ABSTRACT. A Markov-switching observation-driven model is a stochastic process ((S_t,Y_t))_{t∈Z} where (i) (S_t)_{t∈Z} is an unobserved Markov process taking values in a finite set and (ii) (Y_t)_{t∈Z} is an observed process such that the conditional distribution of Y_t given all past Y’s and the current and all past S’s depends only on all past Y’s and S_t. In this paper, we prove the consistency and asymptotic normality of the maximum likelihood estimator for such model. As a special case hereof, we give conditions under which the maximum likelihood estimator for the widely applied Markov-switching generalised autoregressive conditional heteroscedasticity model introduced by Haas et al. (2004b) is consistent and asymptotic normal.

ABSTRACT. A new general class of models which lies between observation driven and parameter driven models is presented. Several examples are discussed in the contexts of conditional mean and variance specifications. Conditions for strong stationarity and existence of moments are derived. Consistency and asymptotic normality of the Maximum Likelihood Estimator are derived for the general model specification and for specific examples. An application in financial econometrics shows the usefulness of the proposed class of models.

ABSTRACT. This paper proposes a novel canonical correlation analysis for semiparametric inference in $I(1)/I(0)$ systems via functional approximation. The approach can be applied coherently to panels of $p$ variables with a generic number $s$ of stochastic trends, as well as to subsets or aggregations of variables. This study discusses inferential tools on $s$ and on the loading matrix $\psi$ of the stochastic trends (and on their duals $r$ and $\beta$, the cointegration rank and the cointegrating matrix): asymptotically pivotal test sequences and consistent estimators of $s$ and $r$, $T$-consistent, mixed Gaussian and efficient estimators of $\psi$ and $\beta$, Wald tests thereof, and misspecification tests for checking model assumptions. Monte Carlo simulations show that these tools have reliable performance uniformly in $s$ for small, medium and large-dimensional systems, with $p$ ranging from 10 to 300. An empirical analysis of 20 exchange rates illustrates the methods.

ABSTRACT. Engle and Russell (1998, Econometrica, 66:1127–1162) apply results from the GARCH literature to prove consistency and asymptotic normality of the (exponential) (Q)MLE for the autoregressive conditional duration (ACD) model, under the assumption of strict stationarity and ergodicity of the durations. Using novel arguments based on renewal process theory, we show in this note that their results hold under the stronger requirement that durations have finite expectation. We also show that this is not the case in general, and provide a counterexample where the MLE is asymptotically mixed normal and converges at a rate significantly slower than usual, even if the durations are stationary and ergodic. The main difference between ACD and GARCH asymptotics is that the latter do not account for the fact that the number of durations in a given time span is random. As a by-product, we provide a new lemma which can be applied to analyze asymptotic properties of extremum estimators when the number of observations is random.

ABSTRACT. We propose a bootstrap generalized Hausman test for the correct specification of unobserved heterogeneity in both linear and nonlinear fixed-effects panel data models. We consider as null hypotheses two scenarios in which the unobserved heterogeneity is either time-invariant or specified as additive individual and time effects. We contrast the standard fixed-effects estimators with the recently developed two-way grouped fixed-effects estimator, that is consistent in the presence of time-varying heterogeneity under minimal specification and distributional assumptions for the unobserved effects. The Hausman test exploits the general formulation for the variance of the vector of contrasts and critical values are computed via parametric percentile bootstrap, so as to account for the non-centrality of the asymptotic χ2 distribution arising from the incidental parameters and approximation biases. Monte Carlo evidence shows that the test has correct size and good power properties. We provide two empirical applications to illustrate the proposed test: the first one is based on a linear model for the determinants of the wage of working women and the second analyzes the trade extensive margin.

ABSTRACT. This paper addresses the challenge of cross-sectional dependence (CSD) in multi-dimensional panel data settings. Extensive research highlights the importance of modeling CSD through multiplicative interactive effects (IE). Two leading methodologies—the principal component (PC) estimation method by Bai (2009) and the common correlated effects (CCE) estimator by Pesaran (2006)—have been widely adopted and expanded upon. Recent empirical needs have driven the development of econometric techniques suitable for multi-dimensional panel datasets, which arise in contexts like trade flows, employer-employee matches, and social networks. Building on the LM procedure introduced by KSS (2023) for two-dimensional panels, this study extends the test to a three-dimensional (3D) framework where units i and j are observed across time t. We propose a 3D version of the LM test to validate the conditional independence of factor loadings and regressors. Additionally, we introduce unit-specific LM tests to identify subsets of data where loadings remain conditionally independent. This allows for valid estimation without fully accounting for CSD, offering a significant advantage over two-dimensional approaches by isolating and excluding correlated units. This innovation underscores the potential of multi-dimensional datasets in overcoming the complexities of CSD in econometric analysis.

ABSTRACT. Many studies in finance use difference-in-differences (DID) methods to analyse the impact of a change in policy or regulation. Frequently, the outcome variable is measured over many time periods and exhibits serial correlation which is not typically modelled by existing approaches. We propose an autoregressive DID model which accounts for the dynamic behaviour of financial outcomes. It also enables the estimation of dynamic treatment effects many periods into the future, and can incorporate structural breaks. We derive the properties of the OLS estimator of the model, showing that it has a bias which can be attenuated using a novel bias correction formula. We apply the method to revisit the effect of tax reform on the financial assets of large firms. When dynamics are added to a static DID model, the implied long-run treatment effects can be much larger in magnitude.

ABSTRACT. in recent econometric literature, univariate panel models with interactive fixed effects have been extensively examined. Our research advances this field by proposing a multivariate panel vector autoregression (PVAR) approach incorporating group-based factors. The proposed model offers substantial methodological advantages. Its key strengths include remarkable flexibility, as it does not require predetermined specifications of group numbers or group membership. Moreover, the model can be extended to accommodate group-specific heterogeneous coefficient specifications in Panel VAR frameworks. Methodologically, we demonstrate that the model represents a parsimonious structural approach with computational efficiency. Through rigorous analysis, we derive the asymptotic distribution of the estimator and establish its consistency as both cross-sectional (N) and time (T) dimensions approach infinity. An empirical investigation illustrates the model's potential, revealing that group-based pattern analysis can yield substantively different insights compared to conventional Panel VAR methodologies. These findings underscore the model's potential to enhance econometric understanding of complex panel data dynamics.

ABSTRACT. We propose a new measure of price discovery, which we will refer to as the In dependent Component based Information Share (IC-IS). This measure constitutes a variant of the widespread Information Share, with the main difference being it does not suffer the same identification issues. Under the assumptions of non-normality of the shocks, a rather general theoretical framework leading to the definition of the IC IS and its estimation via a pseudo maximum likelihood (PML) approach is illustrated. After testing the robustness of the measure in a Montecarlo simulation environment, we illustrate two separate empirical analyses encompassing different price discovery applications.

