Coming soon. 

This paper provides a comprehensive review of various measures of uncertainty and their asset pricing implications in the cross-section of U.S. stock returns. We use a broad array of uncertainty measures, adding to those previously studied in the literature with novel survey-based measures sourced from three professional forecast datasets. Through both portfolio analyses and Fama-Macbeth regressions over the sample period between 1989 and 2020, we observe that exposure to these uncertainty measures can explain a significant portion of the cross-sectional dispersion in future stock returns. Specifically, for some of the measures examined in this paper, the relation between uncertainty and future returns is consistently negative and statistically significant across various asset pricing tests. This negative relation persists over long-term investment horizons, extending up to 36 months, and cannot be explained by the well-established return-predicting factors. Furthermore, our analysis reveals that, across all uncertainty measures, the return predictive power of uncertainty is more prominent in the subperiod from 2005 to 2020 compared to the subperiod between 1989 and 2004.

We propose a new model to decompose inflation swaps into genuine inflation expectations and risk premiums. We develop a no-arbitrage term structure model with stochastic endpoints, separating macroeconomic variables into transitory parts and long-run, economically-grounded, determinants, such as the equilibrium real interest rate and the inflation target. Our estimations deliver new insights as to how macroeconomic variables affect market-based inflation expectation measures.

We study the effect of oil price shocks on bond risk premia. Based on Baumeister and Hamilton (2019), we identify the different sources of oil price shocks using a structural vector autoregressive (SVAR) model of the global market for crude oil. These structural factors are then used as unspanned factors in an a ne term structure model based on the representation of Joslin et al. (2014). This is done for a total of 15 countries. Bond risk premia of net oil-exporting countries show a reaction to the structural shocks which is often statistically significant and in line with the expectation. For oil-importing developed countries, mainly the reaction to economic activity shocks is statistically significant and with the expected sign. The results for oil-importing developing countries are most of the time not statistically significant or run counter what one would expect. Among the unspanned factors, global economic activity explains most of the variability in bond risk.

Fine wine and alcohol prices can be predicted, but the accuracy of the prediction depends on the chosen forecasting horizon. In our study, we analyse the fine wine indices, as well as the retail and wholesale alcohol prices in the US from January 1992 to March 2022. We use comprehensive datasets that include economic, survey, and financial variables to identify the most relevant determinants of price. Using various combination and dimension reduction techniques, we develop a range of models to forecast prices across different selling levels and alcohol categories. Our research shows that fine wine prices can be predicted for twelve-month horizons, while retail and wholesale alcohol prices can be predicted for horizons ranging from one month to two years. To improve the accuracy of these predictions, we recommend incorporating consumer survey data and macroeconomic factors, such as international factors and mature market equity risk factors.

We study the role of the cost of inflation channel in determining the risk premium in a (nonlinear) New Keynesian DSGE model. Relying on a Calvo (or Rotemberg) price setting, we show that while the cost of inflation channel generates the desired term premium moments, it suffers from nontrivial, counterintuitive approximation errors in the price dispersion function. In addition to documenting the issues, we propose ways to alleviate them, including a quasikinked demand function as a risk-generating mechanism.

Abstract. In recent years, the international community has been increasing its efforts to reduce the human footprint on air pollution and global warming. Total carbon dioxide (CO2) releases are a crucial component of global greenhouse gas emissions, and as such, they are closely monitored at the national and supranational levels. This study presents different models to forecast energy CO2 emissions for the US in the period 1972–2021, using quarterly observations. In an in-sample and out-of-sample analysis, the study assesses the accuracy of thirteen forecasting models (and their combinations), considering an extensive set of potential predictors (more than 260) that include macroeconomic, nature-related factors and different survey data and compares them to traditional benchmarks. To reduce the high-dimensionality of the potential predictors, the study uses a new class of factor models in addition to the classical principal component analysis. The results show that economic variables, market sentiment and nature-related indicators, especially drought and Antarctic wind indicators, help forecast short/medium-term CO2 emissions. In addition, some combinations of models tend to improve out-of-sample predictions.