Factor-augmented Vector Autoregression Research Articles

Estimation and Inference of FAVAR Models

The factor-augmented vector autoregressive (FAVAR) model is now widely used in macroeconomics and finance. In this model, observable and unobservable factors jointly follow a vector autoregressive process, which further drives the comovement of a large number of observable variables. We study the identification restrictions for FAVAR models, and propose a likelihood-based two-step method to estimate the model. The estimation explicitly accounts for factors being partially observed. We then provide an inferential theory for the estimated factors, factor loadings, and the dynamic parameters in the VAR process. We show how and why the limiting distributions are different from the existing results. Supplementary materials for this article are available online.

The effect of banking sector’s business conditions on the transmission mechanism of monetary policy

In this paper, we estimate a dynamic factor model for Korean macro economy and banking sector's business conditions by using the FAVAR (Factor augmented vector autoregressive) model, and analyze impulse responses of various variables such as macro aggregates and banks' financial ratios.Our empirical analysis shows that the macro economy tends to affect the banking sector unilaterally over time. Next, in our counter-factual analysis where we artificially remove the effect of banking sector on the macro economy in the FAVAR model, we find that there is no substantial effect of banking sector's business conditions on the transmission mechanism of monetary policy.

Effects of Stock Indices of Developed and Emerging Markets on Economic Activity in Colombia: a FAVAR Approach

Analizamos el efecto de las condiciones financieras de economias emergentes y desarrolladas sobre el comportamiento de la economia colombiana. Para ello, estimamos indices de condiciones financieras para mercados accionarios de paises desarrollados y emergentes usando analisis de componentes principales. Posteriormente, estos indices se incluyeron como regresores en un modelo tradicional de vectores auto-regresivos (VAR); es decir, estimamos un modelo de vectores auto-regresivos aumentado con factores (FAVAR) que incluye un factor no observable y una variable observable de la actividad economica real con el fin de evaluar el efecto de las condiciones financieras internacionales sobre las variables macroeconomicas colombianas. De esta manera, encontramos que las condiciones accionarias de los paises emergentes tienen un efecto positivo sobre el comportamiento macroeconomico de Colombia cuando se lleva a cabo el analisis con datos mensuales.

The Impact of US Monetary Policy and Other External Shocks on the Hong Kong Economy: A Factor-Augmented VAR Approach

This paper uses the factor-augmented VAR (FAVAR) framework to study the impact on the Hong Kong economy of the diverging monetary policies by the Fed, ECB and BoJ as well as the Mainland economy slowdown. The empirical results show that changes in US monetary policy mainly affect interest rate-sensitive sectors in Hong Kong; while real variables such as real GDP growth, unemployment rate are more sensitive to the economic slowdown in Mainland China. Monetary easing from the ECB and BoJ to some extent offsets the tightening of the Fed. The transmission channels of external shocks are through trade and capital markets. It is estimated that the combined effect of the four external shocks will on average lower Hong Kong’s quarterly GDP growth by 0.6 percentage points and quarterly inflation by 0.2 percentage points in the first 4 quarters. However, Hong Kong’s financial stability, particularly with regard to loan quality, banks’ capital and liquidity, is well maintained by macro-prudential policies suggesting that Hong Kong’s financial system is resilient to external shocks.

The macroeconomic effects of uncertainty shocks: The role of the financial channel

This paper studies the macroeconomic effects of uncertainty shocks with an emphasis on the interaction between elevated uncertainty and credit market conditions when the economy is in different regimes (recessions vs. non-recessions). We use a smooth-transition factor-augmented vector autoregression (ST-FAVAR) and a large monthly panel of U.S. macroeconomic and financial indicators in our estimation. Our findings are twofold. First, while an unanticipated increase in uncertainty has adverse effects on the real economy and financial markets, the effects are quantitatively larger during recessions. Second, the financial channel is important in the transmission of uncertainty shocks, with a greater role during recessions and in the short run.