ABSTRACT. This study examines the price dynamics of ten commodities across three sectors: clean energy, fossil fuels, and raw materials. Using daily closing prices from May 6, 2014, to October 31, 2023, we propose a two-step approach combining a cointegration analysis with a spillover volatility transmission model. We find that a long-run relationship between fossil and clear energy exists, consistently with the ongoing energy paradigm. We document a transmission of volatility over time and spillovers increase in magnitude, especially after the outbreak of the Covid-19 pandemic and the Russia-Ukraine conflict. In this context, the primary source of spillovers is the clean energy sector. Our results confirm the growing “Environmental Social Governance” market sentiment influencing financial markets favoring the transition towards more sustainable energy financial products. Finally, our study could provide valuable insights for investors and policymakers interested in financial products or decisions in line with this transition.

ABSTRACT. We propose a general framework for testing the significance of parameters in large panels of multiple linear regression models, focusing on mean-variance spanning (MVS) tests. The proposed methodology is versatile and applicable even when the number of equations is large, requiring only stationary data, and allows the number of regressors to grow asymptotically toward the sample size. Monte Carlo simulations demonstrate that the testing procedure maintains correct size and power, even when residuals exhibit asymmetry, fat-tails, serial correlation, and GARCH effects, outperforming existing methods. We apply the methodology to assess whether including blue-chip stocks from the U.S., Europe, and Switzerland enhances each country's domestic mean-variance efficient frontier. The findings suggest that the benefits of international diversification depend on economic conditions and vary by country, with the rejection of the MVS hypothesis linked to variance reduction within the domestic global minimum-variance portfolios.

ABSTRACT. We derive local (over a small time interval ∆) Edgeworth-like expansions of the bivariate conditional characteristic function of the level/volatility ∆-increments of a semi-martingale. We do so without and with compound Poisson discontinuities (idiosyncratic and joint) in levels and volatility. The first order (in adds skewness to the bivariate Gaussian characteristic function through the time- varying correlation between the level/volatility changes and the volatility/volatility of volatility changes. The second order (in sqrt(∆)) of the expansion adjusts kurtosis through, e.g., the volatility of volatility and the volatility of the volatility of volatility. We show how recovery of the conditional density by virtue of Fourier-inversion is intimately related to the differentiability of the assumed process, which we formalize through the notion of W -differentiability, where W defines the leading Brownian motion. We discuss characteristic function-based nonparametric estimation of all processes driving dynamics.

ABSTRACT. Understanding how US monetary policy affects global economic conditions is of fundamental importance, as the literature shows that the effects are substantial. In this paper, I provide novel evidence on how income inequality shapes the heterogeneity of US monetary policy spillovers to GDP across foreign economies. I employ state-dependent local projections and exploit variation in disposable income inequality across a panel of 87 countries over the period 1966-2020. The empirical findings suggest that household heterogeneity significantly influences how foreign GDP responds to a US monetary policy tightening. GDP contracts two to three times more when inequality is above average. However, while higher inequality amplifies negative spillovers in advanced economies, it mitigates them in emerging markets. To rationalise this finding, I use a three-country open economy Two-Agent New Keynesian (TANK) model, which suggests that this divergence is driven by differences in participation in international financial markets. Households in emerging market economies face higher barriers to investing internationally, limiting their ability to re-balance portfolios towards higher-return bonds after the shock. This, in turn, reduces the macroeconomic effect of higher inequality on domestic conditions.

ABSTRACT. We estimate a time-varying parameter vector autoregression to examine the evolution of international spillovers of U.S. monetary policy in light of increasing globalization in real and financial markets. We find that the adverse international effects of a U.S. tightening have substantially increased over the past three decades, peaking during the Great Recession. Based on a cross-country analysis and counterfactual simulations, we argue that such amplification can primarily be attributed to the surge in trade integration, while the role of rising financial integration in explaining the time-variation is limited.

ABSTRACT. Shedding light on the dynamics of international government bond yields is essential for mastering global financial and macroeconomic stability. This study provides new empirical evidence on the complex interplay between global macro trends, sovereign credit risk, and the cross-border effects of monetary policy in an International Asset Pricing framework. By constructing a novel high-frequency database for the period 1970-2023, I disentangle the contributions of global and country-specific factors to yield movements across 20 developed economies. The empirical findings confirm that global inflation and real interest rate trends significantly affect yield dynamics. Safer countries closely align to global Macro factors, while the higher-risk economies exhibit greater deviations mainly due to country-specific credit risk. Employing this decomposition, I document the asymmetric international transmission monetary policy. I focus on the spillovers of unconventional measures - particularly quantitative easing - for three Central Banks. U.S. Federal Reserve actions show the most pronounced global effects. In contrast, the ECB’s unconventional policies display heterogeneous effect among Euro countries, with some weaker spillovers outside the Euro Area and Europe. Lastly, the Bank of England’s impacts remain largely domestic.

ABSTRACT. This paper investigates the potentially non-linear transmission of U.S. monetary policy surprises given the intensity of carry trade activity in currency markets. A threshold vector autoregressive model discriminates between different regimes of speculative carry trade activity in a set of developed and emerging market currencies against the U.S. dollar. Overall, in the course of monetary policy tightening, the U.S. dollar appreciates in response to a conventional monetary policy shock but depreciates to a growth shock. Across regimes, the transmission of monetary policy adjusts through the interest rate differential and less through exchange rates. U.S. monetary policy shocks generate higher excess returns when carry trade intensity is elevated. These effects are not driven by risk aversion but through speculative trading. A currency trading strategy exploiting the information on central bank announcement days substantially outperforms the carry trade in terms of the Sharpe ratio and downside risk.

ABSTRACT. We study the propagation of resource productivity shocks in a large-scale dynamic production network model. Using production network data from the UK, we examine the implications of policy instruments that target greater material recovery on the propagation of total factor productivity (TFP) shocks. We find that highly connected sectors benefit more from an increase in productivity. Furthermore, with the introduction of a subsidy program targeting businesses in the manufacturing sector, these businesses commit to transforming their production cycles towards more efficient and sustainable material flow systems. A positive TFP shock leads to large output gains relative to the baseline model and generates positive spillover effects on non-targeted sectors.

