-nested models.

ABSTRACT. Causality is a widespread concept in theoretical and empirical economics. The recent nancial economics literature used Granger causality to detect the presence of contemporaneous links among nancial institutions and, in turn, to recover a network structure. Later studies combined the estimated networks with traditional pricing or risk measurement models to improve their t to empirical data. In this paper we provide two contributions: we show how a linear factor model can be used as a device to estimate a combination of several networks that monitor the links across variables from dierent viewpoints; we point out that, when the focus is on risk propagation, that Granger causality should be combined with quantile-based causality. The empirical evidence support the latter claim.

ABSTRACT. This paper evaluates the impact of access to credit from banks and other financial institutions on household welfare in Mauritania. Household level data are used to evaluate the relationship between credit access, a range of household characteristics, and welfare indicators. To address potential endogeneity issues, the household isolation level is used to instrument access to credit. Results show that households with older and more educated heads are more likely to access financial services, as are households living in urban areas. In addition, greater financial access is associated with a reduced dependence on household production and increased investment in human capital.

ABSTRACT. In this paper we use Structural VAR analysis to disentangle credit demand and supply shocks and their effect on real economic activity in Italy. The three endogenous variables considered are loan interest rate, the loans growth rate and the employment to population ratio. The data are observed at annual frequency, from 2008 to 2014, for each of 103 Italian provinces. The structural shocks are identified through heteroscedasticity, by letting the variance of the shocks to switch across four Italian macro-regions: North, Centre, South and Islands. The empirical findings suggest a more important role of credit supply shocks in shaping the level of real economic activity. Furthermore, the results show that credit crunch is hitting the North of Italy less than the remaining macro-regions, especially the South-Italy.

ABSTRACT. Since Basel II was introduced in 2008, multiple approaches to calculating regulatory capital requirements have co-existed in the mortgage market, leading different lenders to calculate different requirements for identical risks. Using a unique UK dataset, 2005-2015, and novel identification, we show this causes specialisation. Larger lenders obtain a comparative advantage in capital requirements for lower risk mortgages, and pass this through to interest rates (1.3bp per 1pp reduction in risk weight), with a corresponding portfolio shift. Conversely, risk concentrates more on the balance sheets of smaller lenders, with implications for systemic risk. Our results are relevant to live policy debates on reforms to the Basel framework.

ABSTRACT. The paper discusses the exercise recently carried out at ISTAT concerning the quarterly estimate of GDP by 30 days after the reference quarter. Main results are presented together with a coherent set of real time revision errors able to establish reliability of estimates with respect to those produced at 60 days when the full set of data is available. The modeling setup is built on main sub-components of GDP from both the production and the demand side, respectively 11 and 6 components. For each component a suitable set of autoregressive distributed lag models are fitted and estimations are carried out by generalized least squares within the Kalman filter methodology. Monthly indicators, for which last month of the quarter to estimate is not yet available, are preliminarily nowcasted by means of ARIMA models. GDP is obtained aggregating production components and changes of inventories are derived as residual, therefore including also statistical discrepancies. The proposal is novel with respect to applied research on Italian data for detail of GDP components, as well as the broad number of related indicators used in the exercise. Moreover an extensive real time analysis is conducted for both model selection and accuracy evaluation of estimates at T+30. Real time analysis provides the revision errors for GDP and its components, relative gains with respect to pure autoregressive models adopted as benchmark and relative losses with respect to the official flash estimate of GDP published at T+45 days.

ABSTRACT. This paper investigates the effect of seasonal adjustment filters on the identification of mixed causal-noncausal autoregressive (MAR) models. By means of Monte Carlo simulations, we find that standard seasonal filters might induce spurious autoregressive dynamics, aphenomenon already documented in the literature. Symmetrically, we show that those filters also generate a spurious noncausal component in the seasonally adjusted series. An empirical application on European in infation data illustrates these results. In particular, whereas several inflation rates are forecastable on seasonally adjusted series, they appear to be white noise using raw data.

ABSTRACT. We apply indirect inference to estimate nonlinear dynamic panel data models. Monte Carlo simulations show that our method compares well with respect to other methods recently suggested in the scientific literature. Indirect inference seems to be particularly competitive in short panels.