Measuring the Channels of Monetary Policy Transmission: A Factor-Augmented Vector Autoregressive (Favar) Approach

Abstract There is more consensus on the effects of monetary policy than its transmission mechanism. Two channels of transmission mechanisms are the conventional interest rate channel and the credit channel. I investigate the channels of monetary policy transmission in the U.S. using the factor-augmented vector autoregressive (FAVAR) models developed by Bernanke, Boivin & Eliasz (2005). The newly developed FAVAR approach allows the researcher to include all relevant macroeconomic variables in the model and analyze them. Therefore, the FAVAR models span a larger information set and generate better estimates of impulse response functions than the commonly used vector autoregressive (VAR) models that utilize only 4–8 variables. I include 154 monthly U.S. time series variables for the period 1970–2014. The findings support the existence of the credit channel in the U.S. The conclusion remains the same when the non-borrowed reserve operating regime (October 1979–October 1982) is removed from the sample period.

A Comparison of Different Short-Term Macroeconomic Forecasting Models: Evidence from Armenia

Abstract We evaluate the forecasting performance of four competing models for short-term macroeconomic forecasting: the traditional VAR, small scale Bayesian VAR, Factor Augmented VAR and Bayesian Factor Augmented VAR models. Using Armenian quarterly actual macroeconomic time series from 1996Q1 – 2014Q4, we estimate parameters of four competing models. Based on the out-of-sample recursive forecast evaluations and using root mean squared error (RMSE) criterion we conclude that small scale Bayesian VAR and Bayesian Factor Augmented VAR models are more suitable for short-term forecasting than traditional unrestricted VAR model.

How Big is the Comeback? Japanese Exchange Rate Pass-Through Assessed by Time-Varying FAVAR

We examine the evolution and the magnitude of exchange rate pass-through (ERPT) to Japanese prices. We employ the Time-Varying-Parameters Factor-Augmented Vector au-toregression model (TVP-FAVAR), which enables us to include a large enough panel of data to better control for variables impacting prices and exchange rates. Our results confirm the decline in ERPT rates until the late 2000s and their resurgence in last years. Our findings provide additional support to the notion that exchange rates can impact import and domestic prices, possibly helping avoid deflation. We also find that the ERPT into aggregate prices can hide considerable variation in price sensitivity to exchange rates across stages of demand and industries. ERPT decreases along the pricing chain, from imported raw materials and intermediate goods prices to domestic prices. Finally, we find price sensitivity consistent throughout, indicating that incorporating extra information leads to more robust estimates.

The interest rate pass-through in the euro area during the sovereign debt crisis

We investigate the pass-through of monetary policy to bank lending rates in the euro area during the sovereign debt crisis, in comparison to the pre-crisis period. We make the following contributions. First, we use a factor-augmented vector autoregression, which allows us to assess the responses of a large number of country-specific interest rates and spreads. Second, we analyze the effects of monetary policy on the components of the interest rate pass-through, which reflect banks' funding risk (including sovereign risk) and markups charged by banks over funding costs. Third, we not only consider conventional but also unconventional monetary policy. We find that while the transmission of conventional monetary policy to bank lending rates has not changed with the crisis, the composition of the pass-through has changed. Specifically, expansionary conventional monetary policy lowered sovereign risk in peripheral countries and longer-term bank funding risk in peripheral and core countries during the crisis, but has been unable to lower banks' markups. This was not, or not as much, the case prior to the crisis. Unconventional monetary policy helped decreasing lending rates, mainly due to large shocks rather than a strong propagation.

Where do jobs go when oil prices drop?

In this paper, we estimate a factor augmented vector autoregressive (FAVAR) model to investigate the effect of oil price shocks on total private job flows as well as on industry-level job creation and destruction. Following an unexpected oil price drop in the first year, we find that in oil and gas extraction and support activities for mining exhibit a reduction in job creation and an increase in job destruction. Instead, industries in construction, manufacturing and services exhibit an increase in the net employment change. An unexpected decline in the real oil price slows down the pace of gross job reallocation. We demonstrate that the increase (decrease) in private job destruction (creation) observed during the first year is primarily driven by the response of closing (expanding) firms in services and manufacturing.