ABSTRACT. We revisit the estimation of the welfare costs of inflation originating from lack of liquidity satiation for 11 low-inflation and 5 high-inflation countries, and for Weimar Republic’s hyperinflation. Our evidence suggests that, contrary to the implicit assumption in much of the literature, these costs are far from negligible. For the U.S. our point estimates are equal to about one-third of those computed by Lucas (2000), and an order of magnitude larger than those obtained by Ireland (2009). Crucially, the most empirically plausible moneydemand functional form points towards sizeable ‘upward risks’ for these costs, with the 90% confidence interval associated with a 4% nominal interest rate stretching beyond 0.5 per cent of GDP. The welfare costs of inflation in the Euro area are about twice as large as in the U.S., thus suggesting that, ceteris paribus, the inflation target should be materially lower. At the peak of the inflation episodes, welfare costs had ranged between 0.3 and 1.9 per cent of GDP for low-inflation countries; between 4 and nearly 7 per cent for highinflation ones; and between 26 and 36 per cent for Weimar’s hyperinflation.

ABSTRACT. We measure the Euro Area (EA) output gap and potential output using a non-stationary dynamic factor model estimated on a large dataset of macroeconomic and financial variables. From 2012 to 2023, we estimate that the EA economy was tighter than the European Commission and the International Monetary Fund estimate, suggesting that the slow EA growth is the result of a potential output issue, not a business cycle issue. Moreover, we find that credit indicators are crucial for pinning down the output gap, as excluding them leads to estimating a lower output gap in periods of debt build-up and a higher gap in periods of deleveraging.

ABSTRACT. Utilizing critical recent data advances, we analyze empirical evidence on long-run samples of short-maturity real interest rates as well as term spreads based on multi-century data. In contrast to an extensive literature on short-maturity real interest rates over the past few decades, we find strong and consistent evidence of trend stationarity in long horizon series, relatively fast adjustment speeds, and a paucity of structural breaks -- results that we show to survive out of sample tests. The use of very long-run data offers a fresh perspective for ongoing monetary policy debates surrounding r*, and also provides a crucial missing link to reconstructing the long-run properties of term spreads. On balance and against limited post-COVID data, our evidence suggests caution on the idea of a break in short-term real interest rate behavior and instead points to elements of continuity over very long time periods. Relatedly, we show that term spreads are secularly rising while inflation volatility trends in the exact opposite direction -- a finding questioning the emphasis of influential term structure models.

ABSTRACT. Do people learn from experience how to cope with weather shocks? We use a unique four-wave panel household dataset from Uganda, merged with granular historical weather records, to understand the nexus between experience, weather shocks, and agricultural performances. Our identification strategy exploits cross-sectional variation in the climate experience of immigrant members of the households and the temporal variation in the realization of the weather shocks during the survey years. We show that, although temperature shocks may be detrimental to agricultural performance, households with more experience perform differentially better. An additional 10 days of temperature shocks reduce the income of households with little experience by 11 percent, while the effects are negligible for those with higher-than-average experience. Our findings are robust to placebo tests on the timing of shocks and to falsification tests. We further document that the differential effect on performance is independent of an unrelated experience variable capturing households' familiarity with soil characteristics. Suggestive evidence points towards adopting risk-reducing technologies as the driving factor behind the gains of the more experienced households. These findings highlight the relevance of initiatives promoting experiential learning.

ABSTRACT. We study the forecasting heuristics MBA textbooks propose to make firms' plans. Using Duke-Survey data, we document the prevalence and heterogeneity of heuristics use among top US executives. We propose a theory of optimal multidimensional forecasting; evaluate heuristics in terms of forecast coherence; develop tests to distinguish coherence from accuracy. In our normative benchmark, technology parameters coherently link output-inputs forecasts. Our positive model rationalizes some heuristics as second-best responses to noisy signals, yielding a pecking order of heuristics. Consistent with our predictions, firm performance is negatively associated with incoherence and incoherent heuristics use. About one-half of CFOs make incoherent forecasts.

ABSTRACT. Using a Python script to extract data from the SEC website, we collect different types of debt covenants for a panel of US Compustat firms covering the period from 1995q1 to 2020q3. Our goal is to investigate how the evolution of the amount and type of covenants has affected the heterogeneity of corporate capital structure behavior. Specifically, we compare maintenance capital covenants, which require continuous compliance with capital structure restrictions, maintenance performance covenants, which could lead to the transfer of control rights to creditors, and incurrence limitation covenants, which activate predefined restrictions on the borrower's actions once the covenant threshold is exceeded. We use fully heterogeneous panel estimates of key parameters such as speed of adjustment, free cash flow, and debt maturity substitutability, and resort to meta-analysis to condense the estimated results as a function of different covenant types, Bankruptcy Abuse Prevention Act, financial crisis, and firm characteristics.

ABSTRACT. This paper studies the under-reporting of workplace injuries in Italy, leveraging unique administrative data on work accidents. Using a difference-in-differences approach, I analyze injury reporting behavior across provincial economic sectors exposed to the news of a fatal workplace accident compared to those without. The results show a significant increase in reporting of non-severe injuries in the weeks following such news, suggesting that many injuries were previously unreported. Two mechanisms drive this pattern: media coverage, which likely fosters transparency and accountability of firms, and union presence, with a decomposition of the ATT revealing stronger effects in provinces with higher union membership.

ABSTRACT. Using a unique dataset of equity offerings from crowdfunding platforms, we explore the synergy between human insights and algorithmic analysis in evaluating campaign success through business plan assessments. Human evaluators (students) used a predefined grid to assess each proposal in a Business Plan competition. We then developed a classifier with advanced textual representations and compared prediction errors between human evaluators, a machine learning model, and their combination. Our goal is to identify the drivers of discrepancies in their evaluations. While AI models outperform humans in overall accuracy, human evaluations offer valuable insights, especially in areas requiring subtle judgment. Combining human and AI predictions leads to improved performance, highlighting the complementary strengths of human intuition and AI's computational power.

ABSTRACT. This paper examines the effects of part-time job policies on bridge activities into retirement and pension claiming, using the 2003 German Mini-job reform as a natural experiment. Leveraging a cohort difference-in-differences approach, the study reveals unexpected negative consequences. The reform did not significantly reduce unemployment, sickness, or inactivity among older workers. However, it increased transitions from regular to marginal employment without social security contributions, acting as a stepping stone to retirement, and raised the likelihood of pension claims among older workers. Financial incentives alone cannot explain these results, as older workers’ flexibility preferences might play a role. Heterogeneous responses are found by gender, income, and employment history. These findings offer insights into the interplay between policy design and retirement transitions.

ABSTRACT. This paper provides new evidence that public investments in education, healthcare and public welfare significantly influence individual time allocation. We analyze the United States care regime from 2004 to 2021 where the latter is subject, as a public expenditure component, to positive and negative phases, or events. By distinguishing between expansionary and recessionary events, we study the individual time allocation as affected by alternative phases of the regime of care using a diff-in-diff event study design that exploits that extreme public spending investments occur in spikes. Using representative samples from the American Time Use Survey and the US State \& Local Government Expenditure databases, we find that positive spikes in state and local expenditures cause increases in the waged working time for women and reduce the burden of domestic and care responsibilities. Conversely, period of fiscal austerity result in a gendered shift from paid to unpaid activities. These effects are large and highlight the role of institutional factors in shaping gender time inequalities and the need for gender budgeting at both recessionary and expansionary stages of the fiscal budget cycle.