ABSTRACT. I propose a xed e ects expectation-maximization (EM) estimator that can be ap- plied to a class of nonlinear panel data models with unobserved heterogeneity, which is modeled as individual e ects and/or time e ects. Of particular interest is the case of interactive e ects, i.e. when the unobserved heterogeneity is modeled as a factor analytical structure. The estimator is obtained through a computationally simple, iter- ative two-step procedure, where the two steps have closed form solutions. I show that estimator is consistent in large panels and derive the asymptotic distribution for the case of the probit with interactive e ects. I develop analytical bias corrections to deal with the incidental parameter problem. Monte Carlo experiments demonstrate that the proposed estimator has good nite-sample properties.

ABSTRACT. We build a simple diagnostic criterion for approximate factor structure in large cross-sectional equity datasets.Given a model for asset returns with observable factors, the criterion checks whether the error terms are weakly cross-sectionally correlated or share at least one unobservable common factor. It only requires computing the largest eigenvalue of the empirical cross-sectional covariance matrix of the residuals of a large unbalanced panel. A general version of this criterion allows us to determine the number of omitted common factors. The panel data model accommodates both time-invariant and time-varying factor structures. The theory applies to random coefficient panel models with interactive fixed effects under large cross-section and time-series dimensions. The empirical analysis runs on monthly and quarterly returns for about ten thousand US stocks from January 1968 to December 2011 for several time-invariant and time-varying specifications. For monthly returns, we can choose either among time-invariant specifications with at least four financial factors, or a scaled three-factor specification. For quarterly returns, we cannot select macroeconomic models without the market factor.

ABSTRACT. This paper investigates on the base of a theoretical spatial model the determinants of macroeconomic volatility of per capita GDP in a panel of 257 NUTS-2 European regions in the period 1992-2008, separately considering positive and negative fluctuations. We find evidence of strong positive spatial dependence, and of remarkable asymmetric effects on macroeconomic volatility of sectoral output (its composition and concentration), of composition of aggregate demand, and of other regional/country characteristics. In particular, while public expenditure exerts a stabilizing effect on both types of fluctuations, financial depth amplifies negative fluctuations. Finally, inflation fluctuations and the participation in EMU appears to have no effect on macroeconomic volatility.

ABSTRACT. In this paper, we model the sovereign debt default intensities for two groups of European countries: a first group included in the European Union and a second group in the Eurozone. The default intensities are assumed to follow self- exciting point processes depending on two components: a systemic component and a country specific component. The first component is built as a simple weighted average of individual countries' default intensities. The weights are unknown parameters and confer information on the contribution of each country's default risk to the overall systemic risk. Moreover, we measure the impact of the systemic risk on individual risk by introducing a special parameter reacting the speed at which each country's default intensity adjusts to the changes in the default intensity of the EU and Eurozone markets. We show the above model is affine in the state variables and use the results in Duffie et al (2000) to obtain closed form CDS prices. We estimate the model using implied state QMLE based on the CDS spreads on 7 different maturities.

ABSTRACT. We propose parametric and semiparametric IV estimators for spatial autoregressive models with network data where the network structure is endogenous. We embed a dyadic network formation process in the control function approach as in Heckman and Robb (1985). In the semiparametric case, we use power series to approximate the correction terms. We establish the consistency and asymptotic normality for both parametric and semiparametric cases. We also investigate their nite sample properties via Monte Carlo simulation.

ABSTRACT. We propose a new linear regression model with tensor variate response and covariates, that encompasses univariate and multivariate regression models as special cases. For dealing with the over-parametrization and over-tting issues due to the curse of dimensionality, we exploit a suitable parametrization which enables to achieve both parameter parsimony and to incorporate sparsity eects. Inference is carried out in the Bayesian framework combined with Monte Carlo Markov Chain (MCMC). We show the eciency of the MCMC procedure on simulated data.

