Yeah

DIPARTIMENTO DI SCIENZE ECONOMICHE

n.142

Giulio Cifarelli
and
Giovanna Paladino
THE INTERNATIONAL RESERVES GLUT:
IS IT FOR REAL?

January 2006

UNIVERSITA’ DEGLI STUDI DI FIRENZE

Direttore responsabile: Prof. Giovanni Andrea Cornia
Comitato di Redazione: Proff. Antonio Gay, Piero Innocenti, Alessandro
Pacciani, Piero Roggi
Coordinatore Scientifico: Dott. Nicola DoniPubblicazione depositata a norma di legge

1

The International Reserves Glut: Is It for Real?

Giulio Cifarelli* and Giovanna Paladino**
January 2006

Abstract
Monthly data from January 1985 to December 2004 are used to investigate reserves management in
ten Asiatic and Latin American countries. Idiosyncratic explanatory variables enter cointegration
relationships based on a stochasticbuffer stock model, where a reserve variability measure is obtained
via conditional variance approaches. International factors influence the cointegration residuals
(representing

the excess demands for reserves), which tend to co-move within and across

geographical areas. Principal components analysis is implemented then to associate their common
drivers with the US fed fund effective interestrate and real effective exchange rate. This two-step
approach sheds light on some controversial aspects of reserves and exchange rate management in
emerging markets such as “fear of floating” and mercantilist behavior. Our results suggest, contrary to
common belief, that the size of recent excess reserves holdings is probably overstated.
Keywords: Emerging Markets’ International Reserves,Cointegration Analysis, Principal Components Analysis.
JEL Code: F310, F340, G150

_____________________________
* University of Florence, Economics Department, via delle Pandette 9, 50127 Florence, Italy.
giulio.cifarelli@unifi.it
** LUISS University and Sanpaolo IMI Economic Research Dept., viale dell’Arte 25, 00144 Rome, Italy.
giovanna.paladino@sanpaoloimi.com

Introduction
The averagereserve holdings of several emerging markets have risen in recent years,
irrespectively of their official exchange rate regime, both in absolute terms and as percentages
of GDP or of standard international trade and finance adequacy indicators, such as the level of
imports or the short term external debt. They have reached $1.8 trillion at the end of 2004, up
roughly $800 billion from 2002and more than four times their level in 1994, exceeding the
potential requirements of any foreseeable shock. So what is the purpose of possessing
international reserves in excess of common measures of adequacy? This puzzle has attracted
considerable attention from both practitioners and academics. Precautionary behavior of
central banks, fear of the disruptive effects of an exchange ratedepreciation (motivated by the
experience of the financial crises and sudden capital reversals of the 1990s), difficult access to
international capital markets and mercantilist export support are possible explanations set out
in a burgeoning literature.
In this paper monthly data from January 1985 to December 2004 are used to investigate the
national idiosyncratic and international determinants ofreserve changes in five Asiatic and
five Latin American countries. Idiosyncratic explanatory variables are mostly associated with
the tenets of the benchmark buffer stock model of Frenkel and Jovanovic (1981) while
international explanatory factors reflect the pivotal role played by the U.S. monetary
authorities in emerging markets finance.
The use of a monthly frequency has a number ofdrawbacks due to the non stationary nature
of the time series and to the difficulty of estimating consistent cointegration relationships and
error correction structures. The twenty year data span, however, provides a large number of
observations and allows to estimate the relevant relationships for each country in isolation.
Most previous empirical studies (Flood and Marion, 2002, Aizenman and...