ABSTRACT. Dynamic Stochastic General Equilibrium (DSGE) models are the main tool used in Academia and in Central Banks to evaluate the business cycle for policy and forecasting analyses. Despite recent advances in improving the fit of DSGE models to the data, the misspecification issue remains.

In this paper, we deal with a specific aspect of the misspecification: the statistical representation of DSGE models. In particular, we discuss the case of DSGE models with a Vector Autoregressive Moving Average (VARMA) representation as Data Generation Process. We solve several DSGE models (Small-Scale Monetary Policy. We feed VARs and Local Projections with artificial data (output, inflation, and short-term interest rate) with different lag lengths, providing a discussion about the statistical representation of these models.

ABSTRACT. A central question in economics is whether technological innovations complement or substitute workers’ skills, thus enhancing or replacing workers’ productivity. However, occupational output is often hard to measure, especially for high-skilled workers performing abstract tasks, making it hard to answer this question. In this paper, we focus on the effect of technology on productivity and inequality within a specific high-skilled group, that of researchers in economics, for whom we measure research output. Specifically, we study the effect of the introduction of DYNARE, a software designed to solve and simulate dynamic stochastic general equilibrium (DSGE) models. We first develop a dynamic model of research and citation accumulation, in which the arrival of the technology allows some researchers to perform more easily a subset of the tasks needed to write academic papers. Next, we test the model’s implications by leveraging quasi-experimental variation in DYNARE adoption across fields. We implement a difference-in-differences strategy, finding a significant increase in the average number of publications. Consistent with the predictions of the model, the increase in publication is driven by less productive scholars, thus suggesting that the new technology could have led to a decrease in citation inequality.

ABSTRACT. This paper introduces a novel dynamic factor model designed to capture common locally explosive episodes, also known as common bubbles, within large-dimensional, potentially non-stationary time series. The model leverages a lower-dimensional set of common unobserved factors exhibiting locally explosive behavior to identify common extreme events. Modeling these explosive behaviors allows to predict systemic risk and test for the emergence of common bubbles. The dynamics of the explosive factors are modeled using mixed causal non-causal models, a class of heavy-tailed autoregressive models that allow processes to depend on their future values through a lead polynomial. The paper establishes the asymptotic properties of the model and provides sufficient conditions for consistency of the estimated factors and parameters. A Monte Carlo simulation confirms the good finite sample properties of the estimator, while an empirical analysis highlights its practical effectiveness. Specifically, the model accurately identifies the common explosive component in monthly stock prices of NASDAQ-listed energy companies during the financial crisis in 2008 and predicts its evolution significantly outperforming alternative forecasting methods.

ABSTRACT. Common wisdom suggests that investment in aerospace can lead to new space discoveries and to a variety of spillovers in the real economy on Earth. This paper tries to quantify this proposition by studying the macroeconomic effects of aerospace structural shocks. We build and estimate a macroeconomic model with endogenous growth in which the aerospace industry is explicitly included. The model recovers the structural relationship between aerospace investment and the technology spillover, which supports persistent economic growth. We find that aerospace missions provided significant growth spillovers. The analysis reveals that spillovers are larger during the public aerospace initiatives of the 60s and 70s compared to those associated with private sector activities in subsequent decades. Extensive experiments quantify the economic relevance of the estimation results.

ABSTRACT. How do inflation expectations matter for the wage bargaining of men and women? Using experimental data, I show that women associate rising inflation more with high unemployment and lower job finding probabilities, which reduces their wage bargaining relative to men causally. I estimate the elasticity of job-finding beliefs in response to inflation and calibrate a New Keynesian Search and Match model with male and female workers accordingly to quantify the role of belief frictions on the cyclicality of the gender wage gap. The model can replicate the cyclical properties of the gender wage gap in the data.

ABSTRACT. The paper proposes the Sequential Monte Carlo and its online variant for the estimation of multivariate Garch models that features time-varying skewness and kurtosis. Relative to model-specific Markov Chain Monte Carlo, Sequential Monte Carlo has the advantages of generality, parallelizability, and speed. Moreover, it gives as output the marginal likelihood useful in model selection. In the empirical appli- cation, we revisit the forecastability of the exchange rate through an asset allocation problem, using different specification of multivariate volatility models.

ABSTRACT. When in proxy-SVARs the covariance matrix of VAR disturbances is subject to exogenous, permanent, non-recurring breaks that cause the target impulse response functions (IRFs) to change across volatility regimes, even strong, exogenous external instruments can yield inconsistent estimates of the dynamic causal effects. However, if these volatility shifts are properly incorporated into the analysis through \textquotedblleft stability restrictions\textquotedblright , we demonstrate that, under a necessary and sufficient rank condition, the target IRFs are point-identified and estimated consistently. Notably, if the shifts in volatility are sufficiently informative, standard asymptotic inference remains valid even with (i) local-to-zero covariance between the proxies and the instrumented structural shocks, as in \cite{StaigerStock1997}, and (ii) potential failures of instrument exogeneity. Intuitively, under the rank condition, shifts in volatility act similarly to strong instruments that are correlated with both the target and non-target shocks. We illustrate the effectiveness of our approach by revisiting a seminal fiscal proxy-SVAR model from the existing literature. Specifically, we detect a sharp change in the size of the US tax multiplier when inference in the fiscal proxy-SVAR, based on a narrative tax instrument, accounts for the decline in unconditional volatility observed during the transition from the Great Inflation to the Great Moderation regimes. In this framework, the narrative tax instrument remains informative about the effects of the tax shock in both macroeconomic regimes, even though our empirical analysis raises concerns about its validity.

ABSTRACT. We establish the identification of a specific shock in a structural vector autoregressive model under the assumption that this shock is independent of the other shocks in the system, without requiring the latter shocks to be mutually independent, unlike the typical assumptions in the independent component analysis literature. The shock of interest can be either non-Gaussian or Gaussian, but, in the latter case, the other shocks must be jointly non-Gaussian. We formally prove the global identification of the shock and the associated column of the impact multiplier matrix, and discuss parameter estimation by maximum likelihood. We conduct a detailed Monte Carlo simulation to illustrate the finite sample behavior of our identification and estimation procedure. Finally, we estimate the dynamic effect of a contraction in economic activity on some measures of economic policy uncertainty.

