Breaking the UIP: A Model‐Equivalence Result

• Published date: 01 September 2022
• Author: YOSSI YAKHIN
• Date: 01 September 2022
• DOI: http://doi.org/10.1111/jmcb.12883

DOI: 10.1111/jmcb.12883

YOSSI YAKHIN

Breaking the UIP: A Model-Equivalence Result

Breaking the uncovered interest rate parity (UIP) is essential to accounting

for exchange rate dynamics, and is required for modeling sterilized foreign

exchange interventions. Gabaix and Maggiori (2015) account for many ex-

change rate puzzles by introducing nancial intermediaries that absorb sav-

ings imbalances for a premium, thereby deviating from the UIP.Fanelli and

Straub (2021) analyze foreign exchange interventions. In their model, reg-

ulatory exposure limits and participation cost in the international nancial

markets drive a wedge in the UIP. This paper demonstrates that, to a rst-

order approximation, introducing a reduced-form portfolio adjustment cost

is isomorphic to these modeling strategies.

JEL codes: E58, F3, F41

Keywords: UIP, nancial frictions, open economy macroeconomics

B     parity (UIP) is

essential to accounting for the empirical behavior of exchange rates,1and it is a pre-

requisite in theoretical models for the efcacy of sterilized foreign exchange inter-

vention (FXI) through the portfolio balance channel. While the inuential work of

Backus and Kehoe (1989) points to the inefcacy of sterilized FXIs, recent contri-

butions have revived the argument for their use, for example, Benes et al. (2015),

Cavallino (2019), Alla et al. (2020), and Fanelli and Straub (2021). These models

introduce nancial frictions that differentiate domestic bonds from foreign ones; oth-

erwise, to a rst-order approximation, the two assets are perfect substitutes, the UIP

holds, and sterilized FXIs are deemed ineffective.

I thank Jonathan Benchimol, Eliezer Borenstein, Zvi Hercowitz, AlexIlek, Nimrod Segev, Osnat Zohar,

and seminar participants at the Bank of Israel for helpful comments and discussion. I am also grateful to

an anonymous referee for excellent comments. All errors are mine. The views expressed in this paper are

solely my own and they do not necessarily represent those of the Bank of Israel.

Y Y is a Senior Economist at the Research Department, Bank of Israel (E-mail:

Yossi.Yakhin@boi.org.il).

Received March 25, 2020; and accepted in revised form October 31, 2021.

1. See Engel (2014) and references therein.

Journal of Money, Credit and Banking, Vol. 54, No. 6 (September 2022)

© 2021 The Authors. Journal of Money, Credit and Banking published by Wiley Periodicals

LLC on behalf of Ohio State University.

This is an open access article under the terms of the Creative Commons Attribution-NonCom-

mercial-NoDerivs License, which permits use and distribution in any medium, provided the

original work is properly cited, the use is non-commercial and no modications or adaptations

are made.

1890 :MONEY,CREDIT AND BANKING

The contribution of Gabaix and Maggiori (2015), GM hereafter, sketches the mi-

crofoundations of a mechanism that introduces a wedge between the domestic and

foreign interest rate differential and the expected exchange rate movement, thereby

deviating from the UIP condition. In their model, international nancial markets are

segmented and nanciers are willing to absorb savings imbalances for a premium,

which, in turn, breaks the UIP. GM forcefully demonstrate that their model can help

rationalize some of the long-standing empirical exchange rate puzzles, including the

exchange rate disconnect and the forward premium puzzle. In another important con-

tribution, Fanelli and Straub (2021), FS hereafter, lay down principles for FXIs. The

microfoundations of the nancial friction in their model rely on regulatory restrictions

and nancier-specic participation cost in the international nancial markets, similar

to Alvarez et al. (2009). In FS, it is the participation cost of the marginal nancier

that determines the gap between the interest rate differential and the expected change

in the exchange rate. Due to the regulatory limits, movements in the foreign asset

position of the economy are associated with changes in the identity of the marginal

nancier, and hence with the marginal participation cost and the size of the deviation

from the UIP.

This paper derives an equivalence result: To a rst-order approximation, the GM

and FS models are isomorphic to a standard reduced-form portfolio adjustment cost

model, as in Schmitt-Grohé and Uribe (2003), SGU hereafter. In SGU, the purpose of

the friction is to impose stationarity in small open economy models. They achieve it

by endogenizing the effectiveforeign return faced by the agents in the small economy,

as by assumption these agents bear a cost whenever their foreign asset position de-

viates from some benchmark level. Uribe and Yue (2006) provide microfoundations

to SGU’s setup. Since in GM and FS, as in SGU, movements in the foreign asset

position generate a time-varying wedge between the global risk-free rate and the ef-

fective foreign return faced by domestic agents, it is not surprising to nd that the

UIP conditions in these models are closely related. SGU’s simple modeling strategy

is therefore robust to different underlying microstructures.

The implication of the equivalence result is that, to the extent that one is only

concerned with rst-order dynamics and second moments of macrovariables—as is

typically the case in the open economy business cycle literature and in many new-

Keynesian applications—there is no gain from adopting the rich microstructure of

either GM or FS; a simple portfolio adjustment cost, as in SGU, is just as good.

Moreover, the underlying microfoundations of GM and FS do not carry into higher

order terms in the UIP condition, suggesting that the simplicity of the rst-order ap-

proximation does not sacrice important higher order dynamics.

The exposition of the models in this paper is deliberately lean, and they contain

the minimal structure needed for discussing deviations from the UIP. The models ab-

stract from production and use one global good, suggesting that the real exchange

rate is xed at unity. The nominal exchange rate reects the relative price of curren-

cies, which are only used as units of account in the domestic and foreign markets.

Prices are exible, resulting in neutral monetary policy. The models are focused on

the specication of the nancial frictions that generate deviations from the UIP, and