Content uploaded by Chan Guk Huh
Author content
All content in this area was uploaded by Chan Guk Huh on Jan 12, 2018
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=mree20
Download by: [Chungnam National University] Date: 23 June 2017, At: 02:02
Emerging Markets Finance and Trade
ISSN: 1540-496X (Print) 1558-0938 (Online) Journal homepage: http://www.tandfonline.com/loi/mree20
Do Hallyu (Korean Wave) Exports Promote Korea’s
Consumer Goods Exports?
Chan-Guk Huh & Jie Wu
To cite this article: Chan-Guk Huh & Jie Wu (2017) Do Hallyu (Korean Wave) Exports Promote
Korea’s Consumer Goods Exports?, Emerging Markets Finance and Trade, 53:6, 1388-1404, DOI:
10.1080/1540496X.2017.1313161
To link to this article: http://dx.doi.org/10.1080/1540496X.2017.1313161
Published online: 20 Jun 2017.
Submit your article to this journal
View related articles
View Crossmark data
Do Hallyu (Korean Wave) Exports Promote Korea’s
Consumer Goods Exports?
Chan-Guk Huh and Jie Wu
Department of International Trade, Chungnam National University, Yuseong-gu, Daejeon, Korea
ABSTRACT: This study analyzes the link between international trade in tangible consumption goods and
services, and intangible cultural goods using panel data of Korea’s exports to 40 countries of three types of
consumer goods and TV content since the mid-2000s. The growing popularity of Korean TV dramas (the
stand-in for the Korean wave or “Hallyu”phenomenon) has not been confined to the East Asian region with
interest spreading much further afield. This study estimates a one-directional gravity model that uses
various forms of consumer goods exports as well as inbound visitors to Korea as dependent variables and a
set of explanatory variables plus the Korean broadcasting content exports using the Poisson pseudo-
maximum likelihood procedure. From our analysis, we find the Hallyu exports to have a positive effect
on consumer good exports and inbound visitor flows in general, but the strength of the influence seems to
be weakest in the case of durable consumer good exports.
KEY WORDS: broadcasting content, consumer goods, cultural goods, exports, Korean wave
JEL: F10, Z10
Culture is an important non-economic factor that can affect commercial exchanges between countries.
Interest in the relationship between culture and international trade appears to have given rise to two
strands of research. The first strand focuses on the theoretical expositions that examine the explicit
consumer choices that are associated with the international trade of cultural goods and the economic
rationales that are related to various policies such as the regulated importations of Hollywood movies.
For example, Bala and Van Long (2005) observe that “[a] tariff on Hollywood movies can be Pareto
superior to free trade if it makes local movies viable”(ibid, p. 144). Also see Francois and Van
Ypersele (2002) for a similar argument.
The second strand focuses on finding empirical regularities with regard to the role that
cultural factors play as an enabler of international trade in goods and services. For example,
Disdier et al. (2010) focus on the issue of cultural proximity between countries measured by the
bilateral trade flow of tangible cultural goods, as defined by the UNESCO. Subsequently, they
find that trade in cultural goods positively affects trade flows between countries (see also
Felbermayr and Toubal 2010).
This study follows the second strand of research and offers an empirical assessment of the role of a
unique type of cultural phenomena known as Korean Hallyu and the exports of the intangible
attributes of this area to Korea’s trading partners in promoting exports of Korean consumption
goods. Melitz (2008) uses carefully constructed measures of linguistic proximity and confirms the
positive effects of common language on trade flows found by earlier studies by Frankel and Jeffrey
(1997), Frankel and Rose (2002), and Anderson and Van Wincoop (2004).
Since the early 2000s, Korea has experienced a surge in interest in exports of cultural content,
such as TV dramas and popular music, to its East Asian neighbors of Japan and China. The
phenomenon of Korean cultural content gaining popularity in the region has been dubbed the
“Korean Wave,”or “Hallyu.”There have been numerous journalistic and academic narratives related
Address correspondence to Chan-Guk Huh, Department of International Trade, Chungnam National
University, Yuseong-gu, Daejeon 34134 Korea. E-mail: chanhuh@cnu.ac.kr
Emerging Markets Finance & Trade, 53:1388–1404, 2017
Copyright © Taylor & Francis Group, LLC
ISSN: 1540-496X print/1558-0938 online
DOI: https://doi.org/10.1080/1540496X.2017.1313161
to this trend.
1
The widespread showing of several Korean TV dramas in Japan, China, and other
Southeast Asia countries via on- and off-line media as well as the contemporaneous strong export
growth in Korean cosmetic products to the region is viewed as a prima facie evidence of the positive
effects of Hallyu in promoting Korea’s exports. Korea’s total exports (in USD) rose on average
2.05% per year between 2010 and 2015, and exports to Asia making up on average 56.7% of its total
exports. In comparison, Korean exports of cosmetics (HS33) rose 23.5% per year on average, and
83.5%ofthemwenttoAsia.Korea’s exports of broadcasting content rose 15.1%, and Asia’sshare
was 92.7% for the same period, according to the CEIC and the Korea International Trade
Association database.
Hallyu exports can be categorized as a type of information services export, which has in recent
years grown rapidly, particularly in the fields of Internet publishing and service provision, motion
pictures, and TV programming. Exports of motion pictures in particular have been the focus of
studies in the United States because of the dominance of the related US industries over time (Hanson
and Xiang 2009; Marvasti and Canterbery 2005;Siwek2005). Two distinct characteristics separate
Hallyu from the information service trade of motion pictures. First, it is locally produced cultural
content, and thus trade in Hallyu means a one-directional flow of exports from Korea. Second, the
size of Hallyu productions and exports are of a much smaller scale as compared to the motion
picture industry of the United States, which, for example, accounted for more than half of the total
box office revenues for many European countries (Hanson and Xiang 2009;Fig.6.6,p.215).
Despite the relative smaller size, studies carried out by local economists in Korea, as reviewed in the
survey section, have found that Hallyu exports have a significant effect on Korea’s general exports.
However, we admit that the unique nature of Hallyu phenomenon might work to limit to the
applicability of this study.
This study offers an empirical assessment of the role of Hallyu content in promoting the export of
Korean consumption goods. In this instance, we focus on consumer goods instead of the overall
exports motivated by earlier theoretical studies that dealt with the linkage between consumer prefer-
ences and the cultural goods trade. Hallyu is more likely to influence consumers’purchasing decisions
of consumer goods, rather than industrial goods, in importing countries. We motivate the gravity
model used for our estimation analysis with a simple microeconomic exposition of the consumer’s
demand for imports mechanism through which Hallyu could affect demand for Korean consumer
goods by consumers in importing countries.
We use three types of Korean consumer goods exports to 40 countries as well as tourism service
exports, measured by the number of inbound visitors from these countries, as the dependent variables
for our study. In addition to a set of standard explanatory variables, such as size of economies, income,
and prices, we add exports of Korea’s TV content, which is mainly composed of dramas, as the
representative Hallyu variable. Annual data from 2006 to 2014 were used for the estimation analysis.
These countries are chosen because they are important destinations for Korean consumer goods
exports and they include both Hallyu importers and non-importers. Data availability was also an
important selection criterion.
Being able to ascertain whether Hallyu TV content has a significant effect on exports would
provide an important piece of evidence as to the changing nature of cultural proximity variables in
today’s Internet age. That is, with the availability of the Internet everywhere, cultural content such
as TV dramas might have a greater influence on consumer behavior than the days of the American
TV series “Dallas”during the 1980s, when its consumer market penetration was limited by fixed
line distribution channels.
2
The continued expansion of Hallyu content might mean that the
advancement in information technology will weaken the influence of factors such as market size
and economies of scale, which are deemed to be important for the competitiveness of the US film
industry.