ABSTRACT. In empirical applications, results from structural VARs (typically, impulse responses) may lack accuracy, especially in small samples. This issue is particularly evident in macroeconomic applications, since the data are typically observed at low frequencies and their time span is often limited. Shrinkage techniques have been used for a long time in VAR modelling to improve forecast accuracy, but their application to structural inference has not been studied so far. In this paper, we investigate the effects of various $\ell_1$ shrinkage techniques with a view to ascertaining whether they can improve estimation accuracy for the quantities of interest in structural modelling. We perform two simulation exercises on two well-known SVAR applications and we find evidence that shrinkage can improve estimation accuracy for the IRFs, especially when the sample is small.

ABSTRACT. We provide precise conditions for nonparametric identification of causal effects by high-frequency event study regressions, which have been used widely in the recent macroeconomics, financial eco- nomics and political economy literatures. The high-frequency event study method regresses changes in an outcome variable on a measure of unexpected changes in a policy variable in a narrow time window around an event or a policy announcement (e.g., a 30-minute window around an FOMC an- nouncement). We show that, contrary to popular belief, the narrow size of the window is not sufficient for identification. Rather, the population regression coefficient identifies a causal estimand when (i) the effect of the policy shock on the outcome does not depend on the other shocks (separability) and (ii) the surprise component of the news or event dominates all other shocks that are present in the event window (relative exogeneity). Technically, the latter condition requires the ratio between the variance of the policy shock and that of the other variables to be infinite in the event window. Under these conditions, we establish the causal meaning of the event study estimand corresponding to the regression coefficient and the consistency and asymptotic normality of the event study estimator. Notably, this standard linear regression estimator is robust to general forms of nonlinearity. We apply our results to Nakamura and Steinsson’s (2018a) analysis of the real economic effects of monetary policy, providing a simple empirical procedure to analyze the extent to which the standard event study estimator adequately estimates causal effects of interest.

ABSTRACT. In this paper we establish asymptotic theory for a spatial autoregressive (SAR) model where the network structure is possibly endogenous and the spatial parameter can be greater than or equal to unity. The setup is general enough to include as special cases the random walk with a drift model, the local to unit root model (LUR) with a drift and the model for moderate integration with a drift. The control function approach is adopted in dealing with the issue of endogeneity. Consistency of the two stage least squares estimator is proven and its asymptotic distribution is derived. Simulations verify the analytical results and we include a small empirical application to support our findings.

ABSTRACT. This paper argues that cross-sectional dependence is an indicator of misspecification in panel quantile regression rather than just a nuisance that may be accounted for with panel-robust standard errors. This motivates the development of a novel test for panel quantile regression misspecification based on detecting cross-sectional dependence. The test possesses a standard normal limiting distribution under joint N,T asymptotics with restrictions on the relative rate at which N and T go to infinity. A finite-sample correction improves the applicability of the test for panels with larger N. An empirical application to housing markets illustrates the use of the proposed cross-sectional dependence test.

ABSTRACT. This paper extends the Common Correlated Effects (CCE) approach of Pesaran (2006) to semiparametric panel data models with cross-sectional dependence. Unlike the original model, we allow common factors to enter nonparametrically while maintaining linearity in individual-specific regressors. Considering heterogeneous slope coefficients, we propose sieve-based estimators for the nonparametric component when both the number of time periods T and cross-sectional units N is large. The proposed method enables consistent and asymptotically normal estimation in interactive fixed effects models. Monte Carlo simulations demonstrate good finite-sample performance of the estimator.

ABSTRACT. We examine the contribution of large-scale national health service to public health and human capital formation. To this end, we exploit the establishment of the large-scale National Health Service in Cuba in 1960 as a source of variation in long-term public health and human capital formation. Our identification strategy exploits for non-parallel outcome prior to the introduction of universal health care provision to estimate the missing counterfactual scenario. By comparing Cuba with a plausible donor pool of former European colonies in Western Hemisphere that have not introduced a comprehensive universal health care, we estimate counterfactual trajectories of infant mortality, life expectancy and human capital for the period 1870-2022. Our results show that compared to the synthetic counterfactual, the introduction of equitable and accessible public health care provision is associated with a permanent decrease in infant mortality by around 35 percent, sustained improvement in life expectancy at birth by two years and with 1.4 years increment in the average length of schooling. The estimated effects of survive an extensive battery of placebo analyses and do not seem to fizzle out up to the present day.

ABSTRACT. We investigate the trade-off between variable selection and shrinkage in designing optimal parametric portfolios based on a large set of firm characteristics. Using a flexible Bayesian prior, we show that greater emphasis on sparsity reduces model uncertainty but results in high-turnover, under-diversified portfolios with larger exposures to fewer characteristics. In contrast, a more shrinkage-inducing prior fosters better diversification across a broader set of characteristics, resulting in portfolios more aligned with out-of-sample mean-variance efficiency, particularly when transaction costs are considered. Our findings caution against over-reliance on sparsity-inducing tools for predicting the cross-section of stock returns using firm characteristics.

ABSTRACT. Recent studies show that the zero-process of observed intraday financial returns is frequently characterised by non-stationary periodicity. As liquidity varies across the trading day, not only does unconditional volatility change, but also the unconditional zero-probability. While scaling returns by the time-varying intraday volatility may stabilise volatility, it does not make the zero-process of scaled returns stationary. This invalidates standard methods of risk estimation, and recent studies document that the use of such invalid methods can have major effects on risk estimates. Formal tests for non-stationary periodicity in the zero-process can therefore be of great value in guiding the choice of a suitable risk estimation procedure. Despite this, little attention has been devoted to the derivation of such tests. Here, we help filling this gap by developing user-friendly yet flexible and powerful tests that hold under mild assumptions. Next, an empirical study reveals that intraday financial returns are widely characterised by non-stationary periodicity in the zeroprocess. This has important and potentially wide-ranging implications for future research.

ABSTRACT. We identify a sector structure within the digital asset market, where different types of digital assets (the “digital asset sectors”) exhibit different risk and return characteristics. To the best of our knowledge, this study provides the first comprehensive analysis of the sector structure within the digital asset market. We examine the observed sectoral variation through two channels: the systematic risk channel and the idiosyncratic risk channel. Our findings indicate that although sectoral differences exist, they do not translate into variations in sector-level beta exposures. Instead, sector-specific information emerges through the idiosyncratic risk channel. The sector risk factor, which captures this sector-specific information, exhibits significant variability. Further analysis reveals that such sectoral differences are driven by sector-specific events, sector momentum, and inter-sector spillovers.

ABSTRACT. Gaussian graphical models have become a staple in statistical modelling and for estimating partial correlation networks. This article extends the baseline approach to a time series framework by introducing temporal dependence for the entries of the precision matrix. In order to conduct statistical inference, a fully Bayesian approach is adopted, relying on global-local shrinkage priors to deal with high-dimensional data and mitigate (temporal) overfitting. The framework has several special cases of interest, including a variant of the standard Bayesian graphical lasso (Wang, 2012). Closed-form recursions for the filtering and smoothing distributions are obtained. These results are exploited to design a simple yet efficient blocked Gibbs sampler for posterior inference. An interweaving strategy is applied to enhance the mixing of the sampler. As a by-product, the proposed method allows for estimating a time series of (sequentially dependent) networks from partial correlations among the variables in the system.

