ABSTRACT. Mitigating the negative impact of climate change implies a drastic reduction in the emissions of greenhouse gases: moving toward the zero-emission target would require, among other things, dramatically improving the energy efficiency of residential buildings, that in Italy account for 12.5% of greenhouse emissions. This paper estimates the extent of the capitalization of energy efficiency labels into house prices. We find that the best energy-performing houses sell at a 25% premium compared to the worst-performing ones. Our contribution is relevant for two reasons. First, we provide granular estimates of the impact of energy labels on house prices in Italy and show that the energy-efficiency premium is significantly heterogeneous across provinces due to climate conditions and regulatory framework differences. Second, energy labels are salient and taken as the benchmark for some policies: a better-targeted policy may have resulted in a better allocation of public funds.

ABSTRACT. Prompted by the need to reduce the concentration of CO2 in the atmosphere in order to limit global warming, several countries are adopting policies to incentivize the production of clean energy. In this context, a relevant aspect to be examined is the effect of expanding renewable resources on employment. Despite the large use of panel and time series analysis to investigate the topic, most of the econometric models generally consider a very small number of regressors. Furthermore, the spatial component, a potentially important determinant of employment, has been always neglected. By making use of a relatively large dataset of 62 countries spanning for 25 years (from 1990 to 2014), the present paper tries to fill these gaps by specifying a dynamic spatial panel data (SDPD) model with fixed effects. The specification of both the individual and time fixed effects allows us to consider both spatial and temporal heterogeneity. Moreover, their presence and the long panel dimension avoid spurious correlations (Granger and Hyung, 1999). In particular, the temporal fixed effects control for the years of economic crises that affect our dataset. Our results confirm the positive role of expanding renewable energy production on employment, at the 10% level of confidence, while the expansion of non–renewable is statistically insignificant.

ABSTRACT. Can technology neutralize the threat that depletion poses to resource availability? We offer new insights into this long-standing topic by analysing the US mining sector of iron ore, an important primary commodity used in a wide range of industrial productions. We develop a new econometric approach that allows to distinguish the sign of unobserved shocks, and we use it to study potential asymmetries between technology and scarcity. We find that technological progress produces stronger and more persistent effects on productivity and price than the natural action of resource depletion, with global market structure influencing the size of such effect.

ABSTRACT. This paper focuses on the short-term electricity price forecasting. In particular the objective of this paper is twofold: on the one hand it documents the performance of forecasting methods with different assumptions (reduced forms, structural decompositions, nonlinearity) and degrees of mean reversion; secondly, it aims at exploring a procedure for interval prediction, concentrating on a new method, Conformal Prediction, which is an effective procedure for distribution-free predictive inference in regression. The empirical application focuses on the prediction of the time series of the single national price of the Italian Electricity Spot market in the short run, i.e., for forecast horizons that are not larger than 14 days ahead. Regarding the point forecast, our findings suggest that the best performing models are a robust autoregressive predictor and an unobserved components models featuring local trends and seasonality, whereas nonlinear specifications do not show a comparative advantage. With respect to the construction of prediction intervals, our results suggest that CP produces fairly accurate and reliable prediction intervals.

ABSTRACT. In this work we model and forecast commodity price time series using multivariate copula-based time series models. In particular, we consider the Gaussian copula, the t-copula and D-vine copulas to model both the contemporaneous and the temporal dependencies of the series. We focus on daily short-term gas, coal, oil and carbon emission futures in the period from March 2010 to February 2021. Further, we examine how the copula-based model gives rise to non-elliptical conditional probabilistic forecasts. During volatile periods the conditional forecast density can become bi-modular. This can be problematic when not only a distributional but also a mean forecast is desirable. Because for such probabilistic forecasts it is not clear what represents a good mean forecast. A possible solution is to augment the forecasts by an artificial neural network that predicts the best quantile to use as mean forecast based on past values and past best quantiles. In a forecasting study the predictive performances of the models are examined and compared with a benchmark. The distributional forecasts are examined by the CRPS. The mean forecasts are examined by the RMSE. The copula-based models are competitive with benchmark models.

