Page 1 of 7

Archives of Business Research – Vol. 9, No. 9

Publication Date: September 25, 2021

DOI:10.14738/abr.99.10831. Kurihara, Y. (2021). Does the Linder Hypothesis Hold? Recent Japanese Case. Archives of Business Research, 9(9). 69-75.

Services for Science and Education – United Kingdom

Does the Linder Hypothesis Hold? Recent Japanese Case

Yutaka Kurihara

Department of Economics, Aichi University, Aichi, Japan

ABSTRACT

The theory of comparative advantage has for a long time played one of the most

important roles among the theories of international economics. The theory says

that a country has to specialize in producing and exporting goods and services that

the country can produce at lower opportunity cost than other goods and services.

Achieving this, both countries can benefit. The Linder hypothesis focuses on

internal demand in a country. It states that the pattern of export goods is

determined by internal demand structure. This study empirically analyzes this

Linder hypothesis for the recent case of Japan. The Gravity model can be used to

obtain theory-consistent estimates of the Linder hypothesis. This model can

estimate bilateral international trade on the basis of cross-section data. The results

show that the Linder effect has been holding well. Also, FTA (Free Trade Agreement)

has been having positive impacts on international trade.

Keywords: demand structure; income; Japan; Linder hypothesis.

INTRODUCTION

The theory of comparative advantage has for a long time played one of the most important roles

among the theories of international economics. The theory says that an economy has to

specialize in producing and exporting goods and services that the economy can produce at

lower opportunity cost than other goods and services. Achieving this, both countries can

benefit. The has been a lot of discussion for this theory both from both theoretical and empirical

viewpoints, or both from academic and real-world fields.

Among them, [1] extended a [2]’s model, and indicated that companies with different

productivity can coexist in one industry as each company faces concerns for their productivity

level before deciding whether they should enter the industry or not.

While [1] emphasizes supply instead of demand, [3] focuses on demand. The Linder hypothesis

focuses on internal demand in a country. It states that the pattern of export goods is determined

by internal demand structure. Why can such a hypothesis be established? [3] states that

companies tend to produce according to economic needs which usually come from internal

demand. When the product activity starts, such companies face limited domestic demand, so

companies would like to export their products. [3] then indicated that further exposures for

companies to international trade led to reallocations toward much more productive ones taking

into account some aspects such as demand structures. If two countries (such as US and

Germany) have the same demand structure, exported and imported products of the countries

would be the same. However, the products are in reality different in some aspects such as

quality. So, the demand structures can be a deterministic element for international trade

Page 2 of 7

70

Archives of Business Research (ABR) Vol. 9, Issue 9, September-2021

Services for Science and Education – United Kingdom

pattern which are affected by a lot of factors. Empirical analysis would help provide more

definitive answers. For this aspect, income per capita is mostly used for the representative

factor that makes demand structures.

Comparative advantage still exists. However, among US, Germany, Japan and so on, cars from

those countries that have comparative advantage are exported and imported. To explain this

phenomena, other theories would be needed. This study focuses on this point and whether or

not Linder hypothesis holds or not is empirically examined for a recent Japanese case.

This study is structured as follows. Following this section, existing studies are listed. Empirical

methods are described, the analyses are conducted, and the results are examined. Finally, this

study ends with a brief summary.

EXISTING STUDIES

Trade patterns have been disputed a lot since the 1800s. It would be impossible and

unnecessary to summarize, however, new theories have been provided and discussed (for

example, [1][2], [3], and see for example, [4]). However, this section’s discussion is limited to

Linder hypothesis or other related strong hypotheses.

The similarity of income is a key issue for the Linder hypothesis. Moreover, [5] found that trade

could benefit by specialization in demand instead of production. Focusing on demand instead

of supply is an important point for the Linder hypothesis. [6] suggested that the presence of

both demand and supply perform roles for per capita income influencing intra-industry trade.

