This document discusses using cluster analysis to measure economic similarity between East Asian countries and identify optimal groups for economic cooperation. It finds that specifying 4 clusters provides the best solution, with China as its own cluster, Korea and Taiwan in another, and Cambodia, Laos, Mongolia and Papua New Guinea in a third cluster. The analysis suggests cluster composition can help sequence economic cooperation by starting within identified similar country groups.
Marco Kamiya in Medellin World Urban Forum Productive Transformation in Quito...
East Asian Economic Similarity
1. HOW SIMILAR ARE THE EAST ASIAN ECONOMIES?
A CLUSTER ANALYSIS PERSPECTIVE ON ECONOMIC
COOPERATION IN THE REGION
May 4, 2012
Alexandre Repkine
2. INTERCONNECTION AND COOPERATION
1997 Asian financial crisis
Japanese statement about interest rates
Depreciation of Thai baht
Capital flight and currency depreciation in Asia
Economic cooperation
Expanding the set of consumer choices
Institutional framework helping to absorb shocks
3. WHY ECONOMIC COOPERATION?
Comparative advantage
Production possibility frontiers
Gains from trade
Examples of economic cooperation
Free Trade Agreement
Regional Trade Agreement
Customs Union
Currency Union
4. ECONOMIC COOPERATION IN EAST ASIA
Growth in the number of FTAs from 2 in 1975 to 16 in 2000
By 2010: more than 45 FTAs concluded in the region
5. SCOPE AND SEQUENCING OF FTA
The scope of East Asian FTAs is constantly
increasing
The problem of sequencing
How do countries form economic cooperation
groups?
How do smaller groups become larger groups?
What is the principle governing the process of
economic agglomeration?
6. ROLE OF ECONOMIC SIMILARITY
It makes sense for the economically similar
countries to form groups of economic
cooperation prior to doing so with more
different countries
E.g.Philippines first forming alliance with Thailan
d before concluding an FTA with EU
How do we measure similarity
Why are groups of similar countries better off
economically?
7. WHY ARE SIMILAR COUNTRIES
BETTER OFF TOGETHER?
Gravity models
Helpman and Krugman (1985)
Geographically close countries trade more
Countries with similar-sized GDPs trade more
Similar countries in general trade more
(Bergstrand and Egger, 2007)
Baier and Bergstrand (2004)
Cooperation and trade between economically similar
countries increases welfare
Similarity measured in terms of distance, GDPs,
remoteness to ROW, and K/L ratios
8. DEFINING ECONOMIC SIMILARITY
What countries can be viewed as similar and
on what grounds?
Language (Korea vs Japan)
Historical heritage (Korea vs China)
Current trade and political links (Korea vs US)
Once groups, or clusters, of similar countries
are identified, economic integration can be
based on those clusters
9. AN OUTLINE OF CLUSTER ANALYSIS
Every country is a collection of characteristics
GDP size
Population
Human development index
Trade openness
How does one compare collections of number
s?
Cluster analysis employs a measure of
generalized distance based on several charac
teristics
10. AN EXAMPLE OF SIMILARITY MEASURE:
EUCLIDEAN DISTANCE
X and Y are any two
economic parameters
(GDP per capita, % ur
ban population)
11. EUCLIDEAN DISTANCES IN ASIA
• Number of distances grows qu
Share of Urban Population,
ickly with more countries adde
d
• Vietnam appears to be similar t
o Cambodia
• Should we include Malaysia in
one group with China and Thai
land?
%
• Dissimilarity matrix summarize
s the information about eco
Real GDP per Capita, USD
nomic distances between co
untries
12. DATA SOURCES AND SUMMARY STATISTICS
Mean Standar Min Max Source Computatio
d Deviat n
ion
Structural Shares
Agriculture, % 18.35% 0.13 1.46% 4.53% ADB SU501/SU499
Manufacturing, % 23.21% 0.12 5.96% 48.46% ADB SU504/SU499
Trading, % 13.98% 0.05 5.76% 21.66% ADB SU508/SU499
International Trade
Trade Openness, % 106.59% 46.84 24.31% 213.75% Penn openk
Economic Development
GDP per Capita, $ $10167 11010 $1707 $34223 Penn rgdpl
Share of Urban Population 44.10% 19.82 12.52% 84.68% ADB, Trading SU1223
Economics
Human Development Inde 0.62% 0.14% 0.44 0.89% ADB SU1023
x, %
Economic Size
Population, mn people 61.2 65.6 2.9 240.3 Penn POP
GDP in constant 2005 pric 1229 2195 7.46 9276 ADB SU499
es, bn USD
Note: ADB stands for the Asian Development Bank’s statistical database https://sdbs.adb.org/sdbs/index.jsp), Penn for the Penn World Table version 7 (http://
pwt.econ.upenn.edu/php_site/pwt70/pwt70_form.php). Data on urban population shares in Papua New Guinea is taken from the Trading Economics Indicator
s database (http://www.tradingeconomics.com/papua-new-guinea/urban-population-percent-of-total-wb-data.html). The “Computation” column is based on t
he variable names provided by the original databases. Population statistics are given for the subsample that excludes China.