A multi-country approach to forecasting output growth using PMIs

This paper derives new theoretical results for forecasting with Global VAR (GVAR) models. It is shown that the presence of strong unobserved common factors can lead to an undetermined GVAR model. To solve this problem, we propose augmenting the GVAR with additional proxy equations for the strong factors and establish conditions under which forecasts from the augmented GVAR model (AugGVAR) uniformly converge in probability (as the panel dimensions N,T→j∞) to the infeasible optimal forecasts obtained from a factor-augmented high-dimensional VAR model. The small sample properties of the proposed solution are investigated by Monte Carlo experiments as well as empirically. In the empirical part, we investigate the value of the information content of Purchasing Managers’ Indices (PMIs) for forecasting global (48 countries) output growth, and compare forecasts from AugGVAR models with a number of data-rich forecasting methods, including Lasso, Ridge, partial least squares, and factor-based techniques. It is found that (a) regardless of the forecasting methods considered, PMIs are useful for nowcasting, but their value added diminishes quite rapidly with the forecast horizon, and (b) AugGVAR forecasts do as well as other data-rich forecasting techniques for short horizons, and tend to do better for longer forecast horizons.

Monetary Policy, Residential Investment, and Search Frictions: An Empirical and Theoretical Synthesis

Using a factor-augmented vector autoregression (FAVAR), this paper shows that residential investment contributes substantially to GDP following monetary policy shocks. Further, it shows that the number of new housing units built, not changes in the sizes of existing or new housing units, drives residential investment fluctuations. Motivated by these results, this paper develops a dynamic stochastic general equilibrium (DSGE) model where houses are built in discrete units and traded through searching and matching. The search frictions transmit shocks to housing construction, making them central to producing fluctuations in residential investment. The interest rate spread between mortgages and risk-free bonds also transmits monetary policy to the housing market. Following monetary shocks, the DSGE model matches the FAVAR’s positive co-movement between non-durable consumption and residential construction spending. In addition, the FAVAR shows that the mortgage spread falls following an expansionary monetary shock, providing empirical support for the DSGE model’s monetary transmission mechanism.

Consumer confidence and economic activity: a factor augmented VAR approach

ABSTRACTThis study aims to analyse the effects of the consumer confidence on economic activity for the US market. We use the empirical factor-augmented vector autoregression (FAVAR) method, which enables us to incorporate a wide range of economic activity factors into the analysis. The consumer confidence index (CCI) is chosen as the principal variable that is presumed to represent the degree of optimism on the state of economic activity. The results show that consumer confidence and economic activity are strongly correlated for manufacturing-related factors, such as industrial production and inventories. We also observe strong relation among CCI and personal consumption expenditures, as well as housing market variables.

Consumer Confidence and Economic Activity: A Factor Augmented VAR Approach

ABSTRACTThis study aims to analyse the effects of the consumer confidence on economic activity for the US market. We use the empirical factor-augmented vector autoregression (FAVAR) method, which enables us to incorporate a wide range of economic activity factors into the analysis. The consumer confidence index (CCI) is chosen as the principal variable that is presumed to represent the degree of optimism on the state of economic activity. The results show that consumer confidence and economic activity are strongly correlated for manufacturing-related factors, such as industrial production and inventories. We also observe strong relation among CCI and personal consumption expenditures, as well as housing market variables.

