Surviving Asymmetry in Capital Flows and the Business Cycles: The Role of Prudential Capital Controls

• Published date: 01 August 2015
• Author: Yoke‐Kee Eng,Chin‐Yoong Wong
• Date: 01 August 2015
• DOI: http://doi.org/10.1111/rode.12174

Surviving Asymmetry in Capital Flows and

the Business Cycles: The Role of Prudential

Capital Controls

Chin-Yoong Wong and Yoke-Kee Eng*

Abstract

Past literature of different strands has pointed to a potential asymmetry: while portfolio capital inflows are

largely irrelevant to the economy, capital outflows can cause recession. In a model with a convex invest-

ment and portfolio balance adjustment cost, and endogenous credit-in-advance constraint, we find that

investment is determined solely by opportunity cost of physical capital unrelated to portfolio capital inflows

when the constraint is slack. However, once credit availability is tightened up by capital outflows, the nega-

tive liquidity constraint dominates the opportunity-cost factor, causing an economic downturn. Financial

fragility against capital outflows is an outcome of pecuniary externalities, which, however, can be moder-

ated by prudential capital controls. Even when exchange rates float freely, capital controls ease the macro-

stabilizing burden of monetary policy, as they help shield the economy from financial instability. Prudential

tax on foreign debt is most preferred, and works the best when the exchange rate float is managed.

1. Introduction

Cross-border capital flows have drawn enormous attention from academics and

policymakers since the waves of capital account liberalization began in the late 1980s.

Much has been written on the search for favorable effects of capital inflows on the

economy. While some have documented benign growth effects of capital inflows,

which have to be conditional on the existence of strong, good institutions (see, for

instance, Friedrich et al., 2013; Alfaro et al., 2007; Bekaert et al., 2005), robust evi-

dence can rarely be found, leaving alone unconditional positive effects (see Jeanne et

al., 2012; Kose et al., 2009; Gamra, 2009, for more skeptical views). Prasad et al.

(2007) and Gourinchas and Jeanne (2006) even show that the fastest growing coun-

tries did so without large foreign capital inflows.

In fact, an easier task is to demonstrate the flipside of the coin: capital flows more often

than not wreak havoc on the economy, which ends with a slump. Especially when the

economy is loaded with a large amount of foreign-currency denominated debts, a sudden

capital reversal that causes the value of the local currency to crash can devalue the worth

of collateral for borrowing, instigating an unfavorable balance-sheet effect of depreciation

* Wong: Faculty of Business and Finance, Universiti Tunku Abdul Rahman, Jalan Universiti, Bandar

Barat, 31900 Kampar, Perak, Malaysia. Tel: +60-5-468-8888; E-mail: wongcy@utar.edu.my. Also affiliated

to Stockholm China Economic Research Institute (SCERI), Stockholm School of Economics, Sweden;

Eng: Faculty of Business and Finance, Universiti Tunku Abdul Rahman, Malaysia. The paper was written

when the first author was hosted by SCERI as Visiting Research Fellow. He is grateful to SCERI for con-

ducive research environment and Lars-Erik Thunholms Stiftelse For Vetenskaplig Forskning for generous

funding. This research is also part of a project financially funded by Fundamental Research Grant Scheme

(FRGS) from the Ministry of Education, Malaysia (FRGS/2/2013/SS07/UTAR/02/1). An earlier version of

the paper was presented at the International Conference on “Exchange Rates, Monetary Policy and Finan-

cial Stability in Emerging Markets and Developing Countries”. The authors are grateful to two anonymous

referees for constructive comments that have helped sharpen their own understanding on the model.

Review of Development Economics, 19(3), 545–563, 2015

DOI:10.1111/rode.12174

© 2015 John Wiley & Sons Ltd

that dominates the conventional favorable expenditure-switching effect of depreciation

(Jeanne et al., 2012; Bordo et al., 2010). By using Hatemi-J’s (2012) asymmetric Granger

causality approach over selected Asian countries, Eng and Wong (2015) find that while

the null hypothesis that capital inflows do not Granger cause economic expansion cannot

be rejected, capital outflows precede economic slump with a high statistical significance.

In short, we may live in a world of asymmetry: while a rising tide may not lift all

boats, a retreating tide of capital flows may sink all boats. Understanding the mecha-

nism of asymmetry thus becomes warranted given the fact that gross capital inflows

and outflows are distinct in dynamics and implications (Forbes and Warnock, 2012;

Broner et al., 2013). More importantly, it poses a challenge to the design of prudential

capital controls as a tool to safeguard financial stability. A question of relevance is, for

instance, should capital inflows or outflows be controlled?

Extending from Eng and Wong (2015), which focuses on asymmetric growth effects of

capital flows in a Schumpeterian growth model with credit imperfection, this paper sheds

light on asymmetric business-cycle effects of gross capital flows, in particular, portfolio

and debt flows. We do so by laying out an otherwise standard two-country real business-

cycle model with convex investment and capital flow adjustment costs in section 2. While

the former setting allows us to derive Tobin’s qin physical investment that can be easily

expanded to explicitly capture the implications of the credit-in-advance constraint, the

latter gives us “Tobin’s q” in gross capital inflows and outflows, and foreign debt inflows

independently. Later, the model is parameterized in section 3.

That said, this paper has three theoretical novelties. First, we introduce an eternally

binding credit-in-advance constraint that depends on the borrower’s credit-

worthiness, which, in turn, is conditional on the borrower’s collateral–debt ratio with

Frechet distribution. Modeling the foreign-currency denominated borrowing con-

straint is nothing new, but is in the spirit of the literature on sudden stops (Bianchi,

2011; Mendoza, 2010; Korinek and Mendoza, 2014) and prudential capital control

(Korinek, 2011; Jeanne and Korinek, 2012; Benigno et al., 2013). The novelty,

however, lies in the way the constraint is modeled, which enables us to generate a

credit amplification mechanism that interestingly resembles the characteristics of an

occasionally binding constraint without succumbing to the difficulty in solving a

model with occasionally binding credit constraint.

More interesting is that the credit-in-advance constraint with endogenous credit-

worthiness sets the mechanism of asymmetry in motion, as elaborated in section 4.

Here is the intuition. When the credit constraint is slack owing to a stronger collateral

value denominated in foreign currency vis-à-vis foreign borrowing, thanks to capital

inflows that appreciate the local currency, the physical investment decision is deter-

mined by the conventional opportunity cost of physical investment with no role for

liquidity. However, in the incident of a persistent capital outflow that depreciates the

local currency and hence foreign-currency value of the collateral, debt availability is

limited by the lower bound of the credit-worthiness. The tightening liquidity effect is

likely to dominate the opportunity-cost channel, thereby constraining physical invest-

ment and causing accumulation of capital stock used as the collateral to slow down.

When this happens, the credit constraint is further tightened as the collateral–debt

ratio is further lowered, setting the stage for a business downturn.

Second, based on the concept that reversing the direction of capital flows incurs adjust-

ment costs, we formalize the dynamics of gross portfolio capital and foreign debt inflows

by foreign residents, and gross capital outflows by domestic residents. Such classification is

coherent with Forbes and Warnock (2012) and Broner et al. (2013), which empirically

shed light on the dynamics of varied gross capital flows by different agents. While yield

546 Chin-Yoong Wong and Yoke-Kee Eng

© 2015 John Wiley & Sons Ltd