How do manufacturing exports react to the real exchange rate and foreign demand? The Chilean case

• Author: Miguel A. Fuentes D.,Jorge A. Fornero,Andrés Gatty Sangama
• DOI: http://doi.org/10.1111/twec.12891
• Published date: 01 January 2020
• Date: 01 January 2020

274

|

wileyonlinelibrary.com/journal/twec World Econ. 2020;43:274–300.

© 2019 John Wiley & Sons Ltd

Received: 16 May 2018

|

Revised: 12 April 2019

|

Accepted: 1 October 2019

DOI: 10.1111/twec.12891

ORIGINAL ARTICLE

How do manufacturing exports react to the real

exchange rate and foreign demand? The Chilean

case

Jorge A.Fornero

|

Miguel A.Fuentes D.

|

AndrésGatty Sangama

Banco Central de Chile, Santiago, Chile

KEYWORDS

Chile, exchange rates, foreign demand, manufacturing exports

1

|

INTRODUCTION

From 2013Q1 to 2017Q1, the growth of the main trading partners of Chile has gradually been mod-

erating while the Chilean peso has depreciated against the US dollar at a two-digit rate, around 33%,

while the real depreciation has been lower, 9%. A priori, lower growth in external demand should

slow down exports while a depreciation of the Chilean peso would result in greater competitiveness

of Chilean exports and, therefore, a higher growth of its volumes. Between 2013 and 2016, the vol-

ume of total exports accumulated a slightly negative variation, combining a drop of −1.5% in mining

shipments and −2% in manufacturing exports, and an increase close to 15% in agricultural exports.

Similarly, goods' exports have shown a limited dynamism in the years 2014–16 in comparison with the

average growth ofthe pre-Crisis 2008 period. These recent developments are framed within a context

where world trade growth has also shown poor dynamism.

Chile is a small open economy rich in natural resources. Its policy framework builds on two pil-

lars. First, independent monetary policy implemented with an inflation target and flexible exchange

markets since 2001 (Central Bank of Chile, 2007) and, second, a structural balance fiscal rule which

determines spending to meet balance targets. Standard measures of trade openness fluctuate easily

above 60% in last 20years. Recent figures of export's share (average 2014–18) indicate that mining is

the largest category weighting about 47% of the value of total exports (copper represents 90% of it),

and manufacturing is the second largest category with 33%, followed by agricultural exports whose

share amounts to about 7% of total exports. Finally, services' share is about 13%.

This paper is framed in the traditional trade-equation approach where the premise is that the exter-

nal demand and the real exchange rate (RER) have a primary role in influencing the domesticbalance

of resources through expenditure-switching effects, controlling by exogenous factors (Hinkle &

Montiel, 1999). For example, Central Banks usually include trade blocks in their models for various

purposes interests such as forecasting and policy analysis, which are calibrated using estimates of

|

275

FORNERO Et al.

trade elasticities. Estimating trade elasticities has been an old practice in international economics. A

challenge is that primary exporting sectors, namely mining and agricultural exports' growth, are

largely explained by its out production (and with idiosyncratic supply shocks), showing no correlation

with RER and foreign demand.1 This characteristic is shared by many commodity exporters' emerging

economies. Previous empirical evidence has found that Chilean manufacturing exports respond to the

RER and foreign demand.2 More recently, the literature has shifted the focus towards services' exports

due to its increasing importance in the world trade representing one-fourth of world trade (Loungani,

Mishra, Papageorgiou, & Wang, 2017).3

Given the previous evidence for Chile, this paper concentrates on updating estimates of manu-

facturing exports' elasticities, with small methodological improvements, which roughly yield similar

results to previous literature. In addition, to go deeper in the analysis we examine the domestic value

of manufacturing sectors' exports. Although this yields a complementary analysis, it does not replace

the interest in exports because value-added measures depend crucially on input–output tables that are

published with a lag of 2–3years reducing its scope for junctural analysis. In addition, not all countries

publish regularly these tables, limiting the scope of comparative examinations across countries. From

the academic point of view, observing value added of exports helps to calibrate more realistic trade

models with firm heterogeneity and also it can turn empirical work more appealing economically. In

our particular case, it is valuable to use it because it enriches our analysis of the sensitivity of exports,

because it filters out the negative effect in value caused by higher costs of intermediate inputs used

in exported good. The disadvantage of traditional exports' measures is that it gathers also value of

intermediate imported inputs that previously have crossed borders (value added generated elsewhere

abroad), which likely leads to exaggerate estimated RER elasticities. Our overall assessment is that for

the case of Chile, exports and domestic value added of exports respond essentially the same to RER

variation (we will expand on why this is somehow not surprising).

The motivation that raised the main question of the paper relates to recent international devel-

opments contemporary to the Taper Tantrum talk episode in 2013Q2. Following that date, we ob-

served higher volatility of financial asset prices, financial markets stress, re-orientation of capital

flows towards advanced countries and persistent currency depreciation of emerging markets' curren-

cies against the US dollar. In particular, from 2013Q1 to 2017Q1 the Chilean peso has depreciated

against the US dollar at a two-digit rate, around 33%, while the real depreciation has been lower: (a)

bilaterally with respect to the United States, it approximated 9%, and (b) in multilateral terms, it has

been even smaller, close to 5%. A priori, a depreciation of the Chilean peso would result in greater

competitiveness of Chilean exports and, therefore, a higher growth of its volumes. However, between

2013 and 2016, the volume of total exports accumulated a slightly negative variation, combining a

drop of −1.5% in mining shipments and −2% in manufacturing exports, and an increase close to 15%

in agricultural exports. Similarly, goods' exports have shown a limited dynamism in the years 2014–16

1 We find in regressions that lagged production in mining sector is the best predictor of the annual growth of mining exports.

The same applies for agricultural exports’ growth. Results are available upon request.

2 Main findings are summarised in Table 1. Briefly, we comment that our findings suggest a significant correlation of

manufacturing exports with foreign demand and BRER in the short run. In the long run, it co-integrates with foreign demand,

but the BRER in levels does not exert significant effects on the level of manufacturing exports. These results are not

surprising given that the BRER behaves as integrated of order zero in our 2003–16 sample (see Appendix B for the context of

international trade of Chile).

3 Sheridan (2014) finds that increasing manufacturing exports is important for supporting a vigorous economic growth;

however, the evidence suggests that this implication only holds conditional on reaching a threshold level of development.