Using synthetic and real data, the performance of the model is investigated in comparison to standard Bayesian and frequentist benchmarks in terms of both covariance matrix estimation and graphical structure learning.

ABSTRACT. Penalised Regression methods and in particular the Least Absolute Shrinkage and Selection Operator (LASSO) have become an integral part of modern-day time series analysis. As the performance of LASSO crucially hinges on the assumption of sparsity, which is unknown in practice, we propose a Hausman type test for the latter. The null hypothesis is that there are at most $k_0$ relevant, non-zero regressors. The key difference between our test and existing approaches like Information Criteria is that we allow the number of regressors to be (much) larger than the sample size. We provide two major applications for our test: The first one is to the impact of parameter estimation error in tests for out-of-sample forecast comparison \citep[e.g.,][]{DM1995}. The second application concerns inference on Impulse Response Functions in (Structural) Vector Autoregressions.

ABSTRACT. Latent Space (LS) network models project the nodes of a network on a d-dimensional latent space to achieve dimensionality reduction of the network while preserving its relevant features. Inference is often carried out within a Markov Chain Monte Carlo (MCMC) framework. Nonetheless, it is well-known that the computational time for this set of models increases quadratically with the number of nodes. In this work, we build on the Random-Scan (RS) approach to propose an MCMC strategy that alleviates the computational burden for LS models while maintaining the benefits of a general-purpose technique. We call this novel strategy Multiple RS (MRS). This strategy is effective in reducing the computational cost by a factor without severe consequences on the MCMC draws. Moreover, we introduce a novel adaptation strategy that consists of a probabilistic update of the set of latent coordinates of each node. Our Adaptive MRS adapts the acceptance rate of the Metropolis step to adjust the probability of updating the latent coordinates. We show via simulation that the Adaptive MRS approach performs better than MRS in terms of mixing. Finally, we apply our algorithm to face-to-face interaction data and to climate-related Facebook data. We show how our adaptive strategy may be beneficial to empirical network applications.

ABSTRACT. We propose a multivariate generalization of the square difference estimator of Von Neumann (1941) to infer the residual variance in a d-dimensional nonlinear regression. Our approach employs the minimum spanning tree as a criterion for selecting the point pairs along which the residual variance is estimated, and applies jackknife resampling to correct the leading bias that arises in dimensions d > 3. We establish general conditions on the class of admissible regression functions ensuring square-root consistency of the proposed estimator for dimensions d > 3. Our results hold for very general geometric configurations of the design points. Using this inferential framework, we test whether a candidate parametric model coincides with the true regression function. Asymptotically, our test identifies with arbitrarily large confidence the true model and rejects a misspecified model with probability approaching one. The test overcomes a well-known drawback of existing selection criteria, which consistently identify the true model conditionally on knowing, a priori, that the latter is included in the set of candidate models.

ABSTRACT. We introduce a procedure involving a new business cycle indicator that exploits a large information set (848 time series) to nowcast the GDP evolution in real time. The proposed indicator, called SONG, focuses on the binary event represented by the sign of the one-step-ahead GDP growth rate. We use many monthly indicators to “rate” the next-period economic performance represented as the GDP direction of change. The proposed methodology comprises three steps: selecting the indicators, predicting the GDP sign based on single indicators, and aggregating the single signals. The first step relies on using the Directional Accuracy Change (DAC) test, Receiver Operating Characteristic (ROC), and spectral coherence (SpCoh) together. In the second step, the probability of the event ”sign of the GDP change”, delivered by each selected indicator, is estimated by either a bivariate logit model or the binary point prediction based on the ROC. Finally, in the aggregation step we adopt alternative weighting schemes. The methodology’s performance was tested by predicting the Italian GDP q-o-q directional changes in pseudo-real time from Q2-2014 up to Q3-2022. The results are compared with the traditional benchmark models used to forecast the GDP, showing a better performance of SONG.

ABSTRACT. We develop a novel multivariate semi-parametric framework for joint portfolio Value-at-Risk (VaR) and Expected Shortfall (ES) forecasting. Unlike existing univariate semi-parametric approaches, the proposed framework explicitly models the dependence structure among portfolio asset returns through a dynamic conditional correlation (DCC) parameterization. To estimate the model, a two-step procedure based on the minimization of a strictly consistent VaR and ES joint loss function is employed. This procedure allows to simultaneously estimate the DCC parameters and the portfolio risk factors. The performance of the proposed model in risk forecasting on various probability levels is evaluated by means of a forecasting study on the components of the Dow Jones index for an out-of-sample period from December 2016 to September 2021. The empirical results support effectiveness of the proposed framework compared to a variety of existing approaches.

ABSTRACT. It is also known that there is a need to monitor the trend of public sector debt with much greater frequency than that permitted by the availability of annual data. One of the possible ways of satisfying this need is to use the data collected on a monthly basis on the main income and expenditure items of public administrations. This paper aims to verify whether, through consolidated Temporal Disaggregation techniques in combination with a seasonal adjustment procedure, it is possible to obtain nowcast of the main public debt components.

ABSTRACT. We tackle the issue of producing point, sign, and density forecasts for the monthly real price of carbon within the European carbon market, EU ETS. We show that a Bayesian Vector Autoregressive (BVAR) model, augmented with factors based on economic fundamentals, provides accuracy gains over a set of benchmark forecasts in both point and density forecast. We also provide a qualitative comparison of model-based forecasts with survey expectations and forecasts released by data providers. Moreover, we consider verified emissions and demonstrate that adding stochastic volatility can further improve the forecasting performance of a single-factor BVAR model. Lastly, we rely on forecasts to build market monitoring tools that track demand and price pressure in the EU ETS.

ABSTRACT. This paper presents a high-frequency structural VAR framework for identifying oil price shocks and examining their uncertainty transmission in the U.S. macroeconomy and financial markets. Leveraging the stylized features of financial data — specifically, volatility clustering effectively captured by a GARCH model — this approach achieves global identification of shocks while allowing for volatility spillovers across them. Findings reveal that increased variance in aggregate demand shocks increases the oil-equity price covariance, while precautionary demand shocks, triggering heightened investor risk aversion, significantly diminish this covariance. A real-time forecast error variance decomposition further highlights that oil supply uncertainty was the primary source of oil price forecast uncertainty from late March to early May 2020, yet it contributed minimally during the 2022 Russian invasion of Ukraine.