[7} revealed that international trade diversification effects are larger for the European

countries with per capita incomes which are close to the European average. [7] also indicated

that non-EEC countries with per capita incomes near to the European average income have

strong negative international trade diversification effects. [8] showed that developed countries

in APEC (Asia Pacific Economic Cooperation) being near per capita incomes have inclined to

trade more with each partner than developing countries in APEC. [9] found that while countries

which have differences in incomes between countries increase international trade, countries

which have differences in income dispersion decline international trade. There are some

differences, however, most studies show that income per capita can be employed to check

whether the Linder hypothesis fits or not. There seems to be a consensus about using income

per capita when examining the Linder hypothesis.

Whether the Linder hypothesis holds well or not should be examined by empirical analyses.

Sixty years have passed since the publication of [3], however, a lot of studies have been

provided to examine the validity of the Linder hypothesis by now. It should be noted that most

of them accept the hypothesis, not only the theoretical views but also empirical views.

[10] found that Linder hypothesis fits well. [11] showed that the Linder effect could be

confirmed for both developed and less developed economies. [12] indicated the validity of the

Linder hypothesis for 18 of 19 OECD (Organization for Economic and Cooperation and

Development) countries. The empirical results to check whether the Linder hypothesis holds

or not were robust. [13] found support for the Linder hypothesis. It was also found that the

Linder hypothesis gained approval in the 1990s using panel data of international trade from 63

Page 3 of 7

71

Kurihara, Y. (2021). Does the Linder Hypothesis Hold? Recent Japanese Case. Archives of Business Research, 9(9). 69-75.

URL: http://dx.doi.org/10.14738/abr.99.10831

countries for 1970, 1980, 1990, and 1992. [14] also used panel data and found support for the

Linder hypothesis. It used panel data of bilateral international trade between the triad (EU15,

USA and Japan) countries and their 57 most related trading partners’ countries for the period

1986–1997. [15] used the panel data of OECD economies and showed some grounds for the

support of the Linder hypothesis. [16] found the validity of the Linder hypothesis using data

which Germany performs international more intensively with trading partners with similar per

capita incomes. However, [17] revealed that the support of the Linder hypothesis for Croatia

could not be admitted. Instead, the structure of Croatia’s international trade is in accordance

with the gravity model of international trade. There are also studies which focus on sector levels

of international trading. [18] used the data of Russia and five regional groups, and showed that

Russia’s manufacturing products and raw material integration employing the Heckcher-Ohlin’s

theory is not similar to that of other BRICs members from the view of the Linder hypothesis.

[19] indicated that the Linder hypothesis was not in support of international trade of

differentiated agri-food products. [20] found that the Linder hypothesis fit at the total level.

Concretely, it showed that similar per capita incomes and the inequality of incomes in two

countries increased international service trade. However, it also found that the precise impact

was different among each service trading sector. [21] indicated that once controlling for the

effect of commonality on the extensive margin was taken into account, the Linder hypothesis

fit well at more aggregate levels. [22] suggested that the Linder hypothesis could be accepted

but held only when the hypothesis is obeyed as a sector level prediction. Moreover, [23] showed

that FDI (foreign direct investment) occurred between countries with similar per capita income

levels which predicts a Linder hypothesis for FDI. However, this study does not address FDI. It

examines whether the Linder hypothesis holds or not between Japan and its major trading

partners.

EMPIRICAL ANALYSIS

The Gravity model can be used to obtain theory-consistent estimates of the Linder hypothesis.

The gravity model seems to be a simple but robust analysis to estimate bilateral international

trade on the basis of cross-section data. Economic size and distance between trading partners

are usually used for theoretical and empirical analysis.

The typical equation is as follows (1)

Xij = α!"!#

$"# (1)

X stands for international trade. Y denotes economic size (income per capita), D denotes

distance, i, j denote countries, and α is a constant term.

The model has been said that it was introduced by [24] etc., and its theoretical foundation has

been discussed in many studies. [25] showed that the unobservable trade costs are large, and

that omitting them sometimes caused not only a bias of the observable trade costs but also a

bias of country-specific variables that are either solved implicitly or estimated explicitly. [26]

showed that the impact of distance on trade has fallen steadily over time. For the empirical

methods, [27] found that heterogeneity problem is not taken into account adequately, and

gravity models could overestimate the effects of international trade volume. [28] suggested that

the variety of intra-regional trade costs have effects on relative bilateral trade costs and