This article is organized as follows. Related literature is reviewed in the first section. The next
section offers discussions of the Hallyu phenomena and a sketch of the theoretical justifications for our
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1389
empirical study. The following section describes the details of our estimation analysis and presents the
results. The final section offers relevant policy implications and conclusions.
Literature Survey
There is a large body of literature that examines the various issues of the relationship between culture
and international commercial exchanges. On the one hand lies a theoretical focus, which analyzes the
nature of cultural goods and how they affect, and are affected by international trade. Bala and Van
Long (2005) offer a rationalization of barriers to cultural goods imports that can change the prefer-
ences of the local population using the biological evolution model of preference transmissions across
generations. Olivier, Thoenig, and Olivier (2008) define cultural identity as a consumption externality
and analyze the divergence/convergence cultural diversity outcomes that occur as a result of economic
as well as social integration. Disdier et al. (2010) offer an interesting way to model cultural goods
consumption for importing country consumers by claiming that it is the stock of past cultural goods
consumption that is relevant to consumers’demand for cultural goods imports, as highlighted in the
economic models of addiction.
In addition, many empirical studies examined the influences of culture and other non-traditional
factors on trade. Linguistic proximity has been one key explanatory variable in the gravity equations of
many studies (Boisso and Ferrantino 1997; Hutchinson 2005; Isphording, and Otten 2013; Kónya
2006; Frankel and Jeffrey 1997; Lohman 2011; Melitz 2008). For example, sharing a language could
increase trade flow by around 65% (Disdier and Mayer 2007, p. 1142).
Colonial links and ethnic linkage (through migration) are also common factors used in many
gravity models. Eichengreen and Irwin (1998) and Rose (2004) are often cited. Studies have analyzed
the linkage between the effects of migration and ethnic network, such as those of regionally scattered
Chinese diaspora and trade flows. Articles by Gould (1994), Head and Ries (1998), Rauch and
Trindade (2002), Wagner, Head, and Ries (2002), Girma and Yu (2002), Felbermayr and Toubal
(2010) each focused on particular ethnic groups, countries, or regions and found fairly large and
significant effects. Ethnic networks tend to reduce various transaction costs that would deter trade
between international trading partners without the connections.
From the literature, we take away two points. First, one needs to pay close attention to the basic
economic mechanism through which cultural goods affect the consumer demand behavior of an
importing country. Second, the list of non-traditional factors that were discussed earlier does not
necessarily apply to Korea. That is, the Korean language is quite unique and is only used by
approximately 75 million Koreans on the Korea peninsula. Furthermore, Korea does not have much
of an ethnic diaspora like the Chinese networks with the exception being the small communities
located in Japan and the United States. In addition, with the exception of Taiwan and Japan, Korea
does not share common historical colonial links with any of its trading partners. Consequently, Hallyu
might be the only non-economic factor that could help explain the trade patterns of Korea’s consump-
tion goods exports.
Next, we turn our focus to the studies of the Hallyu and non-Hallyu export relationships in Korea.
In this instance, the studies are all empirical by nature and could be arranged into different categories
depending on the key dimensions of their analysis.
Many studies use UNESCO’s cultural commodity definition, as specified in terms of HS codes, as
the Hallyu exports (Kang 2009; Kim and Ahn 2012; Park and Choe 2009). Others use different types
of Hallyu variables, such as survey-based cultural content exports as compiled by the Korea Creative
Contents Agency (KCCA), a government-supported entity that includes both tangible and intangible
cultural contents exports, Google search and YouTube page views of major K-pop stars (Lee, Kim, and
Woo 2014), and Google trends query indexes of “Korean dramas”(Park 2015), have also been used.
There is also a study (Park 2014) that analyzed the exports of Hallyu contents and TV broadcasting
contents to 11 Asian countries using various explanatory variables.
1390 C.-G. HUH AND J. WU
The dependent variables of the studies mentioned above range from Korea’s overall exports,
consumer goods exports, other two- or four-digit HS commodities, TV broadcasting contents exports,
and FDI (Kang 2009). The scope of export destinations also varies. Most common are some subsets of
East Asian countries, 25 Asia Pacific countries (Park and Choe 2009), 10 Southeast Asian countries
(Kang 2009), 23 countries (10 Asian), chosen by the order in magnitude of Korea’s cultural goods
exports (Kim and Ahn 2012), 11 Asian countries (Park 2014), and finally a 74-country study involving
Korean cosmetic product exports (Park 2015).
Without exception, studies that used Hallyu exports as a key explanatory variable and focus on
Asian countries found it had a significantly positive effect on various types of Korean exports when
gravity models and OLS, Tobit, Poisson pseudo-maximum likelihood (PPML) methods were used.
Studies have mostly used gravity model specifications that consist of contemporaneous variables,
which could cause a simultaneity problem between Hallyu exports and other non-Hallyu exports. As a
consequence, to ensure statistical pre-determinedness some used a lagged Hallyu variable.
We aim to improve upon these previous studies in several ways. To date, many have focused on
regionally concentrated countries already well connected by regional trade networks, and for which
Hallyu trends have already made visible headways. As such, the issues that could be addressed from
the analyses of such a biased and relatively small sample could be quite limited. In order to overcome
this, we employ, as part of our analysis, a sample of 40 countries that are regionally dispersed. Values
of Hallyu imports from Korea also vary widely, starting with zero observations.
The use of Korea’s aggregate exports as the dependent variable in some studies is questionable
because the largest component of Korea’s export base is its capital or industrial intermediate goods in
such areas as electronics, machinery, and chemical industries. Implicitly assuming that purchasing
managers of importing countries make their decisions on non-economic factors such as Hallyu is not
very realistic. On the other hand, viewing a Korean drama, which features a refrigerator as a product
placement advertisement, might affect the purchase decision of a shopper for white goods in a Hallyu
importing country. Hence, we focus on Korea’s consumer goods exports.
Hallyu Contents and Korea’s Exports
Hallyu Content
We use annual survey data compiled by the KCC on TV content exports as the Hallyu variable in our
study. The reason for this is because official trade data such as the UN Comtrade database or national
customs database have detailed records mainly of tangible goods. For example, they measure “motion
pictures in terms of the value of cinematographic film exposed or developed, which is a commodity
rather than a service.”(Hanson and Xiang 2009, p. 205). For TV content, customs database will have
detailed data of recorded CDs and DVDs, but they do not have export records of intangible goods
export such as TV content.
3
Until 2002, Korea had recorded deficits in the trade of its broadcasting contents, with the US being
the dominant exporter of TV content for Korean viewers. In 2002, Korea reversed this trend as its
exports exceeded imports (USD 28.8 million vs. USD 25.1 million), thus recording its first TV
broadcasting trade surplus. As of the end of 2014, Korea’s exports in this category have risen more
than tenfold to USD 313.8 million, while imports have only doubled to approximately USD 59.3
million.
Japan has been by far the largest importer of Korean broadcasting content, composed predomi-
nantly of TV dramas. However, since the early 2000s China and Hong Kong have emerged as major
importers, thereby reducing the relative importance of Japan. The respective shares of Korea’sTV
broadcasting exports in 2014 to Japan, China, and Hong Kong are 25%, 18%, and 17%, respectively.
These numbers help to demonstrate how exports having once started in Japan have now spread to
China and other Asian countries. This development is geographically limited to East Asia. Though
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1391
relatively small in value terms, Korea’s broadcasting contents have reached many countries throughout
the America’s, Europe, and the Middle East regions in recent years.
4
There seems to be two reasons why this type of Hallyu export is gaining much attention.
Broadcasting contents display the strong public good characteristics of non-excludability and non-
rivalry in consumption; thus they could have substantial externalities. Presumably, these characteristics
and the associated effects of popular American culture exports, such as Hollywood movies and TV
dramas, were the key motivating facts behind the theoretical expositions of Bala and Van Long (2005)
and others that found economic justifications for implementing trade barriers in the trade of cultural
goods. The spread of American broadcasting content has been associated with increases in the global
recognition of American brands such as Coca Cola and American automobile majors. In a similar vein,
Hallyu content exports offer an interesting basis from which businesses and policy makers in Korea
can spread and reinforce Korean brands in a similar fashion to that of America’s Hollywood movies.