ABSTRACT. This paper tackles the issue of producing point, direction-of-change and densityforecasts of the real price of carbon at monthly sampling frequency in the world’slargest carbon market, the EU Emission Trading Scheme (EU-ETS). Our researchpursues two intertwined targets. First, we aim to uncover supply- and demand-sideforces that can contribute to improving the prediction accuracy of models at shortandmedium-term horizons. The second main objective of the paper is to highlightwhich methodological choices have the potential to improve the forecast accuracyof models. Our results are relevant for researchers and policymakers interested inquantifying the desired and unintended macroeconomic effects of putting a priceon carbon. These assessments, coupled with the need to embed climate and carbonmarket modules into macroeconomic models used by central banks, suggest thathaving access to reliable short- and medium-term forecasts of the real price ofcarbon is becoming increasingly important.

ABSTRACT. Climate change poses serious economic, financial, and social challenges to humanity, and green transition policies are now actively implemented in many industrialized countries. Whether financial markets price climate risks is critical to ensuring that the necessary funding flows into environmentally sound projects and that stranded assets risk is adequately managed. In this paper, we assess climate risks for the European stock market within the context of Alessi et al. (2023) greenness and transparency factor. We show that measures of returns spreads of green vs. brown investment might reflect climate risks and assets' exposition to systematic macro-financial risk factors. These latter factors should be filtered out to measure climate risks accurately. We show that climate risks are priced in the European stock market by focusing on aggregate, industry, and company-level data. We propose a market-based green rating procedure, which might be of particular interest to evaluate non-transparent and non-disclosing companies for which ESG information is unavailable. We illustrate its implementation using a sample of over 800 non-transparent firms

ABSTRACT. In this study, we unpack the ESG ratings of four prominent agencies inEuropeand find that (i) each single E, S, G pillar explains the overall ESGscore differently,(ii) there is a low co-movement between the 3 E, S, G pillars and (iii)there arespecific ESG Key Performance Indicators (KPIs) that are driving theseratings morethan others. We argue that such discrepancies might mislead managersabout theiractual ESG status, potentially leading to cherry-picking areas forimprovement,thus raising questions about the accuracy and effectiveness of ESGevaluations inboth explaining sustainability and driving capital toward sustainablecompanies.

ABSTRACT. The relationship between corporate environmental performance and financial performance remains ambiguous after decades of study. To contribute to the ongoing debate, the relationship between these two variables is modelled as a system of simultaneous equations, with lagged endogenous variables and firm and time-fixed effects. Besides allowing for two-way causal dependence as the main methodological contribution, lagged endogenous variables capture the strong time series persistence in both environmental and financial performance present in the data. Although a standard set of observable firm-specific time-varying factors is conditioned upon, firm and time-fixed effects are also included to control for unobserved heterogeneity that varies across firms and over time. In this study, firms in the S&P 500 index are divided into brown (heavily polluting) and green (less polluting) sectors to investigate separately the relationship between environmental and financial performance for each. The results in this paper are in clear contrast with most of the existing literature. (i) Brown firms pay to be green; for these firms, better financial performance translates into higher environmental scores. Yet, this effect is not found for green firms. In addition, (ii) Neither brown nor green firms with higher environmental scores perform better financially. These results are robust to the estimation method (using Generalized Structural Equation Modelling and the Three Stage Least Squares approach) and to alternative proxy measures of firm financial performance.

ABSTRACT. Investors’ appetite for green finance increased considerably in particular after the Paris Agreement. However, the lack of a standardized definition of green and non-green investments represents a main obstacle to understand under which conditions a green portfolio may outperform the market. We investigate to what extent the European stock market prices climate transition risk, testing different classifications of “green” and “high-carbon” used by investors and financial authorities, including Greenhouse gas emissions intensity and carbon footprint, ESG, the EU Taxonomy, and the science-based classification Climate Policy Relevant Sectors (CPRS). As a difference from standard market-based classifications, the CPRS takes into account forward-looking dimensions of climate transition risk exposure, i.e. energy technology profile, fossil input substitutability and cost of policy and regulation. We develop portfolios and green-minus-brown factors and we analyse the market pricing in an augmented CAPM and Fama-French models. The results confirm that different classifications lead to very different evidence on greenium, i.e. higher expected returns for high-carbon assets w.r.t. green assets. Interestingly, we find that the existence of a greenium disappears when considering forward-looking climate risk characteristics captured by combining the CPRS and the EU Taxonomy. Our approach allows for better hedging of climate transition risk emerging from policy and regulatory announcements, and to consider policy uncertainty in climate-aligned portfolio construction.