The World is Not Enough! Small Open Economies and Regional Dependence

This paper bridges the new open economy factor augmented VAR (FAVAR) studies with the recent findings in the business cycle synchronization literature emphasizing the importance of regional factors. That is, we estimate and identify a three block FAVAR model with separate world, regional and domestic blocks and study the transmission of both global and regional shocks to four small open economies (Canada, New Zealand, Norway and UK). The results show that foreign shocks explain a major share of the variance in all countries, most so shocks that are common to the world. However, regional shocks also play an important role, explaining more than 20 percent of the variance in the variables. Hence in small open economies, the world is not enough. The regional factors impact the four countries differently, though, some through trade and some through consumer sentiment. Our findings of a strong transmission of both global and regional shocks to open economies are in sharp contrast to the evidence from recently developed open economy DSGE models.

Interpreting the Latent Dynamic Factors By Threshold FAVAR Model

This paper proposes a method to interpret factors which are otherwise difficult to assign economic meaning to by utilizing a threshold factor-augmented vector autoregression (FAVAR) model. We observe the frequency of the factor loadings being induced to zero when they fall below the estimated threshold to infer the economic relevance that the factors carry. The results indicate that we can link the factors to particular economic activities, such as real activity, unemployment, without any prior specification on the data set. By exploiting the flexibility of FAVAR models in structural analysis, we examine impulse response functions of the factors and individual variables to a monetary policy shock. The results suggest that the proposed method provides a useful framework for the interpretation of factors and associated shock transmission.

Divergent EME Responses to Global and Domestic Monetary Policy Shocks

We assess the effect of tighter monetary policy in the U.S. and emerging market economies (EMEs) on EMEs using a panel factor-augmented VAR model. We find that a U.S. policy rate hike outstrips an equivalent domestic rate hike in its impacts on EMEs. In addition, EMEs show divergent policy responses and their macro-financial responses differ depending upon their economic fundamentals in the face of tighter U.S. policy. In particular, we find that high-inflation than low-inflation EMEs are more susceptible to the shock stemming from a U.S. federal funds rate hike.

Monetary Policy Transmission Mechanism in Nigeria: A Comparative Analysis

Empirical evidence has shown that the result of the exact channel of monetary policy transmission in Nigeria is mixed. This paper compared the Factor-augmented vector–auto regression (FAVAR) framework which exploits large data set of 53 with the traditional VAR model that estimates 6 variables to ascertain the exact channel of transmission. Findings from the two models conclude that although both methods generate qualitatively related results, but the FAVAR model is a superior alternative over VAR on grounds that monetary policy shocks are better identified using the FAVAR model. Also the FAVAR model does not exhibit the prize puzzle problem found in the VAR but allows for the computation of impulse responses of a large number of variables. Results from both models show that interest rate and credit channels are dominant and strongest channels of monetary policy transmission in Nigeria. Exchange rate and money channels were not significant and pronounced.

What happens when the Kiwi flies? Sectoral effects of exchange rate shocks on the New Zealand economy

We use a data-rich approach, a factor-augmented vector autoregression (FAVAR), to identify idiosyncratic exchange rate shocks and examine the effects of these shocks on different sectors of the New Zealand economy. We find that an unexpected shock to the exchange rate has significant effects on relatively tradable sectors of the economy. Whilst this is expected, relatively ‘more’ non-tradable sectors of the economy are also influenced by shocks to the exchange rate, presumably due to their linkages to more trade-exposed sectors. We also find that exchange rate shocks explain a small proportion of overall business cycle variability, implying that the exchange rate acts as a buffer rather than as a source of shock.

Empirical identification of factor models

In the conventional factor-augmented vector autoregression (FAVAR), the extracted factors cannot be used in structural analysis because the factors do not retain a clear economic interpretation. This paper proposes a new method to identify macroeconomic factors, which is associated with better economic interpretations. Using an empirical-based search algorithm, we select variables that are individually caused by a single factor. These variables are then used to impose restrictions on the factor loading matrix, and we obtain an economic interpretation for each factor. We apply our method to time-series data in the USA and further conduct a monetary policy analysis. Our method yields stronger responses of price variables and muted responses of output variables than what the literature has found.