ABSTRACT. Seemingly unrelated regression (SUR) models are used in studying the interactions among economic variables of interest. In a high dimensional setting and when applied to large panel of time series, these models have a large number of parameters to be estimated and suffer of inferential problems. We propose a Bayesian nonparametric hierarchical model for multivariate time series in order to avoid the overparametrization and overfitting issues and to allow for shrinkage of the SUR coefficients toward multiple prior means with the location, scale and shape parameters of these means unknown. We propose a two-stage hierarchical prior distribution. The first stage of the hierarchy consists in a lasso conditionally independent prior distribution of the Normal-Gamma family for the SUR coefficients. The second stage is given by a random mixture distribution for the Normal-Gamma hyperparameters, which allows for parameter parsimony through two components. The first one is a random Dirac point-mass distribution, which induces sparsity in the SUR coefficients; the second is a Dirichlet process prior, which allows for clustering of the SUR coefficients. We provide a Gibbs sampler for posterior approximations based on introduction of auxiliary variables. Some simulated examples show the efficiency of the proposed methods. We study the effectiveness of our model and inference approach with an application to macroeconomics.

ABSTRACT. This paper contributes to the literature on the effects of operational audits on individual taxpayers’ subsequent tax compliance. It focuses on Italy, a country characterized by low-tax morale and by high opportunity of tax evasion, which has not yet been considered in the literature. The empirical analysis is based on a large administrative tax returns panel dataset of Italian taxpayers merged with a tax audit database, both made available by the Italian Revenue Agency. Our identification strategy relies on a difference in difference estimator with ex-ante matching. We find that, on average, after-audit annual reported income increases among audited taxpayers relative to untreated individuals. The positive impact is increasing in the first three years after audit and then decreasing, disappearing four years after the audit. When disentangling results according to audits’ final outcome, we do not find any effect when taxpayers are found compliant, while the effect is the largest when their assessment is fully accepted by the taxpayer. Our results are particularly relevant from a policy perspective, providing suggestions for improving the criteria used for selecting the taxpayers to be audited.

ABSTRACT. This article contributes to the vast empirical literature on companies' capital structure by using an estimator for dynamic panel data models with correlated random effects. This approach is helpful to understand the determinants of the capital structure of the European foreign owned subsidiaries. As will be shown, when applying a dynamic panel data model the estimated impact of taxes and profitability on subsidiaries' financial choices varies across different quantiles of the (skewed) distribution of the leverage.

ABSTRACT. We use confidential data on Value Added Tax payments at the sector level, in two large Italian cities, to estimate the effect of audits publicity on tax compliance of local sellers. By employing a Difference-in-Differences identification strategy, we find that such publicity has a positive effect on fiscal declarations made shortly after. The results suggest that increasing awareness on future audits via the media can be an important instrument in the hands of tax authorities.

ABSTRACT. In this paper, we investigate the determinants of the primary deficit of the regional government, focusing in particular on political economy hypotheses. To test our model, we apply the System Generalized Method of Moments on a panel dataset for the 20 Italian regions over the period between 2000 and 2013. In line with the political budget cycle theory, we find that the primary deficit is lower in the first year after elections than in the subsequent years. Moreover, primary deficit appears negatively related to EU structural funds and payments to service the debt. On the contrary, primary deficit does not seem significantly related to the political orientation of the president of the region and to various macroeconomic and political indicators of the central government.

ABSTRACT. The aim of this paper is two-fold. Firstly, it analyses the impact of banks’ performance on financial stability and its direction taking into account the Italian context during the period 2001-2011. Secondly, it pays particular attention to the role of market structure. The z-score is used as financial stability indicator, while the performance of financial intermediaries is measured using a within-transformation parametric method recently developed (Wang and Ho, 2010). Our data validates one-way direction, where bank performance contributes to make the financial system more stable. In other words, the higher is the performance of the financial intermediaries, the higher is the financial stability of the system. Based on the role of market structure, the evidence suggests that moderately concentrated market contribute to financial stability in term of performance, consistently with the “concentration view” (Freixas and Rochet, 1997). Results hold when robustness checks are performed.