ABSTRACT. We propose a new approach to detect and quantify informal employment resulting from irregular migration shocks. Focusing on a largely informal sector, agriculture, and on the exogenous variation from the Arab Spring wave on southern Italian coasts, we use machine-learning techniques to document abnormal increases in reported (vs. predicted) labor productivity on vineyards hit by the shock. Misreporting is largely heterogeneous across farms depending e.g. on size and grape quality. The shock resulted in a 6% increase in informal employment, equivalent to one undeclared worker for every three farms on average and 23,000 workers in total over 2011-2012. Misreporting causes significant increases in farm profits through lower labor costs, while having no impact on grape sales, prices, or wages of formal workers.

ABSTRACT. The long-term effects of climate policies crucially depend on firms' capacity to successfully adapt by changing the input mix and investing in carbon-saving technologies. However, financially constrained firms will find it more difficult to change, invest and ultimately to reduce the carbon content of production. Little research has been conducted on the interaction of climate policy and financial constraint at the firm-level. This paper fills this gap by exploiting a unique database on French manufacturing companies containing detailed information on carbon emission, exposure to policies, and financial variables allowing one to build reliable indicators of financial constraints. Preliminary results suggest that after an expansionary monetary policy shock, more financially constrained tend to do more carbon-saving. Furthermore, our preliminary results also suggest that this effect is even stronger among the financially constrained firms that are regulated by the EU ETS.

ABSTRACT. This paper employs insights from earth science on the financial risk of project developments associated with some key critical minerals to develop an economic theory of critical minerals. Our theory posits that back-ended critical mineral projects, such as those involving lithium and cobalt, exhibit an additional risk for investors which we term the “back-ended risk premium”. We show that the back-ended risk premium increases the cost of capital and, therefore, reduces investment in the sector. We posit that the back-ended risk premium has the potential to reduce the gains in productivity expected from artificial intelligence (AI) technologies in the mining sector. Progress in AI may, however, reduce the back-ended risk premium itself through reducing the duration of mining projects and required rate of investment through reducing the associated risk.  We conclude through observing that the best way to reduce the costs associated with energy transition is for governments to invest heavily in AI mining technologies and research.

ABSTRACT. We consider bootstrap inference for estimators which are (asymptotically) biased. We show that, even when the bias term cannot be consistently estimated, valid inference can be obtained by proper implementations of the bootstrap. Specifically, we show that the prepivoting approach of Beran (1987, 1988), originally proposed to deliver higher-order refinements, restores bootstrap validity by transforming the original bootstrap p-value into an asymptotically uniform random variable. We propose two different implementations of prepivoting (plug-in and double bootstrap), and provide general high-level conditions that imply validity of bootstrap inference. To illustrate the practical relevance and implementation of our results, we discuss five examples: (i) inference on a target parameter based on model averaging; (ii) ridge-type regularized estimators; (iii) nonparametric regression; (iv) a location model for infinite variance data; and (v) dynamic panel data models.

ABSTRACT. Panel data with a relatively long time span often exhibit trends. We study a class of time-varying coefficient panel regression models with cross-sectional dependence, serial dependence, and heteroskedasticity in both the regressors and the error processes. Trending behaviors are allowed in the dependent and explanatory variables. Since missing observations often occur in applications, especially in climatic studies, we propose a local linear dummy variable estimator which can handle missing values in the dependent variables. We establish a first set of (pointwise) asymptotic results. One of the key ingredients in our theory is a newly derived sub-optimal uniform convergence rate for Lp-NED processes in kernel estimation of large panels (N, T → ∞). We find that the limiting distribution reflects the patterns of missing values, heavily relying on various nuisance parameters. For inference, we propose a new bootstrap procedure, the autoregressive wild bootstrap, to construct pointwise confidence intervals and simultaneous confidence bands around coefficient curves. An extensive simulation study shows good finite sample performance of the proposed methods. As an illustration, we apply our methods to study two empirical problems. First, we study the relationship between global air pollution and economic development. Second, we examine global trends in atmospheric ethane emissions. We find significant time-varying patterns using our methods in both examples.

