Bayesian Dynamic Factor Model Research Articles

Low- and middle-income countries’ external debt cycle and global economic prospect

The rapid growth of peripheral countries has brought significant changes to the global economic and financial system, making it crucial to examine the relationship between their cyclical debt fluctuations and the world economy. Using the Bayesian dynamic factor model, this paper documents the presence of a set of common factors, known as global debt cycles, in the external debt accumulation of the low- and middle-income countries (LMICs). The interesting finding is that this debt cycles differ from well-known factors for global financial cycles in capital flows, such as U.S. monetary policy and global risk aversion. Even after controlling the factors for global financial cycles, our results reveal a significantly negative predictive effect of estimated debt cycles on global economic growth. Moreover, we demonstrate that this negative effect is primarily dominated by LMICs occupying prominent positions in international debt market networks, especially during the post-2008 financial crisis period. These findings emphasize the increasing importance of peripheral countries in the world system, which has been largely ignored in existing literature.

The comovement of bubbles’ responses to monetary policy shocks

In this paper, we present a two-step methodology designed to elucidate the extent of co-movement in the behavior of asset bubbles across a selected group of developed and emerging economies, with observations extending from the year 2000 onwards. We characterize the behavior of these bubbles, conceptualized as the disparity between fundamental and market values, by examining their responses to monetary policy shocks. Our investigation into their co-movement dynamics post-2000 involves the estimation of a Bayesian Dynamic Factor Model that incorporates the reactions of asset bubbles to monetary policy shocks, both in the short term and the long term. Notably, our findings reveal a convergence in short-term responses among these bubbles prior to the onset of the financial crisis. However, as the crisis unfolded, our results indicate a shift towards divergence. Concerning the long-term response of bubbles to monetary policy, divergence was observed before the crisis, but once the crisis had taken hold, there was a noticeable trend towards convergence. An influential idiosyncratic factor was identified for countries like the United States, South Korea, and Japan, impacting both short-term and long-term behavior. Conversely, the long-term behavior of bubbles in most emerging economies appears to be influenced by idiosyncratic factors. Furthermore, our results largely maintain their robustness even when utilizing an alternative measure of monetary policy stance during the period of the zero lower bound. Nevertheless, this secondary analysis suggests heightened volatility in the factors affecting both emerging and advanced markets.

Impact of China on commodity exporters

AbstractWe compare the effect of a domestic shock in China and the US on the real economy and financial markets of various commodity‐exporting countries. To obtain a reliable indicator for China's macroeconomic conditions, we estimate a Bayesian dynamic factor model using block‐exclusion restrictions to identify a China factor and a US factor from monthly macroeconomic data for China and the US. We, then, assess the implications of a negative shock to both factors on the macroeconomy of a commodity‐exporting nation using Bayesian FAVARs based on recursive identification. A negative China shock leads to output loss and a fall in stock prices in these countries. China shock affects the output of only a subset of countries in our sample compared to the US shock, which affects all countries. China shock has a larger, quicker and more persistent impact on the stock markets of commodity‐exporting countries compared to the US shock. Countries with weaker institutional or business environments experience a larger negative real effect of the China shock, whereas countries with less stable financial systems demonstrate stronger financial effects. Using historical decomposition, we establish a growing role of the China factor over time, in particular for large emerging economies such as Brazil and Russia.

International comovements of public debt

AbstractThis study presents evidence of global and regional comovements in public debt and examines their trends and features. We employ a Bayesian dynamic factor model with time‐varying parameters based on the debt‐to‐GDP ratio of 115 countries. We find the global factor accounts for 38 percent of the total variation in the debt ratio worldwide, but its share decreases over time. Countries with more financial openness and a greater share of external public debt are more exposed to global/regional comovements. The assertion that international financial integration encourages higher government borrowing applies primarily to high‐income countries and it does not hold globally.

Advances in nowcasting economic activity: The role of heterogeneous dynamics and fat tails

A key question for households, firms, and policy makers is: how is the economy doing now? This paper develops a Bayesian dynamic factor model that allows for nonlinearities, heterogeneous lead–lag patterns and fat tails in macroeconomic data. Explicitly modeling these features changes the way that different indicators contribute to the real-time assessment of the state of the economy, and substantially improves the out-of-sample performance of this class of models. In a formal evaluation, our nowcasting framework beats benchmark econometric models and professional forecasters at predicting US GDP growth in real time.