ABSTRACT. In this paper we examine the relationship between the default risk of banks and sovereigns, i.e. the ‘doom-loop’. Specifically we try to assess the effectiveness of the implementation of the new recovery and resolution framework in the European Union. We use a panel with daily data on European banks and sovereigns ranging from 2012 to June 2016 in order to test the effects of the BRRD on the two-way feedback process. We find that there was a pronounced feedback loop between banks and sovereigns from 2012 to 2014. However, this feedback loop seems to have disappeared after the implementation of the new regulatory framework. This finding is robust across several specifications.

ABSTRACT. Since macro-prudential stress tests have become the main instruments of the supervisory authorities toolkit, the debate on the effect of their results disclosure inflamed. Our work aims at providing a framework that, via a dynamic estimation of the betas, allows to observe the impact of the new information flow on the stability of the banking system. What we find is that, contrary to literature wisdom, almost all banks betas decrease, as the transparency shock contributes to an overall systemic risk drop.

ABSTRACT. Job disruptions is one of the most visible effects of financial crises. We contribute to the empirical literature on the employment effects of credit supply, investigating individual heterogeneity across firms, workers and jobs in the response to a financial shock. We use an extremely rich data set on job contracts in one Italian region, matched with the universe of firms and their lending banks. To isolate the effect of credit supply shocks we build a firm-specific time-varying measure of credit restrictions. Our findings indicate that a 10 percent supplydriven credit contraction reduces employment by 2.5 percent. The effect is concentrated among temporary contracts and less educated workers, consistently with a skill upgrading effect of the credit crunch.

ABSTRACT. This paper presents a nonparametric structural analysis of a class of contract models à la Baron Myerson (1982). Our analysis is based on a well-posed inverse problem linking the quantile function of the observations and the functional parameter of interest. The resolution of this problem gives the identification properties of the model and leads to an estimation procedure. We provide implementation and asymptotic properties of this type of L-estimator. We extend our analysis by introducing an instrumental variable estimator of the cost function.

ABSTRACT. This paper considers linear rational expectations models from the linear systems point of view. Using a generalization of the Wiener-Hopf factorization, the linear systems approach is able to furnish very simple conditions for existence and uniqueness of both particular and generic linear rational expectations models. The paper provides two applications of this approach; the first describes necessary and sufficient condition for exogeneity in linear rational expectations models and the second provides an exhaustive description of stationary and cointegrated solutions, including a generalization of Granger's representation theorem. Finally, the paper provides an innovative numerical solution to the Wiener-Hopf factorization and its generalization.

ABSTRACT. The paper proposes a novel methodology to identify structural shocks in Vector Autoregressions (VARs) by using high frequency information. Structural shocks are recovered in high frequency systems, aggregated at the macro frequency, and used as a proxy for a structural shock of interest in lower frequency VARs. We label this methodology Bridge Proxy-SVAR because the Proxy-SVAR links high frequency information with low frequency variables. The Bridge Proxy-SVAR mitigates time aggregation biases in structural analysis and it is particularly appealing to study macro-financial linkages. In fact, it is applicable even when the dimension of the system is large and with a wide frequency mismatch (e.g. monthly-daily). In this way, we extend the implementability of the high frequency identification in Proxy-SVARs to those cases in which narrative series or key events are not available. We illustrate the properties of the Bridge Proxy-SVAR both analytically and through Monte Carlo simulations. Finally, we present an application to monetary policy in the US finding consistent results with the most recent advancement in the literature.

ABSTRACT. Mainstream macroeconomic theory predicts a rapid response of asset prices to monetary policy shocks, which conventional empirical models are unable to reproduce. We argue that this is due to a deficient information set: Forward-looking economic agents observe vastly more information than the handful of variables included in standard VAR models. Thus, small-scale VARs are likely to suffer from nonfundamentalness and yield biased results. We tackle this problem by estimating a Structural Factor Model for a large euro area dataset. We find quicker and larger effects of monetary policy shocks, consistent with mainstream theory and the observed large swings in asset prices. Our results point to stronger financial stability consequences of an exogenous monetary policy tightening, also in the form of a quicker than expected unwinding of QE, than commonly thought.