ABSTRACT. Quantile-spillover indexes have recently become popular to analyze tail interdepen-dence. In an extensive simulation study we show that the estimation of spillover indexesis aected by a positive distortion when the parameters of the underlying tted modelsare not evaluated with respect to their statistical signicance. This is a consequenceof loss of eciency due to the inclusion of irrelevant variables. The distortion reducesby ltering out non-signicant parameters and also for increasing sample sizes, thanksto consistency of estimators, but is not fully disappearing due to type I error. Wemake a further step by introducing a simulation-based approach for recovering con-dence intervals of quantile-spillover indexes. In addition, we put forward an algebraicdecomposition of quantile-spillover separating the dynamic interdependence from thecontemporaneous interdependence (due to residuals correlation). Empirical evidenceshow that distortions on real data are sizable, and the decomposition point out thatmost of the spillover is due to contemporaneous eects. All of our results extend andare conrmed for the Spillover index of Diebold and Yilmaz (2009).

ABSTRACT. Univariate panel model with interactive fixed effects has been well discussed in previous studies. This paper studies the multivariate panel vector autoregression (PVAR) model with group-based factors. It is flexible as the number of groups, the group membership in each group and the number of group factors in each equation are not specified, and it can be extended to group-specific heterogeneous coefficient Panel VAR. furthermore, it is a parsimonious structural model and easy to compute. We derive the asymptotic distribution and establish consistency of the estimator for N and T that tend to infinity. We use the model to explore the relationship between cross-country economic growth and income inequality.

ABSTRACT. In this study, we contribute to the existing literature on the impact of temperature on growth by examining the orthogonalized seasonal effect using a sample of 225 EU NUTS2 regions. For this purpose, we use a Panel Mixed-Frequency VAR estimated through Bayesian methods. The empirical findings show, first, a worsening impact of temperature on growth over the last sub-sample (2000-2019) of a rolling window estimation. Then, focusing on the aforementioned sub-sample, the empirical evidence shows that the vulnerability to seasonal temperature effects depends not only on the different sector contributions to aggregate GVA growth (we distinguish between the public sector and five private sub-sectors) but also on the initial temperature level (there is a different impact on hot and cold regions). Finally, it is not income per capita, but a proxy of competitiveness to contribute to the resilience of European regions to temperature shocks.

ABSTRACT. This paper tests the feasibility and estimates the cost of climate control through economic policies. It provides a toolbox for a statistical historical assessment of a Stochastic Integrated Model of Climate and the Economy, and its use in (possibly counterfactual) policy analysis. Recognizing that stabilization requires supressing a trend, we use an integrated-cointegrated Vector Autoregressive Model estimated using a newly compiled dataset ranging between years A.D. 1000-2008, extending previous results on Control Theory in nonstation- ary systems. We test statistically whether, and quantify to what extent, carbon abatement policies can effectively stabilize or reduce global temperatures. Our formal test of policy feasibility shows that carbon abatement can have a significant long run impact and policies can render temperatures stationary around a chosen long run mean. In a counterfactual empirical illustration of the possibilities of our modeling strategy, we show that the cost of carbon abatement for a retrospective policy aiming to keep global temperatures close to their 1900 historical level is about 75% of the observed 2008 level of world GDP, a cost equivalent to reverting to levels of output historically observed in the mid 1960s. This constitutes a measure of the putative opportunity cost of the lack of investment in carbon abatement technologies.

ABSTRACT. This paper investigates the impact of extreme hot and cold temperatures on a comprehensive measure of health, quantifying exposure through the duration of heat and cold waves. We adopt a human capital framework, constructing a health capital proxy based on days lost due to disability. This metric encompasses a wide array of health issues potentially linked to temperature exposure and enables us to quantify the time individuals spend in less-than-optimal health states at various life stages. Our results reveal that exposure to heat waves during childhood, especially before the age of 10, has a significant effect on the number of days lost due to disability at later ages. This heightened sensitivity to temperature exposure during early childhood is justified by physiological factors in children's development that may constrain their ability to regulate body temperature effectively. Additionally, children's capacity to employ protective measures against extreme heat or communicate thermal discomfort might be limited. Furthermore, early childhood exposure significantly shapes the following accumulation of health capital, with potential compensatory behaviours unable to fully mitigate its consequences. This effect persists over time and exerts a considerable influence on adult health, surpassing the influence of contemporaneous temperature shocks. This finding contributes valuable insights to the literature on the long-term consequences of early childhood shocks, emphasizing the almost underexplored role of temperature shocks in this context.

