Transcript Monopolistic Competition and International Trade Between

GRAVITY MODEL AND THE COUNTRIES
CLASSIFICATION
Elżbieta Czarny, Warsaw School of Economics
Jerzy Menkes, Warsaw School of Economics
Katarzyna Śledziewska, Warsaw University
Midwest’2008
1
Aims of the paper
• Review the Helpman’s prediction (more similar countries trade
more intensively)
• Propose a new division of the world: OECD, non-OECD
fulfilling membership criteria and non-OECD not fulfilling
membership criteria
• Examine trade between pairs of countries in the period of 28
years (1977-2005) and aggregate results to show trade between
three groups of countries (OECD, non-OECD fulfilling
membership criteria and non-OECD not fulfilling membership
criteria)
• Discuss the impact of the new split on the test of Helpman’s
theorem and compare the results with those obtain in the test
with the traditional division (OECD, non-OECD)
Midwest’2008
2
Hypothesis
• For testing Helpman’s
Theorem (1987) the
established division into OECD and non-OECD
countries is misleading and a new division of
countries should be proposed
• Helpman’s Theorem can be applied only to developed
countries and not to the less developed ones
Midwest’2008
3
Justification 1: Changing criteria of territorial division of
the world
Ternary division of the world (determined by geographical
closeness creating interdependency, and natural obstacles preventing
contacts):
I. states,
II. continents,
III. the globe.
Reasons for change in the above division of the world (a.o.):
a) natural obstacles became less important
b)
geographical discoveries and establishing of transcontinental
empires make intra-imperial tights stronger than tights among
countries on the same continent (in the 18-th century Spain,
economical, social and cultural distance from Spain to its American
possessions was shorter than from Catholic Spain to Protestant
England),
c) international law is a tool of implementing national interests of
states; participation in international organisations and agreements
builds a new form of closeness of states.
Midwest’2008
4
Justification 2: OECD as a base for territorial
division of the world
Organisation for Economic Co-operation and Development
(OECD) – characteristics:
- a transcontinental space (in geophysical sense),
- economically, legally and politically homogeneous,
- main criteria of membership: open market economy,
high level of development, OECD legal instruments
implementation, pluralistic democracy and human
rights respect
Midwest’2008
5
Justification 3: OECD as a base for territorial
division of the world
We have to be aware that still many countries meeting the
OECD - membership criteria stay outside OECD
(Bulgaria, Cyprus, Estonia, Lithuania, Latvia, Malta,
Romania, Slovenia; OECD maintains close relations
with over 70 countries (25 got the status of a
permanent/full participant of of the OECD Committee;
50 cooperate with OECD in many fields).
Many above mentioned countries meet the formal
accession criteria. As this group is more like the OECD
countries than the other ones, it is a mistake to group
those states just as „non-OECD”.
Midwest’2008
6
Econometric tests of Helpman’s prediction in
the past
• Helpman [1987]
– OECD countries
– confirmed his hypothesis for OECD countries
• Hummels and Levinsohn [1995]
– OECD and non-OECD countries
– confirmed Helpman’s findings for OECD countries
– not rejecting Helpman’s hypothesis for non-OECD countries.
• Debaere [2005]
– OECD and non-OECD countries
– confirmed Helpman’s findings only for OECD countries
Midwest’2008
7
The estimated equation:
 X t ij  X t ji 
i
j
ij

ln i




ln(
s

s
)


ln
1

dispersion
ij
t
t
t
j 
Y

Y
t
 t



αij fixed effects: bilateral transportation costs,
language barriers, cultural barriers, tariffs or
distance
 Yi
1   i
j
Y Y
j
Y
s 
2
  Yj
   i
j
 Y Y



2
Size dispersion index
Shows how the volume of trade will be related to the relative size of
countries
j
Yw
Midwest’2008
8
Criteria of OECD membership (= criteria of
our division of the world)
Three main criteria of OECD membership:
1. democracy and respecting human rights,
2. open market economy,
3. relatively high level of development
Midwest’2008
9
The Freedom House - division of the world into free,
partly free and not-free countries
free + partly free
country
+
WTO/GATT
member
+
GDP pc PPP >=
min GDP pc
PPP OECD
developed countries
Operationalisation of the OECD membership criteria
Midwest’2008
10
Research
• Panel data from the World Bank and COMTRADE
• Years: 1977-2005
• Countries: all available
• 2 Results
– Absolute value of t statistics in parentheses
– *significant at 5%; ** significant at 1%
Midwest’2008
11
Tests of the Gravity Equation with Value of (Trade/current GDP) for pairs of
countries as dependent variable; results for authors’ classification of developed
(OECD + non-members fullfilling membership criteria) and less developed
countries
developed countries with developed
countries
developed countries with less developed
countries
(1)
lns (GDP - current USD)
less developed countries with less
developed countries
(2)
(3)
0.336
0.356
-0.173
(11.83)**
(15.73)**
(4.81)**
1.010
0.872
0.081
(46.56)**
(57.72)**
(2.43)*
9.317
8.507
3.890
(65.67)**
(68.02)**
(15.74)**
observations
22929
55470
23536
number of group (k1 k2)
1843
5684
3478
R-squared
0.35
0.32
0.02
lnsim (GDP - current USD)
constant
Midwest’2008
12
For comparison:
Tests of the gravity equation with value of (trade/current GDP) for
pairs of countries as dependent variable, results reported for countries’ groups, estimation
for OECD and non-OECD countries
OECD countries with
non-OECD countries
OECD countries with
OECD countries
(1)
(2)
non-OECD countries with
non-OECD countries
(3)
0.154
0.485
-0.013
(4.33)**
(21.31)**
-0.48
0.955
0.963
0.383
(28.57)**
(77.14)**
(15.51)**
9.368
9.501
5.29
(61.91)**
(85.51)**
(29.38)**
8268
51215
42452
number of group (k1 k2)
434
3825
4915
R-squared
0.25
0.45
0.12
lns (GDP - current USD)
lnsim (GDP - current USD)
constant
observations
Midwest’2008
13
For comparison: Tests of
the gravity equation with value of (trade/current GDP) for
pairs of countries as dependent variable, for countries grouped depending on their level of
development
Developed
economies &
Developed
economies
lns (GDP - current USD)
0.829
(25.83)**
lnsim (GDP - current USD)
1.639
constant
0.087
(2.58)**
0.934
(34.17)**
-0.544
(3.50)**
Developing
economies &
Developing
economies
0.414
(11.23)**
(61.36)**
observations
Developing
economies &
Developed
economies
0.721
(22.65)**
-4.939
(23.33)**
-7.291
(30.47)**
12566
27424
41235
number of group(k1 k2)
1124
3688
7670
R-squared
0.25
0.05
0.02
Midwest’2008
14
Results
• Confirmation of Helpman’s theorem for developed countries
• Division of countries into two groups:
• OECD together with non-OECD countries fulfilling membership criteria
• non-OECD countries not fulfilling membership criteria
... turned out to be better fitting with the theory than the
simple division into OECD and non-OECD countries (and
better than the split developed and developing countries in the
World Bank setting)
Midwest’2008
15