The Korean tourism industry is viewed as a key beneficiary of Hallyu. There has been a noticeable
surge in foreign visitor traffic, growing on average 10% per year since around 2010, paced by Chinese
tourists. To put it in perspective, tourists from Asia made up approximately 75% of the total inbound
foreign tourists of 5.15 million in 2001 according to the Korea Tourism Organization. The share of
Chinese visitors in the Asian group was 12.5%. In 2016, tourists from Asia made up approximately
85% of the total of 17.24 million visitors. Approximately 55% of the Asian tourists were from China
(8.07 million). Many journalistic narratives interpret this trend as evidence of the influence of Hallyu.
5
Hallyu and Consumer Preference of Importing Countries
Many theoretical expositions of culture and trade, as seen in the previous section, start with the
consumer preference specifications that view cultural goods as a key argument of the consumer’s
utility function which sees cultural and non-cultural goods as being complimentary to each other.
Theoretically, consumption of Korean commodities and Hallyu enter into the utility function of the
representative consumer of importing country as follows:
Ut¼uct;ckt;Cit ;Ht;...ðÞ (1)
Here, c
t
denotes consumption of home consumption goods, c
kt
denotes Korean import consumption,
C
it
is a vector of imported consumption goods from other countries, and H
t
denotes the consumption
of Hallyu contents, all of period t. We assume a well-behaved utility function (i.e., u'(∙)<0,u'(∙)>0
for all augments). At the optimum, which is obtained following an optimization subject to a budget
constraint, we can derive the demand for the import function of the following form:
ckt ¼cpt;Ht;Zt
ðÞ:(2)
Here, p
t
stands for relative price of import, assuming the price of the domestic consumption goods
as numeraire. A vector Z
t
includes all other variables that could affect demand for imports from
Korea, c
kt
. This demand is depicted in Figure 1, in which the vertical and horizontal axes represent
the price of Korean imports and the quantity of imports from Korea, respectively. In this frame-
work, Hallyu acts as a shifter of the demand curve for Korean imports as shown in the figure (H
0
<
H
1
<H
2
). Empirical studies that find the export enhancing effects of Hallyu can be interpreted as
confirming this aspect of derived demand. That is, contemporaneous exports amount to confirming
the positive sign of the cross-derivative of (2) @ck
@H>0
. It is a confirmation of co-dependency.
The discussion so far provides an outline of a model that relates to Korea’s consumer good exports,
equivalently, a demand for Korean consumer goods imports for each of the sample countries to the
factors of p
t
,Z
t
, and H
t
. The estimation model is an empirical couterpart of the demand curve (2). The
actual elements of Z
t
will be disussed in the following section. In this instance, a couple of issues
related to the estimation model specification need to be addressed.
1392 C.-G. HUH AND J. WU
First is the simultaneity of H
t
and c
t
in (2). One could derive an implicit demand for the Hallyu as a
function that contains c
kt
as a determinant. Since both c
kt
and H
t
are normal goods, the following
derivative term @H
@ck
must also be positive. That is, they are simultaneously determined, and thus a
regression equation that includes both c
kt
and H
t
in the same equation is likely to give rise to a positive
bias for the coefficient estimate of the H
t
explanatory variable. Although an increase c
kt
would bring
about a second round positive effect on H
t
, different specifications of Hare going to be used in the
estimation equations that correspond to each assumption to gauge the extent of the bias.
Three different specifications of Hwill be considered in the next section (Case I). The first case
uses the contemporaneous H
t
as the Hallyu variable. The other two specifications are designed to
secure the statistical pre-determinedness of Hwith respect to c
kt
. The first is to use H
t–1
in place of H
t
.
This is also a simple way to model the habit formation aspect of the influences of cultural goods
consumption. The second way to accommodate the habit formation aspect is to adopt the formulation
of Disdier et al. (2010),
6
which in turn appealed to the Chaloupka (1991) model of rational addiction.
In this setup, it is the accumulated stock of past exposure to Hallyu that influences consumer’s
purchasing behavior toward Korean imports. Specifically, we use a transformed Hallyu variable that
captures both the dynamic and habit forming nature of cultural consumption. As such, a new variable
SUMH is defined as follows:
SUMHt1¼X
t1
i¼0
1δðÞ
ti1Hti1:(3)
Here, δand idetermine the rate of decay of the influence of past consumptions as well as the length of
lag.
Using a dynamic panel model specification is a solution if one suspects a large influence of past
consumption on the current consumption because of the persistency in consumer behavior. However, if
imports of Korean consumer goods and Hallyu are contemporaneously positively related, then
including lagged Hallyu terms as discussed above in estimation of (2) might have a similar effect to
that of using a dynamic panel model specification that includes a lagged dependent variable, which
reduces usable data.
The second part of the analysis compares the two cases of H
t
= 0 and H
t
> 0 (Case II). Intuitively,
this could be explained as examining the consequences of the shift of c
k
(H
0
)toc
k
(H
1
)as shown in
Figure 1, where H
0
denotes the demand for Korean imports when H
t
= 0, while H
1
denotes a nonzero
level of Hallyu consumption.
Figure 1. Demand for Korean imports curve.
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1393
The third issue we examine empirically is whether the demand curve for Korean imports shifts
monotonically to the right in Figure 1 as Hincreases (Case III). The results from this examination
would shed light on how Hallyu affects the importing countries’demand for Korean exports. One
possibility is that the presence of Hallyu in a particular country only acts as a minimum hurdle of
familiarity with Korean products. Monotonic increases in Hallyu exports might not lead to commen-
surate increases in exports of other goods if this were the case.
The fourth case relates to the regional concentration of Korea’s exports and inbound tourists, as
well as Hallyu exports (Case IV). The most common narratives about Hallyu effects reflect the
regional concentration of Hallyu exports to East Asian countries. If Hallyu and Korea consumer
goods are complementary as we posited here, then we should observe a positively correlated pattern
between the two across different regions. That is, the regional distributions of the two sets of exports
should look alike.
Empirical Analysis
Data
This study assesses the effects of Korea’s cultural content exports on other types of exports to Korea’s
top 40 trading partners, based on consumer goods export values (Table 1). The dependent variables,
which represent c
kt
of the demand for Korean imports in Equation (2), are included in our gravity
model which consists of three types of tangible goods exports plus the number of visitors from the
same group of 40 trading partners. The list of commodity exports of Korea to these partners and their
mnemonics are, for the purpose of this study, called Nondurable, Semidurable, and Durable, respec-
tively. The nondurable consumer goods category refers mainly to foodstuffs. The second type of
consumer goods consists of clothing, footwear, and home furnishing items. While passenger auto-
mobiles, home electronics are examples of durable consumer goods and they make up the majority of
Korea’s total consumer goods exports. Table 2 shows details of the significance of each type as well as
the top export destinations for Korean exports in the three categories of consumer goods. The last
dependent variable is the number of visitors from these countries to Korea (Vstr).
The demand function for Korean consumer good exports (2) is a simple version of the gravity
models of trade flows. A key difference is that this pertains to a one-directional trade flow, i.e., from
Korea to 40 countries. Therefore, the vector Zof (2) will contain a number of variables that affect the
demand for Korean consumer goods, such as income, the prices of consumer goods from countries
other than Korea, and so forth. The common variable list consists of the following three sets of
Table 1. List of 40 countries grouped into five groups of country and regional dummy variables.