Time-variation in the effects of push and pull factors on portfolio flows: Evidence from a Bayesian dynamic factor model

The extent to which push and pull factors affect international capital flows is widely debated. We contribute to this strand of literature by estimating the relative importance of push and pull factors for portfolio flows over a time span, encompassing the global financial crisis, the European sovereign debt crisis as well as the beginning of the Covid-19 pandemic. To do so, we extract common and country-specific components from fund flow data using Bayesian dynamic factor models with time-varying coefficients and stochastic volatility. Assuming that the common component represents push factors and the country-specific component pull factors, we show that (i) time-variation matters and (ii) there is a substantial amount of heterogeneity in the importance of factors across regions (advanced versus emerging market economies) and asset classes (equity versus bonds). We find that the relative importance of push factors for flows into advanced economies has on average increased over time, particularly for EU countries. With respect to flows into emerging market economies, we find very heterogeneous results between individual countries. Moreover, we identify risk measures, US stock market returns, US real interest rates, the US real effective exchange rate and the oil price as important push factors. Pull factors seem to covary with domestic stock market returns, in particular.

ОСОБЛИВОСТІ ДІЛОВИХ ЦИКЛІВ У КРАЇНАХ ЦЕНТРУ ТА ПЕРИФЕРІЇ ЄС: ДОСВІД ДЛЯ УКРАЇНИ

Economic theory suggests that economic integration can promote business cycle co-movement, which in turn facilitates the institution of common counter-cyclical policies. However, evidence from empirical studies on co-movement within the EU and Europe as a whole is mixed, particularly concerning a so-called group of peripheral countries. This article argues that the existence of large international shocks and their heterogeneous impact on national economies changes the interpretation of co-movement. A decomposition of business cycles into common and country-specific components via a flexible Bayesian dynamic factor model with time-varying parameters and stochastic volatility reveals that the patterns of co-movement among the EU core and periphery are similar after common shocks – such as the Great Recession and the COVID-19 pandemic – and the upward bias introduced by the use of the Pearson correlation coefficient is accounted for. However, it is found that there is another important distinction between the EU core and periphery; that is, during the period of the Great Re-moderation that followed the Great Recession, the business cycles of the core EU countries converged to a lower level of volatility than those of the periphery. Moreover, it is shown that various standard measures of business cycle co-movement can conflate co-movement and volatility convergence, which alters their interpretation. Importantly, this article relates the experience of the EU core and periphery to that of Ukraine. In particular, it is found that the business cycle of Ukraine is similar to those of the EU periphery in terms of the level of its volatility and co-movement vis-à-vis the core EU countries, which has important implications for further development of Ukraine's European integration policy.

Does commodity price uncertainty matter for the cost of credit? Evidence from developing and advanced economies

In this study, we focus on the effect of commodity price uncertainty on the cost of bank credit for a broad sample of loans traced to firms operating in developing and advanced economies. Using loan-level data for the 1990–2019 period, we find novel evidence that commodity price uncertainty, as estimated by a Bayesian Dynamic Factor Model, increases the cost of bank loans particularly for commodity dependent firms operating in developing countries vis-à-vis commodity dependent firms operating in advanced economies. In a further analysis, when examining the effect of group specific commodity price uncertainty on cost of borrowing, we find that agricultural price uncertainty significantly increases the cost of credit of commodity dependent firms operating in developing economies. We also find that commodity price uncertainty rises the cost of bank credit more for loans traced to firms locating in bank-based countries as compared to loans granted to firms operating in market-based economies, suggesting that the financial structure of a country could play an important role in passing through the borrowing costs to firms. Lastly, we also find that the effect of commodity price uncertainty is more pronounced for smaller firms operating in developing countries as opposed to smaller firms operating in developed countries. All in all, the above evidence provides useful policy implications, particularly, for the longevity of corporate sector operating in developing countries whereby the local financial structure poses serious threats to firms’ future earnings.

Time-Variation in the Effects of Push and Pull Factors on Portfolio Flows: Evidence from a Bayesian Dynamic Factor Model

Time-Variation in the Effects of Push and Pull Factors on Portfolio Flows: Evidence from a Bayesian Dynamic Factor Model

Why business cycles diverge? Structural evidence from the European Union

The application of various measures of business cycle synchronization indicates that business cycle divergence is a robust feature of GDP time series in the European Union. An analysis of the sources of the divergence has demonstrated that the determinants of business cycle comovement in economics literature cannot explain this situation. In contrast, they imply that the cycles should converge. An alternative cause of divergence is investigated at the sectoral level using the Bayesian dynamic factor model. The results reveal that, on average, 73 % of the international comovement is driven by sectoral factors, and the manufacturing sector is the main carrier of international business cycles. The unique standing of manufacturing can be attributed to the unparalleled extent of sectoral linkages in the sector. Consequently, business cycle divergence in the European Union could be caused by a rapid decrease in the share of the manufacturing sector and a growing share of services characterized by low intersectoral linkages.