ABSTRACT. This work investigates whether EU cohesion policies aiming at environmental improvement and carbon reduction have an economic impact on adopters. We look at the changes in firms’ performance due to the sustainability-oriented technologies financed by the European cohesion funds during the 2007-13 programming period. We include firms that participated in pilot programs and received public incentives to upgrade their production plants with sustainable technologies, and we use Supervised Machine Learning (ML) algorithms to identify the most appropriate counterfactuals. Our results indicate a strong and positive effect on firms' profitability, for which the public policy is directly responsible, with different dynamics for different levels of public support and implementation. Specifically, the relationship between levels of public aid received and firms’ operating margins presents an inverted "U" shape. Additionally, in the short run, the effect on treated firms tends to diminish, suggesting the possibility of a rebound effect where the gains in production efficiency and energy savings are repurposed by firms to increase production (and profits) instead of reducing absolute emissions. This is perfectly in line with what one can expect from an economic actor at the micro-level: firms’ actions are certainly guided by the search for ways to obtain profit increases.

ABSTRACT. This paper aims to investigate the relationship between different sources of risk related to energy, that is, returns on the energy sector, energy uncertainty, and geopolitical risk. To this aim, we provide a parsimonious and flexible model for extracting common volatility factors (COVOL) from a cross-section of assets. We assume there are groups of assets with different exposure levels across the groups and similar levels within each group. The membership of the assets to the groups is unknown, which naturally calls for using stochastic partition models. The latent factors have a gamma autoregressive structure, which allows for persistence. We provide some theoretical properties of the new clustered COVOL model, a Bayesian inference procedure well suited for latent variable models, and an empirical analysis of the volatility transmission in a multi-country perspective.

ABSTRACT. This paper considers estimation of the moments and the distribution of heterogeneous marginal effects using panel data. We impose no restrictions on the form or dimension of time-invariant heterogeneity. In this setting, we identify the mean, variance, higher-order moments, and the distribution of marginal effects using two periods of data. We propose simple nonparametric estimators for the moments and the distribution, and study their asymptotic properties. The moment estimators are consistent and asymptotically normal. For the distribution estimator, we establish consistency by developing novel results that connect the convergence of distributions to the convergence of their moments. We illustrate the methodology with an application to Engel curves for food at home. Our analysis of variance, higher moments, and the distribution of marginal effects reveals significant heterogeneity. In particular, some households have upward-sloping sections in their Engel curves for lower values of expenditures. In contrast, the average Engel curve is downward-sloping for all expenditure values, in line with the previous literature.

ABSTRACT. Learning about the tail features of unobserved heterogeneity is a complex yet essential task in econometrics, with high relevance for empirical policy analysis. Many policy questions focus on units in the tails—such as identifying the most effective teachers or schools—yet these units are not directly targeted in standard analyses. This paper introduces a novel method for ranking and making inferences on fixed effects specifically within the tails of the distribution of unobserved heterogeneity. Our approach differs from popular nonparametric empirical Bayesian methods in two key ways. First, rather than shrinking the fixed effect estimator, our method expands it in the tails. Second, we avoid imposing parametric assumptions on the error term and allow for dependence between the error term and the important fixed effect term. Applying this method to assess the effectiveness of top- and bottom-ranked schools, we find that our estimates notably diverge from those derived from existing methods, underscoring the value of this new approach.

ABSTRACT. When neighborhood characteristics or shocks are omitted, local unobservables hinder the identification of causal effects. However, if unobservables are smooth over space and units are paired according to proximity, a neighborhood data transformation can effectively rule them out. This paper studies neighborhood differencing and within-neighborhood estimation strategies in a finite population framework. We establish their asymptotic distribution and provide guidance on standard errors adjustment for within and between neighborhood correlations. We also develop a test for smooth fixed-effects, allowing practitioners to select the optimal threshold for data transformation. We examine the behavior of the proposed tools through simulations, and illustrate their usefulness using data from a clustered randomized experiment.

ABSTRACT. This paper develops a novel nonparametric framework for estimating production frontiers and efficiency measures in the presence of endogeneity and environmental variables. Building on existing frontier estimation methods such as Data Envelopment Analysis (DEA) and Free Disposal Hull (FDH), we address key challenges like nonseparability, high-dimensional data, and sensitivity to outliers. Using a panel data setting, the proposed approach introduces a method to adjust inputs and outputs by removing the influence of external factors through a control function methodology. This adjustment ensures unbiased efficiency estimation even in the presence of endogenous environmental variables. The empirical application focuses on foreign direct investment (FDI), a well-established endogenous factor in the economic literature, and its impact on production efficiency over time. Analyzing data from 44 countries over 38 years, we find that FDI positively influences efficiency, particularly after surpassing a certain threshold, while the impact of time diminishes with higher FDI levels. These findings align with the hypothesis that FDI generates significant efficiency gains by reducing adjustment costs and time delays. Our framework offers a robust tool for assessing firm performance under complex economic conditions and environmental dependencies, with particular applicability to scenarios involving endogenous variables such as FDI.

ABSTRACT. In the conventional semimartingale representation of the log-price of a financial security, the volatility of daily log-returns is consistently estimated by their ex-post quadratic variation. A crucial assumption underlying this result is that the drift dynamics are negligible or pre-determined given past information. Recent empirical evidence of intraday unanticipated variations of mean returns call into doubt the validity of this assumption. We propose a semiparametric estimator of the daily return volatility incorporating the effect of stochastic intraday dynamics of the drift process. This framework allows us to test for the presence of drift dynamics in the data and to assess their impact on the daily return volatility. Our empirical analysis shows three main results: (i) there exists compelling evidence of intraday drift dynamics across various asset classes; (ii) when the drift moves significantly, our measure of volatility provides a better description of the return distribution; (iii) the component of the return variance associated with drift dynamics is non-negligible and possesses predictive power.

ABSTRACT. This paper explores market liquidity, focusing on co-liquidity and illiquidity, which are critical for both market participants and policymakers. It addresses the importance of extreme illiquidity events and their risks, particularly in high-frequency trading environments. Using high-frequency data on returns and volume, the study builds on a simple structural model and a nonparametric approach inspired by Lacava et al. (2023), linking market volatility, volume, and liquidity. The paper refines liquidity measurement by capturing price changes and trading volume dynamics, with a focus on illiquidity due to market frictions. It extends the framework to examine the temporal dynamics of co-(il)liquidity across various assets, providing new metrics to quantify liquidity contagion and systemic risks. This methodology bridges theory with empirical applications, offering practical tools for assessing and managing liquidity risks in diverse financial markets.

