Exchange Market Pressure and Monetary Policy in Emerging Market Economies: New Evidence from Treatment‐effect Estimations

• Author: Alexander Erler,Steffen Sirries,Bernhard Herz,Christian Bauer
• Published date: 01 August 2015
• Date: 01 August 2015
• DOI: http://doi.org/10.1111/rode.12169

Exchange Market Pressure and Monetary Policy in

Emerging Market Economies: New Evidence from

Treatment-effect Estimations

Alexander Erler, Steffen Sirries, Christian Bauer, and Bernhard Herz*

Abstract

The way central banks react to exchange market pressure is likely to affect the subsequent economic devel-

opment and the associated economic costs. In a situation of currency pressure the central bank can in prin-

ciple decide to let the currency float freely, to maintain the peg or to implement a managed float policy, i.e.

a mix of depreciation and intervention. As the central bank’s choices are subject to self selection and

endogeneity, we use propensity score matching to adequately cope with these methodical challenges. We

find that monetary authorities have two options to keep down the economic costs in terms of output,

namely stabilizing the exchange rate or letting the currency float freely. In contrast, a managed float under

currency pressure is accompanied by the worst possible outcome with an average loss of gross domestic

product (GDP) between 5% and 6%.

1. Introduction

Negative shocks to the economy can put a country’s currency under pressure in the

exchange market and increase the risk of a currency crisis. Such currency crises are

typically seen as calamitous events as they are quite often accompanied by dramatic

developments in the exchange markets and can be associated with high economic

costs in terms of output losses. However, a closer look reveals that exchange market

crises are very heterogeneous events with economic consequences ranging from deep

recessions to even positive growth effects (see, e.g. Gupta et al., 2007). A comprehen-

sive literature has analyzed the determinants of these economic effects mainly focus-

ing on the usual fundamental suspects, e.g. pre-crisis inflows, capital market

liberalization, external debt (see, e.g. Calvo and Reinhart, 2000; Chiodo and Owyang,

2002; Hong and Tornell, 2005; Gupta et al., 2007; Lahiri and Végh, 2007; Bussière et

al., 2010). Another potentially important determinant, namely the exchange rate and

monetary policy of the central bank, has not caught much attention so far with a few

exceptions (see, e.g. Eichengreen and Rose, 2003; Erler et al., 2014). However, the

central bank’s monetary regime is likely to affect the economic effects of pressure in

the foreign exchange market. In principle, the monetary authority has three options:

it can (i) let the currency float freely, i.e. it lets the currency depreciate, (ii) intervene

in the market to stabilize the exchange rate, or (iii) implement a managed float policy,

i.e. combining (limited) interventions with some degree of depreciation. As exchange

* Erler: University of Bayreuth, Faculty of Law and Economics, 95440 Bayreuth, Germany. Tel: +49-

(0)921-55-6320; Fax: +49-(0)921-55-6322; E-mail: alexander.erler@uni-bayreuth.de. Sirries, Herz: Univer-

sity of Bayreuth, Faculty of Law and Economics, Bayreuth, Germany. Bauer: University of Trier,

Germany. The authors would like to thank conference participants at Bath, Bayreuth, Leipzig and Mumbai

for many helpful comments. In particular, the very helpful comments by Ansgar Belke, Benedikt Heid,

Mario Larch, Ajay Shah, and Lukas Vogel are gratefully acknowledged. The authors would also like to

thank two anonymous reviewers for their very useful comments and suggestions.

Review of Development Economics, 19(3), 470–485, 2015

DOI:10.1111/rode.12169

© 2015 John Wiley & Sons Ltd

market turmoils are a recurring phenomenon in emerging market economies

(International Monetary Fund (IMF), 2009), this policy choice is of particular impor-

tance for these countries. Emerging economies are typically characterized by rela-

tively segmented and less developed currency and capital markets that are vulnerable

to sudden stops or even reversals of capital inflows, e.g. owing to an expected mon-

etary tightening in industrialized economies as seen in the “taper tantrum” in summer

2013 (see, e.g. Neely, 2014). With respect to their monetary regime these countries are

likely to be characterized by relatively weak institutions so that transparency is an

important issue in implementing an adequate policy. An exchange rate peg can be an

interesting approach for the central bank to communicate and conduct a policy of

tying one’s hand to overcome these deficiencies (Aizenman and Glick, 2005, p. 2).

Analyzing the effects of a central bank’s policy in times of exchange rate pressure is

obviously subject to selection bias and endogeneity problems. Selection bias is likely

to be relevant because countries that decide to let the exchange rate float freely

(depreciation) tend to have different economic fundamentals than countries that fend

off the pressure on the exchange rate (peg). A theoretical foundation for these differ-

ences is provided by second generation currency crisis models, which analyze how a

central bank that faces exchange market pressure evaluates the expected costs and

benefits of stabilizing a peg (see, e.g. Jeanne, 2000). The monetary authority takes

into account how an intervention policy via the necessary restrictive policy affects the

subsequent economic development, e.g. through negative wealth effects (see, e.g.

Jeanne, 2000). Thus, the central bank decides to let the currency float freely if the

expected economic costs of the depreciation are lower than the expected costs of

maintaining the exchange rate peg. As a consequence of this selection bias,

endogeneity problems arise, which in turn could lead to inconsistent estimation

results.

Propensity score matching (PSM) might be an attractive approach to adequately

cope with these methodical challenges. In economics, PSM has so far been extensively

used in the field of labor economics, for instance to evaluate the so-called treatment

effects of new labor market programs. In contrast, this approach is fairly new in the

research on international economics and finance.1

In the following we want to adapt this promising approach to analyze central bank

policy choices in times of exchange market pressure. The idea to use PSM can be

summarized as follows. Obviously, countries are not randomly subject to policy

choices, i.e. the treatment—in particular the policy choice—is nonrandom and the

treatment evaluation cannot be based on an experimental design with a random

assignment of the treatments. Economies subjected to a treatment—for example the

decision to let the currency float freely—and non-treated economies may therefore

differ not only in their treatment status but also in country specific characteristics,

which could affect the treatment participation as well as the effects of a treatment. To

eliminate biases that would stem from different characteristics, which influence the

strategy a central bank implements, PSM attempts to match each treated economy

with a “similar” non-“treated” economy to isolate the economic impact of the policy

response. PSM should therefore yield reliable estimates of the difference in the post-

exchange market pressure outcome between economies that implement a certain

policy strategy and those, which do not (Heinrich et al., 2010).

The studies closest in methodology to our empirical analysis are Forbes and Klein

(2013) and Forbes et al. (2013). Forbes and Klein (2013) analyze and compare the

economic consequences of policy measures, which were implemented as a response to

two (global) financial crisis events.2In a similar way Forbes et al. (2013) analyze the

EXCHANGE MARKET PRESSURE AND MONETARY POLICY 471

© 2015 John Wiley & Sons Ltd