Does the Exchange Rate Respond to Monetary Policy in Mexico? Solving an Exchange Rate Puzzle in Emerging Markets

• Published date: 01 December 2023
• Author: PAVEL SOLÍS
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/jmcb.13032

DOI: 10.1111/jmcb.13032

PAVEL SOLÍS

Does the Exchange Rate Respond to Monetary

Policy in Mexico? Solving an Exchange Rate Puzzle

in Emerging Markets

This paper argues that the null or weak response of emerging market cur-

rencies to domestic monetary policy documented in the literature is the re-

sult of wide event windows. An event study with intraday data for Mexico

shows that an unanticipated tightening appreciates the currency and attens

the yield curve, consistent with the evidence for advanced economies. With

daily event windows, however, only the yield curve responds to monetary

policy.Noise in daily exchange rate returns explains the lack of response of

the currency. Such noise gives rise to a bias that declines after controlling

for potential omitted variables.

JEL codes:E43, E52, E58, F31, G14

Keywords:monetary policy, exchange rate, yield curve,emerging

markets, high-frequency data, event study

T     monetary policy in

emerging markets is an open question. Standard open economy models suggest

that an increase in the policy rate leads to an immediate appreciation of the cur-

rency (Dornbusch 1976). Contrary to this prediction, early evidence for advanced

economies (Grilli and Roubini 1995) found that contractionary monetary policy leads

to a currency depreciation, which was referred to as the exchange rate puzzle. This

puzzle owes to the assumptions made to identify the monetary policy surprises, giv-

I am particularly grateful to Jonathan Wright, the editor Kenneth West, and two anonymousreferees.

I also thank Julien Acalin, Derin Aksit, Laurence Ball, Enrique Batiz, Lalit Contractor, Jorge Luis García

Ramírez, Olivier Jeanne, Fabrizio López Gallo Dey,Claudia Ramírez Bulos, Alessandro Rebucci, Jeong-

won Son and seminar participants at Banco de México and Johns Hopkins University for their helpful

comments and suggestions. The views in this paper are the sole responsibility of the author and should

not be interpreted as reecting the views of Banco de México or any other person associated with it. All

remaining errors are mine.

pavel.solis@banxico.org.mx

P S is a Senior Economist, Financial Stability Division, Banco de México (E-mail:

pavel.solis@banxico.org.mx).

Received December 1, 2021; and accepted in revised form October 4, 2022.

Journal of Money, Credit and Banking, Vol. 55, No. 8 (December 2023)

© 2023 The Ohio State University.

2094 :MONEY,CREDIT AND BANKING

ing rise to a problem of reverse causality (Zettelmeyer 2004).1Event studies with

high-frequency data identify exogenous changes in the policy rate and show that a

tightening indeed leads to an appreciation of the currencies of advanced economies;

this effect is detected using intraday (Andersen et al. 2003, Kearns and Manners

2006, Faust et al. 2007) and daily (Wright 2012, Ferrari, Kearns, and Schrimpf 2021)

event windows, and even exhibits persistence over subsequent days (Rosa 2011a,

Ferrari, Kearns, and Schrimpf 2021). For emerging markets, however, event stud-

ies with daily data show that the currency response to monetary policy is low or

nonexistent (Akta¸s et al. 2009, Duran et al. 2012, Pennings, Ramayandi, and Tang

2015). Kohlscheen (2014) identies this as an exchange rate puzzle in emerging

markets.

The null or weak response of the currency to monetary policy in emerging markets

reported in the literature (the puzzle) raises the question of whether their central banks

actually exert an inuence on their own currencies. The question is relevant for three

reasons. First, the transmission of monetary policy via the exchange rate is vital for

open economies. Second, the sensitivity of the currencies of advanced economies

to monetary policy increased after the global nancial crisis (Ferrari, Kearns, and

Schrimpf 2021), even in countries who continued to use conventional tools—like

Australia and Canada—and so it would be striking if emerging market currencies

remain insensitive to monetary policy. Third, the currencies of emerging markets do

respond to foreign monetary policysurprises (Hausman and Wongswan 2011, Kearns,

Schrimpf, and Xia 2022).

This paper studies whether and how the exchange rate responds to monetary pol-

icy in a representative emerging economy. I use an event study methodology and a

new data set of intraday changes in asset prices bracketing all regular monetary pol-

icy announcements in Mexico from 2011 to 2021. Mexico is a small open economy

with relatively liquid nancial markets, a market-based exchange rate, and a cred-

ible ination targeting regime, criteria that make it a reasonable candidate for this

type of analysis (Kearns and Manners 2006, Pennings, Ramayandi, and Tang 2015).

By now, event studies with high-frequency data are a well-established strategy in

macronance to overcome endogeneity concerns because they isolate the surprise

component of policy decisions (Gürkaynak and Wright 2013, Nakamura and Steins-

son 2018). Nevertheless, they have rarely been applied to study the transmission of

monetary policy to asset prices in Mexico.2

1.The traditional approach to identify monetary policy surprises is to estimate a vector autoregres-

sion model using a recursive assumption (see Christiano, Eichenbaum, and Evans 1999). The exchange

rate puzzle is a well-known feature of this approach. Kim and Lim (2016) nd similar results for emerg-

ing markets.

2.For the Mexican case, event studies have been used to analyze the effects of foreign monetary

policy on asset prices (Borensztein, Zettelmeyer, and Philippon 2001, Hausman and Wongswan 2011,

Rosa 2011b, Kearns, Schrimpf, and Xia 2022) and portfolio ows (Hernandez-Vega 2021), and whether

ination expectations are well-anchored (De Pooter et al. 2014). Kohlscheen (2014) includes Mexico to

study the exchange rate response to monetary policy,but does not use intraday data nor swaps to measure

surprises in the policy rate as in this paper.