Region Country
Japan Japan
China China (mainland), Hong Kong*
Other Asian region
(7 countries)
Indonesia, India, Malaysia, Philippines, Singapore, Thailand, Taiwan*
Europe
(16 countries)
Austria, Belgium, England, France, Poland, Germany, Hungary, Italy,
Netherlands, Russia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey
North America
(3 countries)
America, Canada, Mexico
Other regions
(11 countries)
Argentina, Australia, Brazil, Columbia, Chile, Israel, New Zealand, Peru, Saudi Arabia,
South Africa, United Arab Emirates
* Hong Kong and Taiwan are treated as separate countries for the purposes of analysis.
1394 C.-G. HUH AND J. WU
Table 2. Top five countries for different types of consumer goods exports of Korea in 2014
(million USD, %).
Total Nondurable Durable Semi-durable
Total: 88,165 Total: 6,514 Total: 73,956 Total: 7,695
Country Share Country Share Country Share Country Share
USA 0.25 Japan 0.29 USA 0.28 China 0.20
China 0.09 China 0.14 China 0.07 Japan 0.13
Japan 0.06 USA 0.10 Saudi Arabia 0.04 USA 0.11
Russia 0.04 Hong Kong 0.05 Russia 0.04 Vietnam 0.09
Saudi Arabia 0.04 UAE 0.05 Canada 0.03 Hong Kong 0.08
Table 3. Data Description (2006–2014).
Variable
name
a
Unit Data source Obs. Mean
Standard
dev.
Dependent variables that represent c
kt
(Y): Korea’s exports and others
Nondurable Korea’s export of
nondurable consumer
goods, USD thousand
Korea International Trade Association
(KITA)
360 109,095 284,694
Durable Korea’s export of durable
consumer goods, USD
thousand
Korea International Trade Association
(KITA)
360 1,383,371 2,545,460
Semidurable Korea’s export of semi-
durable consumer goods,
USD thousand
Korea International Trade Association
(KITA)
360 114,122 240,126
Vstr Persons Korea National Tourism Organization
(KTO)
360 211,163 628,710
Independent variables
SIZE: variable measuring the size of 40 trading partner countries
Pop Population (thousand) 360 117,719 276,395
X: Other explanatory variables
Dist Distance from Korea
(kilometer)
www.timeanddate.com 360 8,405 4,366
Rexch Real exchange rate, KRWs
per country i’s currency
Author’s calculations using nominal
exchange rate and Consumer Price
Indexes from the International Monetary
Fund
360 633 610
REER BIS real effective exchange
rate of country i
Bank for International Settlements (BIS) 360 98 7.0
Linder Absolute value of difference
of log real GDP per capita
of Korea and country i
b
Author’s calculations using real GDP per
capita from the World Bank
360 0.9 0.7
H: Hallyu export variable
BD Korea’s broadcasting
content export, USD
thousand
Korea Creative Content Agency (KCCA) 146 8,341 21,772
a
Country subscript iis omitted.
b
Real GDP per capita was calculated using 2005 CPI.
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1395
variables. First, SIZE represents variables that measure the size of the importing country. We
considered gross domestic product (GDP), and population (Pop), and chose to use Pop.
7
Second, X
represents explanatory variables that usually appear in the gravity models. They are as follows: the
distance between the capital of the importing countries and Seoul, Korea (Dist); real exchange rates
vis-à-vis Korean won (Rexch); the Linder effect variable (absolute values of the natural log difference
of the GDP per capita of the importing country and Korea (Linder); and the real effective exchange
rate of the importing countries (Reer).
The bilateral real exchange rate acts as the stand-in for the price of Korean imports in the
importing country, i.e., ptof (2). Reer, on the other hand, is a stand in for the prices of imports
from other countries that are substitutes for each type of Korean consumer goods. Ideally, an
import price index of country iexcluding the prices of Korean imports should be used. However,
this import price index for substitutes cannot be calculated due to a lack of data. (Out of 40
countries in our sample, aggregate import price index was available only in 20 countries.) There
are two sources of a real exchange rate appreciation of country i. One is the nominal exchange rate
appreciation, holding the prices of the home and trading partner countries constant. Alternatively,
Reer changes because the relative prices of the home and trading partner countries change, holding
the nominal exchange rate constant. Suppose Reer
i
appreciates because of fall in trading partners’
prices. This would also mean a decrease in the price of imports coming into the country i.Ifwe
further assume that the share of Korean goods in the basket of total imports of country iis small,
then a decrease in her import prices would mean a lowering of import prices from countries other
than Korea. This, in turn, would lead to a decrease in the price of substitutes of Korean imports for
consumers in country i.Inthissense,Reers can be viewed as import prices of goods that compete
with imports from Korea.
The Hallyu variable is the export of broadcasting content. The value of broadcasting content
imports from Korea for country iis referred to as H
i
. Out of the 40 countries, 25 countries had
nonzero observations in total. We use different specifications of Hin our estimation as discussed in the
previous section.
The baseline gravity estimation model is as follows:
lnY h
ki;t¼αhþβh
1lnSIZEi;tþβh
2lnXki þβh
3lnHi;tþγhYeartþεh
ki;t:(4)
Here, Year is a year dummy to capture year fixed effects. We use the PPML estimation method of Silva
and Tenreyro (2006,2011) to account for serious heteroskedasticity problems as well as the zero value
for dependent variables present in data. The mixture of variables in levels and log levels are due to the
PPML methodology.
Next, the contemporaneous cross-correlation patterns are examined, as shown in Table 4, among
the explanatory variables to select estimation models to minimize the potential multicollinearity
problem. BD and Dist show the highest level of correlation (−0.63) across all pairs considered. This
suggests a significant degree of regional concentration of Hallyu exports measured in terms of
Table 4. Contemporaneous correlations of explanatory variables.
Pop Dist Linder BD Rexch REER
Pop 1
Dist −0.18*** 1
Linder 0.57*** −0.25*** 1
BD 0.27*** −0.63*** 0.19* 1
Rexch −0.17* 0.45*** −0.54*** −0.25** 1
REER −0.13 0.06 −0.11 0.05 0.27*** 1
1396 C.-G. HUH AND J. WU
broadcast content. Another pair of interest are the two real exchange rates of Rexch and Reer. Their
correlation of 0.27 suggests that each of the two exchange rates are likely to convey distinct
information.
Estimation Results
The estimation results are presented in Tables 5 through 9. They are organized into groups representing
different cases of the “Hallyu and consumer preference of importing countries”section. For each
estimation model, two results from two alternative specifications are reported. This is because a highly
negative correlation between the bilateral real exchange rate and Linder variable, which is evident in
Table 4, appears to be affecting the estimates of the real bilateral exchange rate coefficients.
(1) Case I: Different specifications of H
t
(BD
t
, SUMBD
t–1
)
The estimation results of (4) in which BD
t
, SUMBD
t–1
were used for H
t
, as shown in Tables 5 and
6, respectively. The results for the alternative specification that uses BD
t–1
are very similar to those of
Table 6, and are therefore not shown here separately. Coefficient estimates are strikingly close to each
other, suggesting that the current TV content exports are a good instrument variable for the cumulated
sum. Thus, we will use the results of Table 5 as the benchmark in the subsequent discussions. The
fixed year effect dummy variable has been used in all specifications as shown. The results show the
models to have substantial explanatory power as the R
2
values range from 0.82 to 0.95.
The SIZE variable, population which measures the size of the economy, is significant and positive.
However, Dist, the physical distance of the importing country from Korea, appears with a positive sign
and is highly significant in the model of durable goods exports (Durable). A literal interpretation is
that the value of Korean durable consumer good exports to her trading partners are directly propor-
tional to their distance from Korea. This might be due to the fact that the main market for Korean
consumer durable goods exports tend to be located in the high income countries of North America and
Europe, as can be seen in Table 2. This explanation will be tested in the latter part of this section by
testing estimating models that utilize regional dummy variables. Otherwise, it appears to be statisti-
cally significant with a negative sign as is the case in typical gravity models.