House price synchronization across the US states: The role of structural oil shocks

This paper analyzes the impact of disentangled oil shocks on the synchronization in housing price movements across all the US states plus DC. Using a Bayesian dynamic factor model, the house price movements are decomposed into national, regional, and state-specific factors. We then study the impact of oil-specific supply and demand, inventory accumulation, and global demand shocks on the national factor using linear and nonlinear local projection methods. The impulse response analyses suggest that oil-specific supply and consumption demand shocks are most important in driving the national factor. Moreover, as observed from the regime-specific local projection model, these two shocks are found to have a relatively stronger impact in a bearish rather than a bullish national housing market. Our results have important policy implications.

Machine Learning Predictions of Housing Market Synchronization across US States: The Role of Uncertainty

We analyze the role of macroeconomic uncertainty in predicting synchronization in housing price movements across all the United States (US) states plus District of Columbia (DC). We first use a Bayesian dynamic factor model to decompose the house price movements into a national, four regional (Northeast, South, Midwest, and West), and state-specific factors. We then study the ability of macroeconomic uncertainty in forecasting the comovements in housing prices, by controlling for a wide-array of predictors, such as factors derived from a large macroeconomic dataset, oil shocks, and financial market-related uncertainties. To accommodate for multiple predictors and nonlinearities, we take a machine learning approach of random forests. Our results provide strong evidence of forecastability of the national house price factor based on the information content of macroeconomic uncertainties over and above the other predictors. This result also carries over, albeit by a varying degree, to the factors associated with the four census regions, and the overall house price growth of the US economy. Moreover, macroeconomic uncertainty is found to have predictive content for (stochastic) volatility of the national factor and aggregate US house price. Our results have important implications for policymakers and investors.

International Co-Movements and Determinants of Public Debt

English Abstract: This paper provides evidence of significant international co-movements of public debt in the form of the common global and regional factors. International events such as the global financial crisis and regional sovereign debt crises suggest the existence of global and regional factors that can generate synchronizations of public debt across different countries. In contrast with previous studies that are focused mostly on domestic economic fundamentals in explaining public debt, this paper finds distinct global factors in the public debt-to-GDP ratio, both from principal components analysis and the Bayesian dynamic factor model. We show that the global factor accounts for a significant fraction of the variation of public debt often more substantial than those explained by domestic variables in many countries. We also find significant regional factors in public debt, especially for the Eurozone member states.

The emergence of regional business cycle in Africa—a reality or myth? A Bayesian dynamic factor model analysis

AbstractThis paper examines the evolution and the underlying forces of the business cycle co‐movements among seven African (A‐7) countries over the period 1970–2016. These countries accounted for over 60% of regional GDP, have abundant natural resources, access to the global capital market and widely viewed as potential drivers of regional (or “African”) business cycle. We model the national business cycle using real output, consumption and investment. We employ a dynamic factor model to decompose fluctuations in these macro‐variables into a regional, country‐specific and idiosyncratic components using Bayesian methods. We also analyse the relative importance of some notable drivers of business cycle fluctuations found in the literature. We find that the idiosyncratic component is dominating cyclical fluctuations in the A‐7 countries, while country‐specific and regional factors play a negligible role suggesting the inexistence of a common regional cycle despite deepening intra‐African trade. Among the driving variables, the terms‐of‐trade shocks exert greater influence on the A‐7 business cycle, while exchange rate movements and changes in money supply explain sizeable fluctuations in consumption and investment in most of the A‐7 countries. Shocks associated with changes in relative domestic oil prices, monetary and fiscal policies cause large output fluctuations.

