The J-Curve Effect in Services Trade: A Disaggregated Analysis
| Author | Ivan D. Trofimov |
| DOI | http://doi.org/10.1177/00157325221126598 |
| Published date | 01 May 2023 |
| Date | 01 May 2023 |
| Subject Matter | Articles |
The J-Curve Effect in
Services Trade:
A Disaggregated
Analysis
Ivan D. Trofimov1
Abstract
This article examines the relationship between the real exchange rate and the ser-
vices trade balance in the UK using the quarterly data for the 2005Q1–2019Q4
period. We consider the aggregate as well as disaggregated trade across five
services categories. We employ linear and non-linear autoregressive distributed
lag (ARDL) models. The non-linear model, in particular, allows consideration of
asymmetric response of the trade balance to depreciations as opposed to appre-
ciations. The findings indicate little evidence of a long-term improvement in the
trade balance following depreciation, and suggest the absence of J-curve effect.
The effects of domestic, ‘rest of the world’ GDP and monetary base on the trade
balance were respectively negative, positive and mixed.
JEL Codes: F14, F31, C22
Keywords
J-curve, cointegration, ARDL, services
Introduction
As far as the relationship between the exchange rate and trade balance is con-
cerned, the international trade and finance models assume the positive response of
the trade balance to currency depreciation, provided that the sum of the price
elasticities of domestic and foreign demand for imports exceeds one, and
both price elasticities of domestic and foreign supply of exports are infinite
Article
Foreign Trade Review
58(2) 199–219, 2023
© 2022 Indian Institute of
Foreign Trade
Reprints and permissions:
in.sagepub.com/journals-permissions-india
DOI: 10.1177/00157325221126598
journals.sagepub.com/home/ftr
1Kolej Yayasan Saad (KYS) Business School, Ayer Keroh, Melaka, Malaysia.
Corresponding author:
Ivan D. Trofimov, Kolej Yayasan Saad (KYS) Business School, KYS Business School Ayer Keroh
Country Resort, Lebuh Ayer Keroh, 75450 Ayer Keroh, Melaka, Malaysia.
E-mail: ivan.trofimov1@gmail.com
200 Foreign Trade Review 58(2)
(Marshall–Lerner condition). This result is likely to hold in the long-run, but the
short-run adjustments in the balance are usually more complex. The J-curve
hypothesis assumes that depreciation (devaluation) of the domestic currency will
produce deterioration of the domestic trade balance immediately after the event,
followed by its improvement further down the track.
This article examines the presence of J-curve in the aggregate services trade of
the UK, as well as in a number of services sub-sectors (transportation, travel,
goods-related, commercial and other services), using the quarterly data that
stretches 2005Q1–2019Q4 period. The empirical literature has tested both the
J-curve hypothesis and Marshall–Lerner condition extensively for the aggregate
and disaggregated trade in goods, in a variety of economies, using a plethora of
econometric methods, model specifications, different definitions of the dependent
variable, generally yielding mixed results (Akoto & Sakyi, 2019; Bahmani-
Oskooee & Ratha, 2004; Bhat & Bhat, 2021; Hsing, 2010; Lal & Lowinger, 2002;
Tandon, 2014). As far as services and international trade link is concerned, the
conventional view has been that services are non-tradeables, resulting in the lack
of testing of hypotheses that were in contrast tested extensively in the manufactur-
ing trade context. In this regard, the four empirical studies of the J-curve effect in
services stand out. While Wijeweera and Dollery (2013) found some evidence of
the J-curve effect for Australian services sector, Prakash and Maiti (2016), in the
study of Fijian trade balance and Yazici (2010) in Turkish services trade, demon-
strated that services trade balance is not highly sensitive to exchange rate changes.
Cheng (2020) presented mixed findings: the J-curve effect was present in the total
services trade, but not in every services category. The aim of the article is there-
fore to address these gaps in the literature.
There were no studies of the J-curve in services conducted for the UK context,
which is a notable shortcoming, given that as of 2018 the services constituted 46%
of UK exports, a high figure in comparison to services exports in other developed
economies (Jozepa et al., 2019, p. 8). An important feature of UK services trade is
its absolute value (the value of services exports and imports of 297 billion and 193
billion of pound sterling respectively in 2018), the prominent position of the UK
internationally as the second largest exporter of services in the world and prime
player in finance, insurance, air and sea transport, and higher education, as well as
the scale of diversification of services trade, both in terms of the number of sub-
sectors and export and import destinations (albeit professional and business ser-
vices constitute roughly one third of total exports or imports of services).
In the remainder of the article we describe the model, the data and econometric
methodology in the second section, then present the empirical results in the third
section, and provide the concluding remarks in the fourth section.
Methodology
Model
We assume that trade balance takes the following general form:
To continue reading
Request your trial