Table 5. Estimation results of equation (4), using BD
t
as Hallyu variable H(2006–2014).
Nondurable Durable Semidurable Vstr
Pop 0.19*** 0.54*** 0.60*** 0.89*** 0.37*** 0.53*** 0.24*** 0.47***
(0.04) (0.07) (0.05) (0.07) (0.05) (0.09) (0.03) (0.05)
Dist −0.41*** −0.37*** 0.60*** 0.53*** −0.24** −0.25*** −0.72*** −0.69***
(0.10) (0.07) (0.09) (0.07) (0.10) (0.09) (0.08) (0.06)
BD(t) 0.33*** 0.20*** 0.20*** 0.13*** 0.26*** 0.21*** 0.32*** 0.22***
(0.03) (0.03) (0.02) (0.02) (0.02) (0.03) (0.02)
Linder −0.85*** −0.83*** −0.39*** −0.52***
(0.11) (0.10) (0.12) (0.08)
Rexch −0.04 −0.14*** 0.17*** 0.01 0.06*** 0.01 −0.04 −0.10***
(0.04) (0.03) (0.04) (0.04) (0.02) (0.03) (0.03) (0.02)
Reer −2.02*** −0.58 −0.27 0.52 −0.67 −0.02 0.62 1.35***
(0.48) (0.43) (0.86) (0.66) (0.60) (0.64) (0.41) (0.37)
Constant 20.25*** 11.68*** 1.59 −2.78 10.83*** 7.17** 10.92*** 6.19***
(2.16) (2.17) (4.15) (3.23) (3.19) (3.66) (1.82) (1.81)
Obs # 146 146 146 146 146 146 146 146
R
2
0.81 0.91 0.81 0.88 0.82 0.85 0.92 0.95
Standard errors in parentheses, * p< 0.10, ** p< 0.05, *** p< 0.01.
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1397
The sign patterns of the coefficient estimates of bilateral as well as multilateral real exchange rates
are somewhat mixed in the full sample that includes all variables. However, they show more consistent
patterns in the model when the Linder variable is not included. Coefficients for the bilateral real
exchange rates are positive and significant implying that a depreciation of Korean Won in real terms
would have a positive effect on both the volume of Korea’s consumption goods exports to her trading
partners. Although not significant in most cases, the coefficient estimates of Reer
i
are negative for the
three consumer goods export models. This effect is also most noticeable in the case of durable goods
exports. This is supportive, albeit weakly, of our proposed interpretation of the real effective exchange
Table 6. Estimation results of Equation (4), using SUMBD
t–1
as Hallyu variable H(2005–
2013, calculating from 2001).
Nondurable Durable Semidurable Vstr
Pop 0.19*** 0.51*** 0.57*** 0.83*** 0.38*** 0.52*** 0.26*** 0.44***
(0.04) (0.06) (0.04) (0.06) (0.04) (0.08) (0.03) (0.04)
Dist −0.42*** −0.41*** 0.58*** 0.51*** −0.21** −0.23*** −0.65*** −0.64***
(0.09) (0.07) (0.08) (0.06) (0.10) (0.09) (0.07) (0.06)
SUMBD(t –1) 0.34*** 0.20*** 0.21*** 0.14*** 0.27*** 0.22*** 0.37*** 0.28***
(0.04) (0.03) (0.02) (0.02) (0.02) (0.02) (0.03) (0.03)
Linder −0.80*** −0.79*** −0.36*** −0.41***
(0.10) (0.09) (0.10) (0.07)
Rexch −0.04 −0.14*** 0.15*** 0.01 0.05** −0.01 −0.04* −0.09***
(0.03) (0.03) (0.03) (0.03) (0.02) (0.03) (0.02) (0.02)
Reer −1.12** −0.09 0.99 1.47*** 0.36 0.81 1.57*** 1.93***
(0.51) (0.46) (0.64) (0.51) (0.64) (0.62) (0.44) (0.41)
Constant 15.53*** 9.51*** −3.68 −6.34** 5.54 3.04 4.59* 2.40
(2.73) (2.38) (3.07) (2.49) (3.59) (3.53) (2.35) (2.08)
Obs # 186 186 186 186 186 186 186 186
R
2
0.81 0.90 0.83 0.88 0.82 0.84 0.92 0.95
Standard errors in parentheses, * p< 0.10, ** p< 0.05, *** p< 0.01.
Table 7. Estimation results of Equation (4), using observations without Hallyu exports (2006–
2014).
Nondurable Durable Semidurable Vstr
Pop −0.19* 0.12 0.29*** 0.77*** 0.16*** 0.51*** 0.28*** 0.21**
(0.10) (0.14) (0.04) (0.04) (0.05) (0.07) (0.06) (0.09)
Dist −1.51*** −1.42*** 0.14 0.02 −1.15*** −1.24*** −2.02*** −2.01***
(0.28) (0.33) (0.12) (0.11) (0.12) (0.22) (0.19) (0.18)
Linder −0.84*** −1.41*** −0.88*** 0.16
(0.26) (0.10) (0.17) (0.15)
Rexch −0.04 −0.11** 0.05*** −0.11*** 0.06 −0.03 0.19** 0.21**
(0.04) (0.05) (0.02) (0.02) (0.05) (0.05) (0.08) (0.09)
Reer −1.12 −1.64 2.19** 1.39** −0.93 −1.43 0.48 0.63
(1.95) (1.77) (0.90) (0.63) (1.08) (0.94) (1.38) (1.42)
Constant 31.24*** 30.77*** −0.80 0.97 23.13*** 23.85*** 22.29*** 21.97***
(10.95) (9.76) (4.54) (3.11) (5.35) (4.63) (6.86) (6.93)
Obs # 214 214 214 214 214 214 214 214
R
2
0.11 0.10 0.21 0.62 0.22 0.27 0.56 0.58
Standard errors in parentheses, * p< 0.10, ** p< 0.05, *** p< 0.01.
1398 C.-G. HUH AND J. WU
rate as a stand-in for an index of prices of country iof the substitute goods imported from countries
other than Korea. However, the opposite sign patterns are seen in the case of inbound tourists. This
implies that visitor traffic is more influenced by the overall competitiveness of their origination
countries, measured by Reer
i
, than the individual bilateral exchange rate vis-à-vis Korea.
The coefficients of the Linder variable and the bilateral real exchange rates are also significant with
the expected signs in all three categories of consumer goods exports. A negative sign on the Linder
variable, which measures the absolute size of income difference for two countries, indicates that the
larger the size of the per capita income difference between Korea and a trading partner, the smaller
Korea’s exports will be to that country.
All in all, the model estimation results suggest that the performance of Korea’s consumer good
exports to a trading partner is robustly and positively affected by Hallyu exports to the country.
(2) Case II: Model without H
t
We repeat the estimations for models that (i) use data for cases where BD
t
equals zero, i.e., no
Hallyu exports. For estimations of the models shown in Table 7, an unbalanced panel data of 214
observations were used.
First of all, the models that do not include the Hallyu variable appear to have significantly smaller
R
2
values when compared with the ones that include Hallyu variables as seen in the previous section
(0.1–0.6 vs. 0.8–0.9). The constant terms (α) are also considerably bigger than those of models with a
nonzero Hallyu, with the exception of the durable consumer good export model.
The coefficient estimates of Dist show divergent patterns. The estimates are considerably more
negative than those of the nonzero Hallyu. Even in the case of durable goods exports (Durable), the
Dist coefficient estimates are substantially smaller in Table 7 than in Table 5 (0.14 and 0.02 vs. 0.6,
Table 8. Estimation results of Equation (4): Full panel with discretized broadcasting export
values.
