To Bi, or not to Bi? Differences in Spillover Estimates from Bilateral and Multilateral Multi-country Models

Abstract

Asymptotic analysis and Monte Carlo simulations show that spillover estimates obtained from widely-used bilateral (such as two-country VAR) models are significantly less accurate than those obtained from multilateral (such as global VAR) models. In particular, the accuracy of spillover estimates obtained from bilateral models depends on two aspects of economies' integration with the rest of the world. First, accuracy worsens as direct bilateral transmission channels become less important, for example when the spillover-sender accounts only for a small share of the spillover-recipient's overall integration with the rest of the world. Second, accuracy worsens as indirect higher-order spillovers and spillbacks become more important, for example when the spillover-recipient is more integrated with the rest of the world overall. Empirical evidence on the global output spillovers from US monetary policy is consistent with these generic results: Spillover estimates obtained from two-country VAR models are systematically smaller than those obtained from a global VAR model; and the differences between spillover estimates obtained from two-country VAR models and a global VAR model are more pronounced for economies for which the US accounts for a smaller share of their overall trade and financial integration with the rest of the world, and for economies which are more integrated with the rest of the world overall. JEL codes: F44, C33

International Effects of Euro Area Versus U.S. Policy Uncertainty: A Favar Approach

“…In this regard, we use the FAVAR as a multicountry framework, assuming that developments of economic variables share strong similarities which can be approximated by common factors. Our empirical approach is also motivated and encouraged by the results of Georgiadis (). His results show that multicountry models such as the FAVAR (and GVAR, global vector autoregressive) are, in contrast to bilateral VARs, more appropriate to model global and regional shocks as well as cross‐country spillovers, because these models are capable of modeling higher‐order spillovers.…”

Exchange Rate Pass-Through and Cross-Country Spillovers: Some Evidence from Ukraine and Russia

“…Although we find that our results are robust to different specifications and consistent with existing studies, identification of spillover effects from the bilateral perspective may result in possible practical consequences. As argued in Georgiadis (2015) spillover estimates obtained from bilateral models are in general inconsistent asymptotically and less accurate than those obtained from a multilateral model in finite samples due to omitted variable bias and failure to account for higher-order transmission channels. In this respect, our results from the bilateral perspective may not indicate all possible cross-country interrelations in the globalized world.…”

What Goes Around Comes Around: How Large Are Spillbacks From US Monetary Policy?

“…15 Brute force alternatives for carrying out counterfactual analysis in VAR models are to set to zero autoregressive parameters after or before estimation (see, for example, Ramey, 1993;Vicondoa, 2019;Degasperi et al, 2020;Dees and Galesi, 2021). However, setting to zero VAR coefficients before estimation implies a mis-specified empirical model and induces biased estimates; in general, the bias is not informative about the strength of the channel that is being shut down (Georgiadis, 2017). In turn, setting to zero VAR coefficients after estimation may be understood similarly as the MRE approach in the sense that it reflects some counterfactual VAR model.…”