Decoupling after the Crisis: Western and Eastern Business Cycles in the European Union

ABSTRACTThe aftermath of the crisis in Europe renewed the interest in the comovement business cycles and the ability of the European Central Bank (ECB) to implement a common monetary policy. Depending on the examined time period, the sample of countries and the utilized methodology, the research shows evidence of both convergence and decoupling of European business cycles. The present paper extends this research agenda by applying Bayesian dynamic latent factor models to the time series of GDP growth and the inflation of the Eurozone and Central and Eastern Europe (CEE) countries from 2000 till 2018. The results show evidence of two distinct business cycles in the Eurozone and the CEE, as well as the progressing decoupling in the European Union and convergence within the Eastern and Western Europe. Comovement of inflation has been slowly declining in the Eurozone, suggesting decreasing effectiveness of the EBC in following the inflation target in the entire euro area.

Globalization or Regionalization of Stock Markets? the Case of Central and Eastern European Countries

This article investigates the degree of financial integration of the Eastern European countries with global and regional financial markets and compares it with the core of the euro area. A Bayesian dynamic factor model was utilized to uncover global, regional and country factors driving the co-movement of rates of return of stock indexes. In the case of the Eurozone, the role of the country and international factor was extremely stable over the analyzed period. The composition of the international factor changed over time with the share of the global factor increasing during the time of the global financial crisis.The share of the country factor in the EMU was around 10%, while it was around 20% in Visegrad 4, indicating very strong international ties between the capital markets.

Measuring financial systemic stress for Turkey: A search for the best composite indicator

In this study, we aim to construct a single financial stress indicator (FSI) for Turkey adopting weekly data from between April 2005 and December 2016. To do so, we compose 15 different FSIs using 14 variables that will represent five different markets, i.e. the money market, the bond market, the foreign exchange market, the equity market and the banking sector. We aggregate these five different markets using a variety of techniques, including principal component analysis (PCA), basic portfolio theory, variance equal weights and the Bayesian dynamic factor model. We compare 15 different FSIs on the basis of their relation to, and the forecasting power of, different variables such as the growth rate of industrial production, the OECD business condition index and the OECD composite leading indicator for Turkey. Our results suggest that there is no simple best indicator for Turkey to measure financial systemic stress. Some indicators offer good forecasting power for economic growth while others have a stronger correlation with systemic risk. Therefore, we offer a final FSI for Turkey conducting a model averaging method via a rolling correlation based weighting scheme to benefit from the information content of all the FSIs and observe that the final FSI successfully indicates the tension periods.

Investigating the Drivers of International Comovement in Real Financial Asset Returns

There is a substantial body of theoretical and empirical research on asset price comovement and determinants. The empirical analysis in this paper differs in that it incorporates a channel for cross-country comovement in asset prices as well in a set of proposed asset price determinants, across a sample of 9 OECD countries. A Bayesian dynamic factor model is utilised to isolate common, or ‘world’, and country-specific shocks in stock, bond, currency, and house markets and in variables representing monetary policy, fiscal policy, productivity, demand, relative commodity prices and macroeconomic sentiment. The results are used to gauge the degree of financial and economic integration. Individual asset returns are then regressed on factors extracted from the driving variables to examine the relative importance of the common and country shocks. Stock and bond markets in particular are found to be driven largely by shocks which are common across all countries and asset markets, though a country level cycle in returns is also evident. Together the world factors in the driving variables are found to be a relatively large source of shocks for all asset markets, with shocks to fiscal policy variables, productivity and sentiment appearing to underpin international linkages in asset return volatility. The country-specific component in relative commodity price growth is a large driving force for individual returns.

An Inflation‐Predicting Measure of the Output Gap in the Euro Area

AbstractUsing a small Bayesian dynamic factor model of the euro area, we estimate the deviations of output from its trend that are consistent with the behavior of inflation. We label these deviations the output gap. In order to pin down the features of the model, we evaluate the accuracy of real‐time inflation forecasts from different model specifications. The version that forecasts inflation best implies that after the 2011 sovereign debt crisis, the output gap in the euro area has been much larger than the official estimates. Versions featuring a secular stagnation‐like slowdown in trend growth, and hence a small output gap after 2011, do not adequately capture the inflation developments.

Boom or Gloom? Examining the Dutch Disease in Two-speed Economies

Traditional studies of the Dutch disease do not account for productivity spillovers between the booming resource sector and other domestic sectors. We put forward a simple theory model that allows for such spillovers. We then identify and quantify these spillovers using a Bayesian dynamic factor model. The model allows for resource movements and spending effects through a large panel of variables, while also identifying disturbances to other shocks. Using Australia and Norway as representative cases studies, we find that a booming resource sector has substantial productivity spillovers on non-resource sectors, effects that have not been captured in previous analysis.