Nondurable Durable Semidurable Vstr
Pop 0.14*** 0.55*** 0.47*** 0.84*** 0.34*** 0.62*** 0.24*** 0.46***
(0.04) (0.07) (0.04) (0.05) (0.04) (0.08) (0.03) (0.07)
Dist −0.51*** −0.51*** 0.56*** 0.43*** −0.33*** −0.37*** −0.79*** −0.78***
(0.10) (0.07) (0.08) (0.06) (0.09) (0.09) (0.07) (0.06)
Linder −1.02*** −1.10*** −0.70*** −0.51***
(0.13) (0.09) (0.15) (0.14)
Rexch −0.01 −0.12*** 0.13*** −0.02 0.08*** −0.01 −0.02 −0.08***
(0.03) (0.03) (0.02) (0.02) (0.02) (0.03) (0.02) (0.02)
Reer −2.22*** −0.62 1.04 1.58*** −0.77 0.37 0.10 0.94**
(0.49) (0.45) (0.79) (0.55) (0.65) (0.67) (0.46) (0.45)
BD1 0.74*** 0.32* 0.39*** −0.03 0.67*** 0.36** 0.80*** 0.57***
(0.19) (0.19) (0.13) (0.12) (0.21) (0.17) (0.14) (0.13)
BD2 1.26*** 1.27*** 0.90*** 0.66*** 1.43*** 1.37*** 1.39*** 1.39***
(0.23) (0.18) (0.23) (0.13) (0.17) (0.14) (0.17) (0.16)
BD3 2.07*** 1.32*** 1.68*** 0.93*** 1.84*** 1.32*** 2.04*** 1.64***
(0.26) (0.23) (0.20) (0.15) (0.17) (0.20) (0.19) (0.16)
BD4 2.65*** 1.45*** 1.70*** 0.67*** 2.14*** 1.30*** 2.59*** 1.94***
(0.23) (0.25) (0.15) (0.19) (0.17) (0.28) (0.17) (0.18)
Constant 23.33*** 13.59*** −1.99 −5.30* 12.97*** 6.45* 14.60*** 9.26***
(2.09) (2.31) (3.90) (2.80) (3.37) (3.74) (2.00) (2.47)
Obs # 360 360 360 360 360 360 360 360
R
2
0.80 0.89 0.74 0.86 0.81 0.85 0.92 0.94
Standard errors in parentheses, * p< 0.10, ** p< 0.05, *** p< 0.01.
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1399
0.53). This suggests that the greater the distance a trading partner is from Korea the more negative
effect it would have on Korea’s exports to her trading partners which do not import Hallyu. This
indirectly confirms the positive role of Hallyu for both Korea’s consumer goods export and inbound
visitors.
In the case of the Linder variable, it is not significant by itself in the Vstr model. Only when it
appears with Hallyu, does it becomes significant. The coefficient estimates of Rexch show a mixed
pattern, similar to those of Case I. The coefficient estimate of the real exchange rate in the Vstr model
is positive and highly significant. This suggests that inbound visitor traffic becomes more elastic with
respect to real exchange rate in the absence of Hallyu.
One general takeaway from the comparison of the results of Cases I and II is that the inclusion of
the Hallyu variable seems to considerably enhance the ability of the models to explain Korea’s
consumer goods exports and inbound tourist flows.
(3) Case III: Models with discrete BD size dummy variables
The next issue is to examine whether increases in Hallyu export values are accompanied by
commensurate increases in Korea’s exports as discussed in the previous chapter. To carry out this
exercise, we convert the continuous BD export values into 5 discrete categories and assign four
dummy variables. The unit of measurement is one thousand USD. The number of observations in each
category are 214 (for BD = 0), 75 (for BD1), 38 (BD2), 18 (BD3), and 15 (BD4). One explanation of
this approach is that the value of BD exports is proportional to the degree of maturity of the Hallyu
trend, meaning how widely known Korean cultural products are in the importing country.
Table 9. Estimation results of Equation (4), using Hallyu variable BDt: Full panel with country
and regional dummy variables.
Nondurable Durable Semidurable Vstr
Pop 0.20*** 0.36*** 0.35*** 0.53*** 0.28*** 0.43*** 0.34*** 0.52***
(0.02) (0.04) (0.04) (0.06) (0.04) (0.06) (0.03) (0.04)
Linder −0.45*** −0.55*** −0.43*** −0.48***
(0.08) (0.12) (0.09) (0.09)
Rexch −0.04 −0.09** −0.05* −0.12*** −0.02 −0.06*** 0.06** 0.02
(0.04) (0.04) (0.03) (0.03) (0.02) (0.02) (0.03) (0.03)
Reer 0.86*** 0.88*** 1.04** 1.07** 0.70 0.65 2.61*** 2.60***
(0.30) (0.28) (0.50) (0.48) (0.57) (0.53) (0.25) (0.21)
BD 0.11*** 0.09*** 0.08*** 0.05*** 0.10*** 0.08** 0.12*** 0.09***
(0.03) (0.02) (0.02) (0.02) (0.04) (0.03) (0.03) (0.02)
Japan 2.24*** 1.93*** −0.86*** −1.30*** 2.00*** 1.70*** 4.66*** 4.36***
(0.36) (0.35) (0.27) (0.29) (0.38) (0.37) (0.28) (0.25)
China 0.96*** 1.05*** −0.63** −0.52** 1.87*** 1.96*** 3.32*** 3.41***
(0.31) (0.31) (0.25) (0.22) (0.32) (0.31) (0.21) (0.20)
Oasia −0.08 0.17 −1.88*** −1.61*** 0.60* 0.85*** 2.62*** 2.92***
(0.33) (0.33) (0.26) (0.25) (0.32) (0.31) (0.23) (0.23)
Namerica 0.98*** 0.82*** 1.03*** 0.84*** 1.85*** 1.69*** 2.46*** 2.28***
(0.33) (0.32) (0.23) (0.23) (0.35) (0.34) (0.22) (0.21)
Europe −0.39 −0.52 −0.41** −0.57*** 0.09 −0.03 1.28*** 1.14***
(0.33) (0.33) (0.19) (0.19) (0.29) (0.29) (0.21) (0.21)
Constant 4.88*** 3.85*** 5.87** 4.82** 4.19 3.56 −7.17*** −8.21***
(1.32) (1.26) (2.28) (2.21) (2.63) (2.51) (1.12) (0.97)
Obs# 146 146 146 146 146 146 146 146
R
2
0.97 0.97 0.94 0.95 0.93 0.93 0.98 0.98
Standard errors in parentheses, * p< 0.10, ** p< 0.05, *** p< 0.01.
1400 C.-G. HUH AND J. WU
BD1 = 1 if 1 ≤BD
t
≤1,000, and BD1 = 0 otherwise;
BD2 = 1 if 1,000 ≤BD
t
< 5,000, and BD2 = 0 otherwise;
BD3 = 1 if 5,000 ≤BD
t
< 20,000, and BD3 = 0 otherwise;
BD4 = 1 if 20,000 ≤BD
t
, and BD4 = 0 otherwise.
The estimation results from the balanced panel of 360 observations are shown in Table 8. The
dummy variable coefficients are to be interpreted in relation to the group not included in the model,
namely, BD
t
= 0. The coefficient estimates for Pop, Dist, Linder, and Rexch and R
2
of Table 8 are very
similar to those of Table 5, although the Reer coefficients of durable consumer goods exports are more
positive in the discretized models.
Next we turn to the estimation results of the BD dummy variables and present a couple of
noteworthy points. First, the coefficient estimates of the four dummies are different with some
regularities. In all eight cases, the magnitudes of the coefficients of BD2–BD4 are larger than those
of BD1. In models without the Linder variable, the magnitude of those coefficients increases mono-
tonically as the BD export value increases, while the increments in the coefficients stagnate after BD2
in the full models.
Second, the coefficient estimates of the BD dummy variables of Durable models are substantially
smaller than those of the other two consumer goods exports. The size of the Durable BD dummy
coefficients are about one half of those of Nondurable and Semidurable. This suggests that the exports
of this category of consumer goods are less affected by Hallyu exports than the other two.