ABSTRACT. This paper assesses the effect of US-China political relationships and geopolitical risks on oil prices. To this end, we consider two quantitative measures—the Political Relationship Index (PRI) and the Geopolitical Risk Index (GPR)—and rely on structural VAR and local projections methodologies. We expand the literature on the macroeconomic consequences of geopolitical risks by considering bilateral political relationships. The bilateral GPR does not focus on the relation between the US and China; rather, it provides an overall picture of the geopolitical uncertainty for China on a multilateral basis. Our empirical investigation shows that improved US-China relationships, as well as higher geopolitical risks, drive up the price of oil. Indeed, unexpected shocks in the political relationship index are associated with optimistic expectations about economic activity, whereas unexpected shocks in the geopolitical risk index reflect fears of supply disruption. Political tensions and geopolitical risks are thus complementary causal drivers of oil prices, the former being linked to the demand side and the latter to the supply side.

ABSTRACT. We analyze the content of the Organization of the Petroleum Exporting Countries (OPEC) communications and whether it provides valuable information to the crude oil market. To this end, we derive an empirical strategy which allows us to measure OPEC’s public signal and test its credibility. Using Structural Topic Models, we identify several topics in OPEC narratives. We show that these topics are related to fundamental fac- tors such as demand, supply, and speculative activity in the crude oil market, highlighting that OPEC narratives are highly linked to oil market volatility and traders’ positions. We also find that OPEC communication is credible, reduces oil price volatility, and prompts market participants to rebalance their positions.

ABSTRACT. This paper analyses the informational efficiency of global crude oil markets using a recently proposed quantitative measure for market inefficiency. The procedure measures how far away observed oil price behaviour is from the Random Walk benchmark which represents an efficient market. The key findings are, first, that crude oil markets are found to be more informationally inefficient during extreme episodes such as the price downturns witnessed in 2008, 2014, and early 2020. During these periods, oil market uncertainty is high; thus there is a Blind Spot in the efficient market hypothesis. Second, the inefficiency measure calculated in this paper resembles in its development over time prominent measures for economic uncertainty which have recently been proposed. Additional findings include that there is no systematic trend in inefficiency over time, the degree of inefficiency varies considerably across markets pre-2006, but then converges. Finally, both oil prices and degrees of inefficiency begin to deviate from each other in 2022.

ABSTRACT. Motivated by the public debate on sanctioning crude oil imports from Russia, we esti- mate the elasticity of substitution between different crude oil types. Using European data on country-level crude oil imports by field of origin, we argue that crude oil is not a homogenous good and that the relevant substitutability for analyzing the impact of trade sanctions must account for the quality of different oil types in terms of their API gravity and sulfur content. Our results suggest that, by neglecting these differences in quality, standard estimates significantly underestimate the production disruptions in crude oil refining resulting from sanctions.

ABSTRACT. This paper proposes a non-parametric test for Granger causality in quantiles for data sampled at mixed frequencies. We extend the test proposed by \cite{JeHaeSo12} resorting to a two-step procedure to detect causality from a high-frequency target to a low-frequency target. In an economic application, we examine Granger causality between inflation, as a low-frequency macroeconomic variable, and a selection of commodity futures, including gold, oil, and corn, as high-frequency financial variables. We find that there exists a causal relationship between the two. The logarithmic returns on these commodity futures are a prima facie cause of inflation at the lower quantiles of the distribution and marginally around the median.

ABSTRACT. In this paper we investigate the dynamic features of energy commodity prices. Using a generalized smooth transition model (GSTAR) we show that dynamic symmetry in price cycles in the energy markets is stronglyrejected. Further, our results show that the proposed model performs well when compared to other linear andnonlinear specifications in an out-of-sample forecasting exercise.

ABSTRACT. This paper investigates the effects of financial shocks on realactivity from a medium-run perspective. We use a structural VAR with anovel identification scheme based on cyclical sign restrictions;intuitively, we build on the idea that financial and housing cyclesdisplay longer duration and higher persistence beyond the traditionalbusiness cycle frequencies. Our main results are threefold: (a)medium-term frequencies are crucial to properly identify the realeffects of financial shocks; (b) in the medium run, the effect of anexcess of credit is detrimental to real economic activity; (c) themedium-run effects emerge only from the mid-1980s.