ABSTRACT. Empirical finance researchers and professional portfolio managers have long considered strategies that exploit individual stock characteristics such as value and momentum, but recently they have shifted their attention to others. Our empirical goal is to assess if portfolios that exploit the profitability and investment characteristics really provide new opportunities for investors by testing whether they can be spanned by dynamic portfolios of traditional strategies. Given that the optimal weights of these dynamic portfolios depend in an unrestricted manner on predictors such as yield spreads, we develop a new econometric methodology called sieve managed portfolios that can handle non-parametric weights. We find that the two new strategies do not increase the conditional Sharpe ratio that can achieved from traditional strategies, but they are relevant for investors that want to exploit the changing investment opportunities that our predictors generate.

ABSTRACT. This study investigates whether (and how) working from home (WFH) affects the gender division of parental unpaid labor. I use the recent COVID-19 pandemic that brought an unanticipated yet lasting shift to WFH combined with a measure of occupational WFH feasibility (Alipour et al. 2023) as a quasi-experiment to employ an instrumental variable (IV) approach and estimate causal effects. I use unique longitudinal data from the “Growing up in Germany” (AID:A) panel study, which administered a pre-pandemic wave in 2019, and a post-pandemic wave in 2023. AID:A contains rich information on mothers’ and fathers’ time use for work, commuting, childcare, and housework. I find that the most robust effects emerge for paternal WFH intensity (at least weekly WFH) on parental division of housework: families in which fathers start weekly WFH in the period 2019 to 2023—due to their occupational WFH capacity in combination with the pandemic WFH-boost—experience a significant decrease in the maternal share of parental housework. Interestingly, this shift appears to be driven by a reduction of maternal time use for housework (combined with an increase of her work hours) rather than an increase in paternal time use for housework suggesting cross-parent effects of WFH.

ABSTRACT. This study examines the impact of the 17th Amendment, which introduced direct elections for U.S. Senators, on their legislative behavior. Using a difference-in-differences approach and NLP analysis of over 6.5 million congressional speeches (1879–1935), we find that direct elections move Senators’ policy agendas more closely with directly elected House Representatives, especially on fiscal policy and taxation, while reducing attention to infrastructure and immigration. Importantly, the effect is also observed when focusing only on incumbent Senators, and we find an overall narrowing focus on their agendas. However, voting behavior showed no significant ideological shift, suggesting the reform primarily influenced public discourse rather than legislative decisions.

ABSTRACT. Although tourism mobility spillover continues to be a key indicator for tourism management, more innovative research must be conducted at the micro level and high sampling frequency. Against the backdrop of an increasing number of global cities, in this paper, we evaluate the daily tourism mobility spillover inside a worldwide city of China: Shanghai. Based on the Granger causal network model and an original mobile positioning dataset, we analyse the causal relationship between local tourism flows and the spillover effects of tourism mobility within Shanghai. By categorising tourists into ‘local tourists from Shanghai’ and ‘tourists from out of Shanghai’, we reveal a significant causal relationship between Shanghai districts and flows generated by ‘tourists from out of Shanghai’. The analysis of the causal network structure also reveals key districts and points of interest that significantly contribute to congestion in tourism mobility and Shanghai's dynamics. This econometric approach offers policymakers a valuable tool to monitor mobility drivers and optimise flows within the city.

ABSTRACT. This study explores the crucial role of local immigration history in shaping the relationship between the arrival of refugees and support for anti-immigration parties. Using unique georeferenced individual-level survey data, we show that opening refugee hosting facilities close to voters polarises support for anti-immigration parties: in neighbourhoods with established immigrant communities such support reduces, while it increases in areas with fewer long-term immigrants. These findings are consistent when using official electoral outcomes. We extend the analysis beyond the direct effect of exposure to immigrants by examining the role of voters’ demographics. Our results are mainly driven by male, childless, and low-skilled working-age voters. Moreover, facilities that highlight the presence of short-term immigrants amplify the main effects. These findings underscore the importance of considering both the composition of the immigrant population and individual demographics to fully understand how contact influences voter behaviour. Crucially, by examining the interplay between long- and short-term immigrant presence, our study provides a unifying framework that explains the polarised voter responses to refugee shocks, reconciling contrasting outcomes observed in different contexts.

ABSTRACT. We investigate the drivers of the Italian debt/GDP ratio and its determinants with a state-of-the-art VAR model able to separate permanent vs. transitory demand and supply shocks. According to our VAR, permanent supply shocks are the number one candidate to explain the evolution of the real GDP growth and the debt-to-GDP ratio during and in the aftermath of the crisis. We document the correlation between such shocks and a measure of total factor productivity, and conclude that the Italian economy was cursed by a "damned cross", i.e., the concurrent increase in public debt/GDP and decrease in productivity.

ABSTRACT. This paper explores the fundamental drivers of U.S. housing price exuberance using a Time-Varying Parameter VAR model with Stochastic Volatility, combined with the recursive right-tailed unit root testing framework of Phillips et al. (2015a,b). The goal is to identify the structural shocks behind periods of exuberance—characterized by rapid increases in house prices beyond their fundamental values—and provide deeper insights into their dynamics. While exuberance is often attributed to non-fundamental factors, our analysis shows that it can also emerge from fundamental shocks, particularly those related to credit supply and demand. The results demonstrate statistically significant differences in the first-moment dynamics of house prices to credit-related shocks during exuberant versus non-exuberant periods, highlighting their critical role in driving housing market fluctuations. Additionally, our approach reveals previously undetected periods of exuberance when house prices are conditioned on structural shocks, which remain hidden when analyzing the broader housing price dynamcis alone.

ABSTRACT. We propose a novel identification strategy based on a combination of sign, zero, and policy coefficient restrictions to identify the exchange rate response to a US monetary policy shock. Our strategy crucially hinges upon imposing a sign on the policy response to exchange rate fluctuations, i.e., we require monetary policy to tighten after a depreciation of the US dollar. This restriction is supported by narrative accounts of the historical period we investigate, and it is particularly relevant to model Volcker’s monetary policy regime. We find evidence consistent with exchange rate overshooting and the existence of a conditional uncovered interest parity condition. Importantly, we show that our identification strategy implies robust impulse responses across samples characterized by different monetary policy conducts. Differently, restrictions imposed only on impulse responses return evidence that is sub-sample specific and associate Volcker’s regime with a delayed overshooting and a forward discount puzzle.

ABSTRACT. In this study we contribute to the literature on labour market resilience by using quarterly employment data for Italian regions. We, mainly, focus on long-term horizon (e.g low frequency band) to detect how a crisis event (localized in time) has an impact on a indicator of resilience. For this purpose, using the Continuos Wavelet Transform filter, we compute the sensitivity (observed over time and across frequency ranges) of each Italian region to a common (nationwide) shock. The empirical evidence suggests that, on average, highest resilience scores are observed for Centre-North region, while the lowest values are recorded for the South. Panel regression results show that EU cohesion policies have a positive effect on labour market resilience especially when the degree of financial development is high.