Third, the BD dummy coefficient estimates of the Vstr models clearly show a monotonically
increasing pattern. This is true in both the models with and without the Linder variable. Thus, it
suggests that inbound visitors are generally more influenced by Hallyu.
(4) Case IV: Models with regional dummy variables
It has been widely noted that Korean TV dramas have rapidly gained popularity in East Asia. In
order to address this change, we added country and regional dummy variables in (4) to gauge the
regional concentration effects. There are two country dummies for China and Japan, as well as three
regional dummies of Oasia (other Asia), Namerica (North America), and Europe. The constituent
countries for each group can be seen in Table 1. A group of countries was left out as a control dummy
group, and thus, the coefficients of the dummy variables are to be interpreted in relation to the other
regional cases. The estimation results from the 360 balanced panel data are shown in Table 9.
The first remarkable result is that the coefficient estimate of the regional dummies, especially in the
Durable equations, is significantly negative. This seems to resolve the puzzling observations of the
distance variable appearing with a positive coefficient estimate for the Durable equations stated in
earlier models as discussed earlier in Case I. The Durable columns of Table 9 show that coefficient
estimates of the regional dummy variables all appear with a negative sign with the exception of the
Namerica. This implies that North American countries are the only market in which Korea exports
more consumer goods than the group of countries that were left out as a control group such as
Australia, Saudi Arabia, and Brazil. It is significantly negative for Japan as well as China, which has
been the number one destination, in terms of value, for Korea’s overall exports as well as broadcasting
content in recent years. This means that both China and Japan have imported much less durable
consumer goods from Korea than trading partners situated farther away.
However, the dummy coefficients turn positive in most cases except for Europe in the nondurable
and semi-durable consumer goods exports categories. This suggests that Korea has exported a lot more
of these two types of consumer goods to its regional and North American trading partners. In the case
of the former (nondurable consumer goods), Japan has been the most important destination followed
by China and North America. China overtakes Japan in the latter (semi-durable) category. It is
interesting to note that North America appears to have imports more goods from these categories
than other East Asian countries.
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1401
The coefficients of regional dummies are all positive in the case of the inbound visitor model,
suggesting that there are more Korea-bound tourists from all of the listed countries and regions than
from the group of countries that were left out. The two groups of exchange rates exhibited show
distinct patterns; the Rexch coefficients are all negative and small in absolute values, while those for
Reer are all positive in the three consumer goods exports. They suggest that once the regional
divergence in consumer goods exports are allowed, bilateral exchange rates do not help in explaining
the variations in Korea’s exports of consumer goods. On the other hand, the real purchasing power of
each importing country represented by Reer does matter in explaining the variations. The two
exchange rates appear with positive coefficients in the case of inbound visitors.
Lastly, the BD coefficient estimates are all significant and positive. However, the size of these
estimates are about half when compared with those of Table 5. This observation suggests that earlier
estimates of the Hallyu effects might be overstated due to factors not considered in the model
specifications.
Summary and Conclusion
This study used a gravity model to examine Korea’s consumer goods exports as well as inbound
visitors from 40 trading partners since 2006 for the existence of any systematic patterns between the
dependent variables and Hallyu TV content and other dependent variables. We find that Hallyu exports
have had a measurable positive effect on consumer goods exports. Evidence based on the estimation
results consists of two parts; the Hallyu appears with significantly positive coefficients, and the
inclusion of Hallyu in the estimation equations tends to reduce the negative effects of the distance
variable, which measures the physical distance between Korea and each trading partner. This second
effect constitutes an example of the cultural proximity helping to mitigate the negative effects of
physical distance on trade flow from Korea to her 40 trading partners. This article’s contribution to the
body of literature that deals with the relationships between international trade and culture is to find a
popular cultural content, whose reach has been greatly enhanced by the spread of Internet technology,
could also be a part of cultural proximity factors such as language and ethnic linkages that affect
international trade.
Once the basic premise that Hallyu is nontrivially present in the derived demand for Korean imports
of a typical importing country is established, we further examine several characteristics of the link
between consumer goods and Hallyu content (BD) exports. First, we estimated the models sequentially
using alternative measures of BD, lagged, as well as an accumulated past BD that captures both the
dynamic and habit forming nature of cultural consumption, to avoid a potential simultaneity problem
between the dependent variable and the contemporaneous BD. We failed to find any perceptible
differences in the estimation results.
Second, this study examined whether an increase in Hallyu exports is associated with a propor-
tionally larger increase in the dependent variables. We compared the coefficient estimates for the
discretized BD dummy variables, which were defined into four groups by ascending order, according
to the magnitudes of BD exports. A discrete jump was seen in the coefficient estimate sizes from the
initial level for the smallest BD export value to the next level and up. However, the magnitudes of
coefficient estimates did not grow proportionately with the increments of BD export values. The
estimates from the durable consumer goods export equation were the smallest of all models.
Third, Korea’s consumer good exports showed a distinctly biased regional distribution: well below
the average of reference countries in East Asia including China and Japan, and well above the average
for North American countries. This is the case that despite the fact that its East Asian neighbors are the
dominant importers of Hallyu exports. This appears to weaken the strength, or economic significance
of the positive effects of Hallyu, seen in the earlier estimation results of this article, as well as findings
of many earlier studies on Hallyu effects. This observation is based on export patterns of durable
consumer goods, which make up close to 80% of the total amount of consumer good exports.
1402 C.-G. HUH AND J. WU
All in all, this study confirms that Hallyu has a measureable positive effect on demand for both
Korea’s consumer good exports and inbound visitor numbers. In regard to the policy implications of
our findings, we note that there is a need to boost Korea’s consumer good exports to its trading
partners in the East Asia region, where consumers are more attuned to things “Korean”as a result of
their increased exposure to Hallyu content. This applies to all types of consumer goods although the
economic benefits would be greatest for durable consumer goods.
Even without demonstrating a large positive externality on Korea’s external commerce, Hallyu’s
role in widening Korea’s cultural understanding across the world is significant enough to be worthy of
Korean government support. To some observers, cultural capital is viewed as no less important as
human and natural capital (Throsby 1999). Our findings of the export enhancing effects of Hallyu
exports appear to be proportional to their values undergirds this recommendation.
Notes
1. For examples of journalistic narratives, see “Riding the ‘Korean Wave’,”Wall Street Journal,May19,
2009 (http://www.wsj.com/articles/SB124267698913031617), “Hallyu, yeah!”Economist, January 25, 2010
(http://www.economist.com/node/15385735), For an example of analysis of Hallyu developments from inter-
national cultural/communication perspective, see Shim (2006). The surge in Hallyu has provoked in some
countries critical examinations of it from a “cultural imperialism”perspective that has been typically associated
with American popular culture. See examples of discussions along this line in Huang (2009)andYang(2012).
2. In 2014, dramas made up approximately two thirds of total TV content exports of Korea in value terms
according to the Korean Communication Commission (KCC), government agency established to oversee Hallyu
content promotion. In terms of types of rights, in addition to the conventional DVD and various airing rights, the
share of sales of digital rights to TV programs have approached close to 20% in recent years.
3. The largest market for this category of Korea’s export items for the past decade and a half was the United
States, far surpassing both Japan and China. This is incongruent to the generally accepted facts regarding the
geographical distribution of Hallyu. Furthermore, the single most important US-bound export item was HS
970110 (paintings, drawings, and pastels, executed entirely by hand). This includes the hand-drawn drawings
that Korean subcontractors have provided to the animation TV and movie contents producers such as Dreamworks
from the United States. It is inappropriate to treat exporting drawings for the “Simpsons,”the popular US TV
animation shows as a form of Hallyu export.
4. For a succinct summary, see “South Korean soap operas hook foreign audiences,”Financial Times,November
13, 2013 (http://www.ft.com/intl/cms/s/0/948719ae-4858-11e3-a3ef-00144feabdc0.html#axzz4BH2fi9IO).
