Skills, population aging, and the pattern of international trade

• DOI: http://doi.org/10.1111/roie.12386
• Published date: 01 May 2019
• Date: 01 May 2019

Rev Int Econ. 2019;27:499–519. wileyonlinelibrary.com/journal/roie

|

499

© 2019 John Wiley & Sons Ltd

1

|

INTRODUCTION

Population aging is the most prominent global demographic trend of the 21st century, which is ex-

pected to alter the labor force composition in a large number of countries. Along with population

aging comes the structural change in the mix of skills that an average worker of a country possesses,

as long as some of these skills change over a person’s life cycle. Some skills are known to deteriorate

with age, causing a decline in the aggregate supply of those skills in aging populations and altering

age-earnings profiles. To understand better the economic implications of aging, one needs to know the

effect of demographics on international trade flows since changes in the supply of a certain skill in a

country may also affect the demand for that skill through exports and imports.1

In this study, we analyze the effect of population aging on trade flows through a reduction in the

supply of a particular age-dependent skill—the ability of an individual to adapt to changes in working

conditions. The neuroscience and behavioral literatures document that the ability of an individual to

Received: 31 January 2018

|

Revised: 14 September 2018

|

Accepted: 29 October 2018

DOI: 10.1111/roie.12386

ORIGINAL ARTICLE

Skills, population aging, and the pattern of

international trade

Ke Gu

|

Andrey Stoyanov

Department of Economics,York University,

Toronto, Canada

Correspondence

Ke Gu, Department of Economics, York

University, 4700 Keele Street, Toronto, ON,

Canada.

Email: kegu@yorku.ca

Abstract

Some cognitive functions, such as the ability to update

skills and adapt to changes in working conditions, are

known to vary with age. With population aging it becomes

increasingly difficult for firms to find workers with up-to-

date skills. As a result, countries with aging populations

start losing comparative advantage in industries that rely

heavily on those skills. We test this hypothesis and find

robust evidence for a significant negative effect of popula-

tion aging on comparative advantage of a country in in-

dustries that are intensive in skill adaptability of labor

force, in both the cross-sectional and the dynamic panel

data sets.

JEL CLASSIFICATION

F14, F16, J11, J24

500

|

GU and STOYA NOV

adjust to frequent changes in the workplace or in the job requirements is getting worse with age. This

implies that younger workers, all else equal, are more productive in tasks that require workers to up-

date their skills regularly. Throughout the paper, we call the worker ability to adapt to changes in the

workplace the adaptability skill.

The age-induced decline in the adaptability skill implies two channels through which aging labor

force affects the pattern of comparative advantage across countries with different demographic com-

position. First, population aging reduces the effective stock of adaptability skill and increases the

skill premium, and industries that are intensive in the adaptability skill will face higher labor costs.

This Heckscher–Ohlin channel implies that aging countries will lose their comparative advantage in

industries that rely heavily on adaptability skills. Second, aging workers may become less productive

in occupations that require adaptability skills. As a result, industries that rely on such occupations will

lose the Ricardian comparative advantage as the pool of available workers becomes older. Therefore,

regardless of the channel through which demographic changes affect comparative advantage, aging

countries are expected to specialize less in industries in which adaptability skills are important.

In order to analyze the effect of aging on bilateral trade flows through adaptability skills, we con-

struct an industry-level measure of intensity in the adaptability skill using two data sources. First, we

use the O*NET data base, which surveys workers, occupational experts, and occupational analysts

in the United States to measure the importance of various skills for different occupations. O*NET

provides a direct measure of the relative importance of adaptability skill across occupations. Second,

we use occupational composition in four-digit NAICS industries in the United States to construct a

weighted-average measure of the adaptability skill intensity for each industry.

Testing our main hypothesis on a cross-section of bilateral trade data for a large set of countries

in year 2000, we find that countries with relatively older labor forces tend to export less in indus-

tries that are intensive in adaptability skill. Interestingly, the effect of demographic differences across

countries on trade is similar in magnitude to the effect of differences in physical capital and skilled

labor, the two conventional factors of production. Furthermore, the role of cross-industry variation in

adaptability skill in bilateral trade is similar to the effect of other age-dependent skills identified in

previous literature, such as physical and cognitive skills. Our baseline results imply that if an industry

has intensity in adaptability skill one standard deviation higher than the average in the economy, then

a one year increase in a country’s median age will decrease exports in that industry by 1.9% relative

to other industries.

Another theoretical prediction that we test is that different rates of population aging across coun-

tries over time should change the patterns of comparative advantage. Specifically, if a country’s popu-

lation is aging faster than in other countries, then we should expect exports of the former to move away

from industries that rely heavily on the adaptability skill. A test based on the dynamic panel structure

has an additional advantage as it allows differentiating out country-specific factors, such as the level

of development, with a proper set of fixed effects. Using the panel data on 80 exporting countries, 136

importing countries, and 71 industries for the time period from 1962 to 2010, we confirm that export

structure of a country depends on the rate of population aging relative to other countries. The mag-

nitude of the effect estimated from the dynamic model is nearly identical to that estimated from the

static model: a one year increase in the median age of a country and a one standard deviation increase

in industry’s intensity in adaptability skill is associated with a 1.9% reduction in exports. This result

confirms that while population aging decreases the relative supply of the adaptability skill, the effect

of this change on labor markets is alleviated, to some extent, by a decrease in the demand for those

skills through an increase in imports of goods that are intensive in the adaptability skill.

The results of our study complement the research investigating the sources of comparative advan-

tage in international trade. Beck (2003), Manova (2008, 2012) and Campello and Gao (2017) find that