ABSTRACT. Market risk premia play an important role in the transmission of monetary policy. If the transmission were to work asymmetrically for positive and negative shocks, monetary authorities would face a problematic trade-off: a temporary stimulus could boost the economy in the short run, but at the same time sow the seeds of a painful medium-run market reversal (the ``financial stability dark side" of monetary policy of Stein, 2014). We study the relation between interest rates, credit spreads and output in the U.S. using monthly data and a range of nonlinear dynamic models. We find clear signs of a reduced-form asymmetry, but no evidence in support of the causal mechanism that underpins the `dark side' argument: spreads rise noticeably ahead of economic slowdowns but they do not appear to cause them directly, particularly if they move in response to monetary shocks. This suggests that the asymmetry is best interpreted as a purely predictive relation, with markets being particularly sensitive to bad economic news; and that it creates no complications for monetary policy or for the exit strategy from monetary accommodation.

ABSTRACT. This paper explores the potential of bootstrap methods in the empirical evaluation of dynamic stochastic general equilibrium (DSGE) models and, more generally, in linear rational expectations models featuring unobservable (latent) components. We consider two dimensions. First, we provide mild regularity conditions that suffice for the bootstrap Quasi-Maximum Likelihood (QML) estimator of the structural parameters to mimic the asymptotic distribution of the QML estimator. Consistency of the bootstrap allows to keep the probability of false rejections of the cross-equation restrictions under control. Second, we show that the realizations of the bootstrap estimator of the structural parameters can be constructively used to build novel, computationally straightforward tests for model misspeci…fication, including the case of weak identi…fication. In particular, we show that under strong identi…fication and bootstrap consistency, a test statistic based on a set of realizations of the bootstrap QML estimator approximates the Gaussian distribution. Instead, when the regularity conditions for inference do not hold as e.g. it happens when (part of) the structural parameters are weakly identifi…ed, the above result is no longer valid. Therefore, we can evaluate how close or distant is the estimated model from the case of strong identi…fication. Our Monte Carlo experimentations suggest that the bootstrap plays an important role along both dimensions and represents a promising evaluation tool of the cross-equation restrictions and, under certain conditions, of the strength of identi…fication. An empirical illustration based on a small-scale DSGE model estimated on U.S. quarterly observations shows the practical usefulness of our apprach.

ABSTRACT. In this paper we propose a new framework for modelling heteroskedastic structural vector autoregressions. The identification of the structural parameters is obtained by exploiting the heteroskedasticity in the data naturally arising during crisis periods. More precisely, we provide identification conditions when heteroskedasticity and traditional restrictions on the parameters are jointly considered. Although the framework is general enough to find potential applications in many empirical economic fields, it proves to be well suited for distinguishing between interdependence and contagion in the literature related to the transmission of financial crises. This methodology is used to investigate the relationships between sovereign bond yields for some highly indebted EU countries.

ABSTRACT. In this paper we develop a new theoretical framework for the estimation of state space models with time varying parameters. We let the driver of the time variation be the score of the predictive likelihood and derive a new filter that allows us to estimate simultaneously the state vector and the time-varying parameters. In this setup the model remains Gaussian, the likelihood function can be evaluated using the Kalman filter and the model parameters can be estimated via maximum likelihood, without requiring the use of computationally intensive methods. Using a Monte Carlo exercise we show that the proposed method works well for a number of different data generating processes. We also present two empirical applications. In the former we improve the measurement of GDP by combining alternative noisy measures, in the latter we construct an index of financial stress and evaluate its usefulness in nowcasting GDP in real time. Given that a variety of time series models have a state space representation, the proposed methodology is of wide interest in econometrics and applied macroeconomics.

ABSTRACT. We develop two algorithms to identifying Smooth Transition Vector Autoregressive models (STVARs) using sign restrictions. The aim is to contribute to the literature by showing that STVARs can be identified using approaches that differ from the recursive identification, which is instead the one almost exclusively used for such models. We show an application to the study of monetary policy shocks in expansions and in recessions. We find no observationally equivalent structural representation alternative to the recursive identification in which monetary shocks are more effective in expansions. This suggests that monetary shocks are more effective in recessions rather than in expansions, confirming previous findings from the literature.