ABSTRACT. We propose a structural vector-autoregressive model for the natural gas market to investigate the impact of the 2022 Russian supply stop on the German economy. We combine conventional and narrative sign restrictions to leverage information about supply cuts for identification and find that supply and demand shocks have large and persistent price effects, whereas output effects are rather short-lived. The 2022 natural gas price spike was driven by negative supply and positive storage demand shocks. Counterfactual simulations of an earlier embargo on Russian gas imports indicate only moderately larger negative output effects compared to what we observe in the data.

ABSTRACT. This paper uses a Max-Share approach to identify oil supply news shocks within a Non-Causal VAR model of standard global oil market variables. News shocks are identified in a way that explain most of the movements in real oil prices driven by global oil production, over a long but finite time horizon. Our findings highlight the prominent role of expectations in propagating oil supply shocks. Negative oil supply news shocks cause a gradual and persistent decline in global oil production and global economic activity and a strong and immediate increase in the most forward-looking variables, namely real oil price and global oil stock. Finally, news about future oil supply shortfalls has substantial consequences in macroeconomic variables leading to disruptions in both real and financial sectors.

ABSTRACT. The COVID-19 pandemic resulted in unprecedented shocks that massively affected macroeconomic variables and the demand for energy in the euro area. The Russian invasion of Ukraine has also generated large energy shocks, leading to an increase in energy prices. At the same time, the Harmonized Index of Consumer Prices (HICP) in the euro area has reached record levels since the introduction of the single currency. In this paper, we analyze the main forces driving these unusual observations and decompose the relative importance of large pandemic and energy shocks. We propose and estimate a two-sector dynamic stochastic general equilibrium model (DSGE) with the explicit introduction of crude and refined energy sources. In this model, an energy sector combines crude sources (oil, coal, gas) and other production inputs to supply refined energy to firms and households in the core sector. Our model describes the transmission mechanism of energy shocks, such as complementarities in production and consumption as business cycle amplification mechanisms. We find that the effects of price shocks on oil, coal, and gas account for about 40% of price increases between 2021:Q1 and 2022:Q2, with oil and gas the most important contributors.

ABSTRACT. How do shifts in the supply of natural gas affect output and inflation? To answer this question, we construct an instrument for gas supply shocks using a large set of daily news on the European gas market over the 2010-2022 period and use the instrument within a Bayesian VAR model. We find that negative supply shocks are stagflationary and that their effects materialize over far longer horizons than those of oil supply shocks, with peaks (troughs) in core inflation (industrial production) that follow the shock by two years or more. This pattern is consistent with the structural features of the gas market, and it suggests that European economies are still grappling with the large price spikes that took place in 2022.

ABSTRACT. This paper is joint with David Hendry and J. Isaac (Zack) Miller

Extreme weather events occurring around the world are a daily reminder that our climate is rapidly changing due to human activity. The long-term pattern of weather, or climate, is determined by human behaviour interacting with the physical properties of Earth’s climate system. To understand how the climate is changing and prepare for the future, both in terms of mitigation and adaptation, we need accurate forecasts of the climate. However, human behaviour is non-stationary from both stochastic trends and location shifts. Consequently, the time-series data generated by the interaction of humanity and the climate are non-stationarity from distributional shifts, resulting in forecasts that are uncertain and prone to failure howsoever they are generated. We discuss how climate scientists and climate econometricians produce forecasts, drawing parallels between the scenario projections of climate scientists based on varying initial conditions and the conditional forecasts of econometricians capturing model uncertainty. The success of forecasts made by either approach hinges on the ability to handle unanticipated shifts as climate change is characterised by ‘the change in the change’. While unanticipated changes cannot be avoided, whether shifts that were once in the future were expected or not, they later become in-sample, so empirical modelling must take account of those to avoid distortions in parameter estimates and the resulting forecasts. We look at why it is important to identify and model location shifts and how doing so improves the verisimilitude of the model and its forecasts. Using indicator saturation estimators to capture in-sample shifts, improved econometric models and their forecasts can be achieved, as demonstrated within a system of four of the key climate variables, atmospheric CO2, global mean surface temperature deviations, ocean heat content deviations and global sea-level rise.