5. “China awash with Korean Wave fever,”Financial Times, April 12, 2016 (https://www.ft.com/content/167338ec-
fa0b-11e5-8e04-8600cef2ca75), “Korea attracts ‘quality visitors’from ASEAN, Korea Times, November 6, 2016
(http://www.koreatimes.co.kr/www/news/biz/2016/11/123_217651.html).
6. See detailed discussion given in the Appendix of Disdier et al. (2007).
7. First, the two variables are highly correlated; the value of the contemporaneous correlation between the two
was 0.65. Second, estimation results were more significant when the population was used as the SIZE variable.
References
Anderson, J. E., and E. Van Wincoop. 2004. Trade costs. Journal of Economic Literature 42 (3):691–751. doi:10.1257/
0022051042177649.
Bala, V., and N. Van Long. 2005. International trade and cultural diversity with preference selection. European Journal of
Political Economy 21 (1):143–62. doi:10.1016/j.ejpoleco.2004.09.001.
Boisso, D., and M. Ferrantino. 1997. Economic distance, cultural distance, and openness in international trade: empirical
puzzles. Journal of Economic Integration 12 (4):456–84.
Chaloupka, F. 1991. Rational addictive behavior and cigarette smoking. Journal of Political Economy 99 (4):722–42.
doi:10.1086/261776.
Disdier, A. C., and T. Mayer. 2007. Je t’aime, moi non plus: Bilateral opinions and international trade. European Journal of
Political Economy 23 (4):1140–59. doi:10.1016/j.ejpoleco.2006.09.021.
Disdier, A. C., S. H. T. Tai, L. Fontagne, and T. Mayer. 2010. Bilateral trade of cultural goods. Review of World Economics 145
(4):575–95. doi:10.1007/s10290-009-0030-5.
Eichengreen, B., and D. Irwin. 1998. The role of history in bilateral trade flows. In The regionalization of the world economy, ed.
J. Frankel, 33–57. Chicago: University of Chicago Press.
Felbermayr, G. J., and F. Toubal. 2010. Cultural proximity and trade. European Economic Review 54 (2):279–93. doi:10.1016/j.
euroecorev.2009.06.009.
DO KOREAN WAVE EXPORTS PROMOTE GOODS EXPORTS? 1403
Francois, P., and T. Van Ypersele. 2002. On the protection of cultural goods. Journal of International Economics 56:359–69.
doi:10.1016/S0022-1996(01)00119-2.
Frankel, J., and A. Rose. 2002. An estimate of the effect of common currencies on trade and income. Quarterly Journal of
Economics 117 (2):437–66. doi:10.1162/003355302753650292.
Frankel, and A. Jeffrey. 1997. Regional trading blocs in the world trading system. Washington, DC: Institute of International
Economics.
Girma, S., and Z. Yu. 2002. The link between immigration and trade: Evidence from UK. Weltwirtscahftliches Archiv 138:115–
30. doi:10.1007/BF02707326.
Gould, D. 1994. Immigration links to the home country: Empirical implications for US bilateral trade flow. The Review of
Economic and Statistics 76:302–16. doi:10.2307/2109884.
Hanson, G., and C. Xiang. 2009. International trade in motion picture services. In International trade in services and intangibles
in the era of globalization, ed. M. Reinsdorf, and M. J. Slaughter, 203–22. Cambridge, MA: National Bureau of Economic
Research.
Head, K., and J. Ries. 1998. Immigration and trade creation: Econometric evidence from Canada. Canadian Journal of
Economics 31 (1):47–62. doi:10.2307/136376.
Huang, X. 2009. Korean Wave’–The Popular culture, comes as both cultural and economic imperialism in the East Asia. Asian
Social Science 5 (8):123–30. doi:10.5539/ass.v5n8p123.
Hutchinson, W. K. 2005. Linguistic Distance as a Determinant of Bilateral Trade. Southern Economic Journal 72:1–15.
doi:10.2307/20062091.
Isphording, I. E., and O. Sebastian. 2013. The costs of babylon—linguistic distance in applied Economics. Review of
International Economics 21 (2):354–69. doi:10.1111/roie.12041.
Kang, H. G. 2009. An economic effect of korean cultural contents on korea’s exports and FDI in southeast asian countries. Korea
Trade Review 34 (1):29–47. in Korean.
Kim, J. G., and S. Y. Ahn. 2012. An empirical study on effects of Korea’s cultural exports. Journal of Korea Trade 16 (2):25–48.
Kónya, I. 2006. Modeling Cultural Barriers in International Trade. Review of International Economics 14:494–507. doi:10.1111/
j.1467-9396.2006.00626.x.
Lee, J., Y. Kim, and W. Woo. 2014. Impact of K-POP on export: Based on youtube page view and google search. Journal of
Marketing Management Research 19 (4):83–97. in Korean.
Lohmann, J. 2011. Do language barriers affect trade? Economics Letters 110:159–62. doi:10.1016/j.econlet.2010.10.023.
Marvasti, A., and E. R. Canterbery. 2005. Cultural and other barriers to motion pictures trade. Economic Inquiry 43 (1):39–54.
doi:10.1093/ei/cbi004.
Melitz, J. 2008. Language and foreign trade. European Economic Review 52 (4):667–99. doi:10.1016/j.euroecorev.2007.05.002.
Olivier, J., M. Thoenig, and T. Olivier. 2008. Globalization and the dynamics of cultural identity. Journal of International
Economics 76:356–70. doi:10.1016/j.jinteco.2008.06.009.
Park, S. C., and J. I. Choe. 2009. The trade creation effects of hallyu. Economics Analysis, Bank of Korea Economic Research
Institute 15 (1):73–96. in Korean.
Park, Y. S. 2014. Trade in cultural goods: A case of the korean wave in Asia. Journal of East Asian Economic Integration 18
(1):83–107. doi:10.11644/KIEP.JEAI.2014.18.1.276.
Park, Y. S. 2015. Does the rise of the korean wave lead to cosmetics export? Journal of Asian Finance, Economics and Business
2 (4):13–20. doi:10.13106/jafeb.2015.vol2.no4.13
Rauch, J. E., and V. Trindade. 2002. Ethnic chinese networks in international trade. Review of Economics and Statistics 84
(1):116–30. doi:10.1162/003465302317331955.
Rose, A. K. 2004. Do we really know that the WTO increases trade? American Economic Review 94:98–114. doi:10.1257/
000282804322970724.
Santos Silva, J. M. C., and S. Tenreyro. 2006. The log of gravity. Review of Economics and Statistics 88:641–58. doi:10.1162/
rest.88.4.641.
Santos Silva, J. M. C., and S. Tenreyro. 2011. Further simulation evidence on the performance of the Poisson pseudo-maximum
likelihood estimator. Economics Letters 112 (2):220–22. doi:10.1016/j.econlet.2011.05.008.
Shim, D. 2006. Hybridity and the rise of Korean popular culture in Asian. Media, Culture & Society 28 (1):25–44. doi:10.1177/
0163443706059278.
Siwek, S. E. 2005. Engines of growth: Economic contributions of U.S. intellectual property industries. Economists incorporated
for NBC Universal. Washington, D.C.: IIPA.
Throsby, D. 1999. Cultural Capital. Journal of Cultural Economics 23 (1):3–12. doi:10.1023/A:1007543313370.
Wagner, D., K. Head, and J. Ries. 2002. Immigration and the trade of provinces. Scottish Journal of Political Economy 49:507–
25. doi:10.1111/sjpe.2002.49.issue-5.
Yang, J. 2012. The korean wave (Hallyu) in East Asia: A comparison of chinese, japanese, and taiwanese audiences who watch
korean TV dramas. Development and Society 41 (1):103–47. doi:10.21588/dns.
1404 C.-G. HUH AND J. WU
