Capital‐flow volatility in emerging markets: A panel GARCH approach

• Published date: 01 August 2023
• Author: Ahmet Ihsan Kaya,Lutfi Erden
• Date: 01 August 2023
• DOI: http://doi.org/10.1111/infi.12427

Received: 3 September 2021

|

Accepted: 26 October 2022

DOI: 10.1111/infi.12427

ORIGINAL ARTICLE

Capital‐flow volatility in emerging markets:

A panel GARCH approach

Ahmet Ihsan Kaya

1,2

|Lutfi Erden

2

1

Economics and Prosperity, British

Embassy, Ankara, Turkey

2

Department of Economics, Hacettepe

University, Ankara, Turkey

Correspondence

Ahmet Ihsan Kaya, Economics and

Prosperity, British Embassy, Sht. Ersan

Cd. 46/A, 06680 Cankaya/Ankara,

Turkey.

Email: ahmetihsan.kaya@fcdo.gov.uk

Abstract

This study analyzes the role of push–pull factors on the

level, volatility and comovement of capital flows in

emerging markets (EMs). Taking the commonality of

capital flows into account, we employ the panel

Generalized Autoregressive Conditional Heteroscedas-

ticity model developed by Cermeño and Grier for 16

EMs. This method not only accounts for country‐

specific heterogeneity and cross‐section dependence

but also allows the examination of the sources of the

level, volatility and comovement of capital flows in a

single step. The results show that domestic factors

explain two‐thirds of the variation in net capital‐flow

volatility. While both global and domestic factors, with

the prominent ones being global risks and domestic

economic growth, influence the comovement, their

impacts somewhat vary by the types of capital flows.

KEYWORDS

capital flows, comovement, panel GARCH, push–pull factors,

volatility

JEL CLASSIFICATION

C33, F21, F37

1|INTRODUCTION

Foreign capital has been a major funding source for most economies in the last few decades.

However, it could also be detrimental for recipient countries, as international capital flows have

become rather volatile, especially after the global financial crisis (GFC) in 2008–2009. Empirical

evidence shows that capital‐flow volatility is considerably higher in emerging markets (EMs)

International Finance. 2023;26:172–188.wileyonlinelibrary.com/journal/infi172

|

© 2022 John Wiley & Sons Ltd.

and developing economies (Broner & Rigobon, 2004; Pagliari & Hannan, 2017). Compared to

advanced economies, they are more prone to the adverse effects of flow volatility on

macroeconomic and financial stability, particularly due to their lack of well‐established

institutions and well‐developed financial markets. Thus, dealing with the volatility of these

flows has become an important policy concern for those countries. The main question is

whether there is any course of action that host countries can take to alleviate the disruptive

impacts of capital‐flow volatility and insulate their economies against unfavorable global

financial conditions. In addition to its volatile nature, the degree of commonality in

international capital flows has risen in recent decades (Reinhart et al., 2016). Obviously, the

success of domestic policies in reducing financial instabilities resulting from capital‐flow

volatility also depends closely on the degree of flow synchronization. Hence, it is essential for

policymakers and academics to put more effort into understanding the drivers of both the

volatility and the comovement of flows.

Following the influential work by Calvo et al. (1993), one line of research in this literature

considers many pull (domestic)–push (global) factors to investigate the determinants of the

volatility of capital flows. For this purpose, the previous studies mostly adopt a two‐step

approach by focusing on the conditional or unconditional variance properties of capital flows

(Broto et al., 2011; Lee et al., 2013; Neumann et al., 2009; Pagliari & Hannan, 2017). In the first

step, the time‐varying volatility for individual countries based on standard deviations,

Autoregressive Integrated Moving Average (ARIMA) or Generalized Autoregressive Condi-

tional Heteroscedasticity (GARCH) models is obtained. In the second step, the sources of this

volatility are analyzed with an application of panel regression models. Another line of research

focuses on the other second‐moment properties; namely, covariance or cross‐section

correlations, investigating whether capital flows of different countries comove. These studies

mostly employ some form of dynamic factor analysis (two‐level, multilevel, latent and

hierarchical) and generally report a high degree of commonality of capital flows, pointing to the

presence of a global financial cycle (Barrot & Servén, 2018; Kaminsky, 2019; Lafuerza &

Servén, 2019; Lee et al., 2013; Rey, 2015) except for the studies by Förster et al. (2014) and

Cerutti et al. (2019).

The present study combines these two strands of the literature, focusing on the second‐

moment properties (variance and covariance/correlation) of capital flows, into a novel

empirical framework suitable for investigating the role of global and domestic factors

in both the volatility and comovement of capital flows. To this end, we employ the

panel GARCH model proposed by Cermeño and Grier (2006; also known as the dynamic

panel data model with conditional covariance), which not only explicitly incorporates

the cross‐section dependence (CD) into the analysis of volatility drivers, but also enables

us to investigate if some observable push and pull factors lead to the comovement of

capital flows. We estimate the panel GARCH model employing panel data from 16

emerging economies over the period 1995–2019. After analyzing the volatility dynamics of

gross inflows/outflows and net capital flows, we run the same models using the

subcategories of net flows, such as foreign direct investment (FDI), portfolio and other

flows. As the model is ideally suitable for capturing the potential sources for the

comovement of capital flows, we further investigate this issue by modeling the

covariances of capital flows, providing additional insights on the synchronization

of capital flows to EM. Overall, our results indicate that domestic factors explain two‐

thirds of the variation in net capital‐flow volatility, suggesting EMs are not unprotected

against the global financial cycle.

KAYA AND ERDEN

|

173