13. EAST ASIAN COUNTRIES COVERED
1. Mongolia 7. Papua New Guine
2. Korea a
3. China 8. Vietnam
4. Taiwan (Chin 9. Indonesia
a) 10. Malaysia
5. Cambodia 11. Philippines
6. Laos 12. Thailand
Japan is excluded because of its special status of the second largest econo
my in the world (until recently) and its currency being the only hard currenc
y in the region.
14. CLUSTERING BY K-MEANS
Specify the number of groups in advance
Make sure that the overall distance within each cluster of the
individual observations from the cluster center (i.e. centroid)
is minimized
Proceed in iteration so that each country may change its cl
uster several times in the process
Random assignment of group centers initially
Realistic group membership: clusters of 2, 3, and 4 countries
15. K-MEANS CLUSTERING,
EUCLIDEAN DISTANCE MEASURE
2 groups 3 groups 4 groups
Group 1
• Results similar to the ca
China
Indonesia
se when Manhattan
Korea Korea Korea (city block) measure is u
Malaysia Malaysia sed
Philippines
Taiwan Taiwan Taiwan
Thailand • China forms one –count
Group 2 ry cluster
Cambodia Cambodia Cambodia
Laos Laos Laos
Mongolia Mongolia Mongolia • Korea and Taiwan
Papua New Guinea Papua New Guinea Papua New Guinea
Vietnam Vietnam • Cambodia, Laos, Mongo
Indonesia lia, Papua New Guinea
Thailand
Philippines
Group 3 • Indonesia, Philippines,
China China Thailand
Group 4
Indonesia
Malaysia
Vietnam
Philippines
Thailand
16. K-MEDIAN CLUSTERING,
EUCLIDEAN DISTANCE MEASURE
2 groups 3 groups 4 groups
Group 1
China
Indonesia • Mongolia joins more adva
Korea Korea Korea nced group with Thailand
Malaysia Malaysia and Indonesia in 4-gro
Philippines
Taiwan Taiwan Taiwan up solution
Thailand
Group 2 • China is still forming a
Cambodia Cambodia Cambodia
Laos Laos Laos
separate cluster in 4-grou
Papua New p solution
Papua New Guinea Papua New Guinea Guinea
Vietnam
Mongolia Mongolia
• Korea and Taiwan still sti
ck together
Group 3
• Indonesia, Philippines an
China
Indonesia Indonesia d Thailand
Malaysia Mongolia
Philippines Philippines
Thailand Thailand
Vietnam Vietnam
Group 4
China
17. HOW MANY CLUSTERS TO CHOOSE?
Both K-means and K-median clustering proc
edures need a priori the number of clusters
Hierarchical procedures start with each count
ry being its own cluster, then agglomerating u
p
Stopping rules
Pseudo-F value
Duda-Hart value
18. AGGLOMERATION INTO CLUSTERS
Based on dissimilarity matrices
(normally Euclidean distances)
Merging clusters that are similar
Single-linkage
Complete linkage
Average linkage
Cluster centroid
Ward’s method
19. OPTIMAL NUMBER OF CLUSTERS
Average Single Complete Centroid Ward’ s
Linkage Linkage Linkage Method Four clusters a
Number of F DH F DH F DH F DH F DH
Clusters
ppears to be th
2 e optimal sol
5.33 0.6 5.33 0.92 5.33 0.58 5.33 0.79 4.47 0.46
ution
3 6.99 0.55 2.98 0.75 7.3 0.41 4.26 0.56 7.3 0.41
4 8.94 0.54 3.19 0.68 9.13 0.47 6.62 0 9.13 0.47 Single linkage i
5
s exceptional
9 0.33 3.91 0.64 9 0.33 5.02 0.77 9 0.48
6 8.34 0 4.86 0.68 8.34 0.48 4.86 0.49 8.6 0.33 Average numb
7 7.76 0.48 5.36 0.82 8.93 0 7.76 0 8.93 0
er of clusters 4.
8
2
9.54 0 4.62 0.58 9.54 0.23 6.72 0.65 9.54 0.23
20. CLUSTER SEQUENCING: DENDROGRAMS
4-cluster grouping
coincides with
K-means solution
(Euclidean distanc
e)
China staying
separately
Korea and Taiwan
Cambodia, Laos,
Papua New Guine
a
Mongolia
ambiguous
21. CONCLUSIONS
Economic theory suggests similar countries should trade an
d cooperate more since such cooperation increases t
heir total welfare
Cluster analysis determines what countries are similar and s
uggests two approaches
Composing clusters if number of groups known
Hierarchical approach (dendrograms)
Results surprisingly stable over various procedures
Results could be used as background for future policy makin
g on regional economic cooperation in East Asia