Measuring and understanding trade in service tasks

• Author: Renato YSLAS,Daniel CHIQUIAR,Martín TOBAL
• DOI: http://doi.org/10.1111/ilr.12133
• Date: 01 March 2019
• Published date: 01 March 2019

International Labour Review, Vol. 158 (2019), No. 1

Copyright © The authors 2019

Journal compilation © International Labour Organization 2019

*

Directorate General of Economic Research, Banco de México, emails: dchiquiar@banxico.

org.mx (corresponding author); martin.tobal@banxico.org.mx; renato.yslas@banxico.org.mx. The

opinions expressed here are those of the authors and do not represent the opinions of Banco

de México or its board of governors. The authors wish to thank Pedro Martins, Pierre Sauvé,

Lucian Cernat, Huong Dinh, Stijn Vanormelingen, Christian Viegelahn, Anirudh Shingal, Kar-

ishma Banga and Johannes Schwarzer for their valuable comments.

Responsibility for opinions expressed in signed articles rests solely with their authors,

and publication does not constitute an endorsement by the ILO.

Measuring and understanding trade

in service tasks

Daniel CHIQUIAR,* Martín TOBAL* and Renato YSLAS*

Abstract. The revolution in information and communications technologies has put

service tasks with strong tradability characteristics at high risk of being offshored.

This article reviews studies proposing indicators of service tradability, exploring

the labour market implications of service offshoring and developing theories to

rationalize the facts. It suggests that both skill intensity and tradability are deter-

minants of wage and employment effects. Nonetheless, the lack of widely accepted

denitions of tradability, the absence of high-quality data on service trade ows

and the difculty of measuring import competition at higher disaggregation levels

pose difculties in achieving further progress, pointing to areas for future research.

By spreading the use of technologies such as the Internet, cell phones and

teleconferencing, the information and communications technology

(ICT) revolution has facilitated the electronic delivery of labour tasks. This

has resulted in a reduction in the costs of offshoring service tasks and a boom

in service offshoring. In this context, and given that service offshoring involves

moving service jobs from one country to another, it is natural to think that

the offshoring boom will have had implications for wages and employment

in various countries. Moreover, trade in services is expected to have signi-

cant consequences for labour markets, not only in aggregate terms, but also

through the creation of employment and wage redistribution effects within

trading countries.

The fact that trade in services affects labour market outcomes in ag-

gregate terms is suggested by a recent strand of the literature investigating

its effects on employment. In this literature, a salient result is that service ex-

ports are an important source of job creation (Martins, 2016; Rueda-Cantuche,

International Labour Review170

Cernat and Sousa, 2019). Moreover, recent contributions document the fact

that lower barriers to service trade in one country generate employment in

that country’s trading partners without necessarily diminishing employment in

its own economy (Shingal and Sauvé, 2019; Kühn and Viegelahn, 2019).

Beyond these impacts, the recent boom in service offshoring induced by

the ICT revolution has also generated – and is expected to further generate –

important redistribution effects. The fall in offshoring costs as a result of the

ICT revolution has been more signicant for some labour tasks than for others,

implying that certain workers have become more strongly exposed to global

competition. In this context, one would expect the effects of service offshoring

to vary depending on the tradability of labour tasks. In other words, the de-

gree to which service outputs are tradable has become the predominant factor

shaping the wage and employment effects of international trade.

Recently, this prominence has led to literature focused on the construc-

tion of a wide range of measures of tradability. However, as explained in the

present review, this literature has been confronted with a major problem: the

concept of service tradability is extremely hard to dene, partly because there

is no agreement on what characteristics of a service job make it easier to off-

shore. In addressing this topic, the literature has proposed three potential de-

terminants of service tradability: Levy and Murnane (2006) argue that trad-

ability increases with the degree of routine in a job; Blinder (2006 and 2009)

claims that tradability is stronger for services that do not require personal

interaction; and Garner (2004) holds that ICT-enabled jobs tend to be “off-

shorable” because their output can be sent over long distances at little cost.1

The importance gained by tradability has also raised the question of

whether traditional labour market distinctions between skill groups must be

complemented with information on the tradability characteristics of labour

tasks. Interestingly, the answer to this question depends on the nature of the

correlation between skill intensity and tradability: if these two labour dimen-

sions were perfectly correlated, the additional information on tradability would

be redundant and there would be no need to complement the labour market

distinctions traditionally used in international trade.

Along these lines, existing work has explored the direction and strength

of the correlation between skill intensity and tradability. Leading scholars have

argued that there is in principle no reason to believe that the correlation is

strong (Krugman, 1996 and 2011; Grossman and Rossi-Hansberg, 2006; Bald-

win, 2006; Blinder, 2006 and 200 9). In this regard, empirical studies show that

the correlation could be positive but far from economically signicant (Blinder,

2009; Jensen and Kletzer, 2010).

The new, predominant role of service tradability has been more thor-

oughly investigated by Crinò (2010) and by Liu and Treer (2011), using so-

1 As will be discussed below, these job characteristics have been used in the construc-

tion of tradability indices. Depending on whether these indices rely on personal judgement or

on the number of characteristics considered, they can be classied according to two dimensions:

(i)subjective or objective and (ii) simple or compound.