Building on a proper selection of macroeconomic variables for constructing a Gross Domestic Product (GDP) forecasting multivariate model (Kazanas, 2017), this paper evaluates whether alternative Bayesian model specifications can provide greater forecasting accuracy compared to a standard Vector Error Correction model (VECM). To that end, two Bayesian Vector Autoregression models (BVARs) are estimated, a BVAR using Litterman’s prior (1979) and a BVAR with time-varying parameters (TVP-BVAR). Two forecasting evaluation exercises are then carried out, a 28-quarters ahead forecast and a recursive 4-quarters ahead forecast. The BVAR outperformed the other models in the first, whereas the TVP-VAR was the best-performing model in the second, highlighting the importance of having adjusting mechanisms, such as time-varying coefficients in a model.
This paper documents five stylised facts relating to price adjustment in the euro area, using various micro price datasets collected in a period with relatively low and stable inflation. First, price changes are infrequent in the core sectors. On average, 12% of consumer prices change each month, falling to 8.5% when sales prices are excluded. The frequency of producer price adjustment is greater (25%), reflecting that the prices of intermediate goods and energy are more flexible. For both consumer and producer prices, cross-sectoral heterogeneity is more pronounced than cross-country heterogeneity. Second, price changes tend to be large and heterogeneous. For consumer prices, the typical absolute price change is about 10%, and the distribution of price changes shows a broad dispersion. For producer prices, the typical absolute price change is smaller, but nevertheless larger than inflation. Third, price setting is mildly state-dependent: the probability of price adjustment rises with the size of price misalignment, mainly reflecting idiosyncratic shocks, but it does not increase very sharply. Fourth, for both consumer and producer prices, the repricing rate showed no trend in the period 2005-19 but was more volatile in the short run. Fifth, small cyclical variations in frequency did not contribute much to fluctuations in aggregate inflation, which instead mainly reflected shifts in the average size of price changes. Consistent with idiosyncratic shocks as the main driver of price changes, aggregate disturbances affected inflation by shifting the relative number of firms increasing or decreasing their prices, rather than the size of price increases and decreases.
Large swings in cross-border capital flows can have consequences for domestic stability and open a channel for the transmission of shocks and spillovers across economies, including the euro area. Against this backdrop, the present paper reviews new evidence for the effectiveness of capital flow management policies in achieving macroeconomic and financial stability. Particular attention is paid to literature that has been used by the International Monetary Fund (IMF) to underpin its so-called Integrated Policy Framework, in which the roles of monetary, exchange rate, macroprudential and capital flow management policies are considered jointly. The literature published since the global financial crisis continues to affirm the effectiveness of capital flow management measures (CFMs) in addressing financial stability risks resulting from capital flow reversals; at the same time, however, it also continues to underscore that such policies should not substitute for warranted economic adjustments and structural reforms. Even so, recent literature also provides a case for considering, under certain circumstances, “precautionary” CFMs which could be applied to capital inflows to prevent a boom-and-bust cycle from being set in motion. This paper also highlights the need for further work on the long-term effects of such precautionary instruments, as well as their joint use with monetary policy instruments. Regarding capital flow management policies within the domain of central banks, the literature points to the usefulness of foreign exchange interventions (FXIs) in mitigating financial stability risks in countries with specific characteristics such as currency mismatches, borrowing constraints and shallow foreign exchange markets that are common to emerging market and developing economies alike. However, the literature also warns that such measures may reduce economic agents’ incentives to hedge against currency risks, with the result that unfavourable initial conditions beco
In this study, we explore the impact of COVID-19 pandemic on the default risk of loan portfolios of the Greek banking system, using cutting edge machine learning technologies, like deep learning. Our analysis is based on loan level monthly data, spanning a 42-month period, collected through the ECB AnaCredit database. Our dataset contains more than three million records, including both the pre- and post-pandemic periods. We develop a series of credit rating models implementing state of the art machine learning algorithms. Through an extensive validation process, we explore the best machine learning technique to build a behavioral credit scoring model and subsequently we investigate the estimated sensitivities of various features on predicting default risk. To select the best candidate model, we perform comparisons of the classification accuracy of the proposed methods, in 2-months out-of-time period. Our empirical results indicate that the Deep Neural Networks (DNN) have a superior predictive performance, signalling better generalization capacity against Random Forests, Extreme Gradient Boosting (XGBoost), and logistic regression. The proposed DNN model can accurately simulate the non-linearities caused by the pandemic outbreak on the evolution of default rates for Greek corporate customers. Under this multivariate setup we apply interpretability algorithms to isolate the impact of COVID-19 on the probability of default, controlling for the rest of the features of the DNN. Our results indicate that the impact of the pandemic peaks in the first year, and then it slowly decreases, though without reaching yet the pre COVID-19 levels. Furthermore, our empirical results also suggest different behavioral patterns between Stage 1 and Stage 2 loans, and that default rate sensitivities vary significantly across sectors. The current empirical work can facilitate a more in-depth analysis of AnaCredit database, by providing robust statistical tools for a more effective and responsive micro and macro supervision of credit risk.
We explore the relation between sound institutions favouring innovation and technology investment and firms’ emissions reduction. Even though emission abatement is achieved at the firm or plant level, we postulate that structural and institutional fac- tors underpinning green innovation, skills and technology adoption at the country level are of material importance. Advances in technology and infrastructure are the main drivers for the reduction of emissions and are, in turn, intrinsically linked to overall country characteristics. Sound institutions can act as enablers and accelerators for firms and industries in the green transition process, hence we find an attenuating effect on emissions conditional on firm attributes.