ABSTRACT. In this paper we propose a new class of Dynamic Mixture Models (DAMMs) being able to sequentially adapt the mixture components as well as the mixture composition using information coming from the data. The information driven nature of the proposed class of models allows to exactly compute the full likelihood and to avoid computer intensive simulation schemes. An extensive Monte Carlo experiment reveals that the new proposed model can accurately approximate the more complicated Stochastic Dynamic Mixture Model previously introduced in the literature as well as other kind of models. The properties of the new proposed class of models are discussed through the paper and an application in financial econometrics is reported.

ABSTRACT. We propose a decomposition to distinguish between Knightian uncertainty (ambiguity) and risk, where the first measures the uncertainty about the probability distribution generating the data, while the second measures uncertainty about the odds of the outcomes when the probability distribution is known. We use the Survey of Professional Forecasters (SPF) density forecasts to quantify overall uncertainty as well as the evolution of the different components of uncertainty over time and investigate their importance for macroeconomic fluctuations. We also study the behavior and evolution of the various components of our decomposition in a model that features ambiguity and risk.

ABSTRACT. So far, little work has been done on directly estimating the difference of gaps in earnings. Studies estimating pay differentials, generally compare them across different subsamples. This comparison does not allow to conduct any inference or to confront the respective decomposition components across the subsamples. Using Italian micro-data, we estimate the difference in the gender pay gap across time (2005 and 2014) in a path-independent way. By applying our proposed decomposition, we find that the convergence of the gender pay gap over time is only driven by the catching-up of women in terms of labor market characteristics, while the impact of anti-discrimination legislation is found to be negligible.

ABSTRACT. This paper documents the transgenerational effects of exposure to World War II hardships on second generation outcomes. We use linked survey data on two generations from different European countries. The focus is on cohorts of parents born between 1920--1956, most of whom were exposed to military operations, bombing, even hunger. Our findings indicate that children whose parents were exposed to hardships have a lower education attainment. Moreover, mothers' education seem to be very important for girls but not for boys. These results are robust to a series of specifications.

ABSTRACT. This paper studies parental decisions on resource allocation in a condition of a strong and evident ability imbalance among the offspring: when one of the children is considered mentally or physically disabled. Standard economic models posit that parents should optimally invest more in the more able child, however empirical evidence has shown that when deciding resource allocation, parents' equality concerns seem to prevail despite children's innate ability levels. This analysis identifies causal changes in parental inputs given to children, more specifically quality time, due to living with an ill sibling among sibling couples in the Panel Study of Income Dynamics by using matching techniques. Parents seem to devote more quality time to girls and older children. Moreover, mothers are the ones to dedicate more time to the offspring, while fathers do not seem to be affected. Further investigations highlight that children receive the same amount of time than their siblings, very likely being involved in activities with the ill child.

ABSTRACT. We use a large and unique administrative dataset from the district of Verona (Italy), covering the period 2008-2015, to compare the effectiveness of alternative policies to select employees to monitor through a medical visit, among those on sick leave. We find that private employers are more effective than the public insurer in selecting for monitoring employees who are more likely to be fit to work. However, the advice of appropriate data-driven tools can close the gap. We discuss the impact of using direct measures of health, such as the outcome of a medical visit, on the study of the determinants of opportunistic behavior and argue that simply looking at days of work lost, without appropriately controlling for health status, may lead to inaccurate conclusions when studying moral hazard.

ABSTRACT. Let us consider a discrete-time $n$-dimensional stochastic process $z$, with components $x=(x_1,...,x_{m_1})'$ and $y=(y_1,...,y_{m_2})'$, $m_1+m_2=n$. We want to study causality relationships between the variables in $x$ and $y$. Suppose that we find that $y$ Granger causes $x$. Then we would expect to be able to pick out at least one of these variables, say $y_j$, having a causal impact on $x$. It turns out that, when we consider the conditioning information set defined by the past observations of $x$ and all the $y_i$, $i\neq j$, it may be that $y_j$ has no causal impact on $x$, irrespective of the particular $j=1,2,...,m_2$ that we tried to pick out. This is a puzzling property. The paper provides a condition under which this puzzling property cannot hold.

ABSTRACT. We consider a system of spatially correlated cointegrating relationships. In addition to the correlation induced by the spatial autoregressive formulation, we also allow for cross-unit correlation of the integrated regressors as well as the error terms. Cointegration amongst the regressors is not allowed, as is standard in the cointegrating regression literature. The convergence rate of the spatial correlation parameter is determined by the order of the deterministic trend polynomial. The fully modified ordinary least squares principle is extended to this setting to allow for applying for standard asymptotic inference. The theoretical results are complemented by a small simulation study. Finally, the methodology is applied to investigate credit risk correlation, where { by using a data set comprising the main U.S. corporate default swap dealers { highly significant spatial correlation is observed.

ABSTRACT. The goal of this paper is to provide further evidence for the impact of the Bureau of Labor Statistics PPI component for all primary commodity series (PPI) into the consumer price index (CPI) by shedding light on the interaction between their short- and long-term dynamics. To do so, we use recent developments on threshold cointegration allowing for a nonlinear adjustment between the variables to some long-term equilibrium and a battery of linear and nonlinear causality tests. The empirical findings suggest, in terms of short-run behavior, the presence of a clear leading of PPI growth rates when for the long-run dynamics these regimes are characterized as "normal". The very interesting exception is observed over the "turbulent" regime from December 1973 to February 1986, where the nonlinear causal link from the PPI changes to inflation is dominated by the tendency for the commodity and consumer prices to move towards long-run equilibrium. The asymmetric adjustment to the long-run equilibrium between PPI and CPI offers extremely important insights for the nature of the information flow from the commodity price changes to inflation in the short-run.

ABSTRACT. This paper provides some exercises (circuits) aimed at evaluating the performance of Likelihood maximization algorithms for I(1) and I(2) Vector Error Correction Models in terms of effectiveness (i.e. ability to find the true maximum of the likelihood) and efficiency (i.e. speed in finding the maximum, conditionally on the fact that the algorithm has reached the maximum)

ABSTRACT. We analyze the pass-through of monetary policy measures to firms and household lending rates in the euro area using a novel bank-level dataset. Capital adequacy, exposure to sovereign debt and magnitude of non-performing loans are responsible for the heterogeneity of pass-through to conventional monetary policy changes. The location of a bank is instead irrelevant. ECB non-standard measures normalized the capacity of banks to grant loans. Banks with high level of non-performing loans and low capital ratio were most affected. Banks' lending were also affected. Policy implications are discussed.

ABSTRACT. In this paper we extend a deep learning approach for predicting bank distress from news text trough the integration of numerical financial data to the model. The model is composed of an unsupervised part to learn the semantic representation of news and of a supervised neural network that classifies banks between distress and tranquil states, based on a small set of known distress events. The classifier is trained on vectors composed of a semantic and a numerical indicator part in order to learn to distinguish distress based on text in the context of the financial situation of a bank. The objective is to enhance the predictive capability of the model with the ultimate aim of providing accurate realtime descriptions of distress.

ABSTRACT. Giudici and Parisi (2016) have proposed a novel econometric approach that measures systemic risk as a probabilistic "add-on" to the idiosyncratic probability of default of an economic sector (sovereign, corporate or bank). In this contribution we extend their approach to financial institutions and, doing so, we investigate the relative advantage, in terms of systemic risk, of a bail-in versus a bail-out scenario. We apply our methods to the Italian bail-out private intervention scheme named Atlante. The results show that the bail-out of a troubled bank and, specifically of the Banca Popolare di Vicenza, is more convenient for the smaller, safer and highly correlated banks.

ABSTRACT. We develop a measure of systemic risk for international capital flows between banking systems that takes into account the probabilty of default (PD) of each country and also how countries interact via multiple linkage types. We employ tensor decomposition to extract an adiacency matrix from a multilayer network. The PD of each country is extracted from the corresponding CDS spreads. We adapt the alpha-centrality measure to incorporate the PDs into a new network centrality measure that can be probabilistically interpreted.