ARTNeT- GIZ Capacity Building Workshop on Introduction to Gravity Modelling:
19-21 April 2016, Ulaanbaatar
Session 2: Introduction to the basic gravity model
Dr. Witada Anukoonwattaka
Trade and Investment Division, ESCAP
anukoonwattaka@[Link]
1
Introduction
• Gravity model is a very popular econometric model in
international trade
• Origins with Tinbergen (1962). Thousands of published
articles and working papers since then.
– “Some of the clearest and most robust findings in empirical
economics.” (Leamer & Levinsohn, 1995)
• The name came from its utilizing the gravitational force
concept as an analogy to explain the volume of bilateral trade
flows
• Initially, it was not based on theoretical model, but just
intuition only
• Later on, a range of rigorous theoretical foundation has been
given.
2
Introduction
• Gravity’s main comparative advantage lies in its ability to use
real data to assess the sensitivity of trade flows with respect
to policy factors we are interested in.
• Numerous applications looking at different types of factors
affecting trade costs, and their impacts on trade flows:
– Transport costs.
– Tariffs and non-tariff barriers.
– Regional integration agreements, currency unions, and the
– GATT/WTO.
– Time delays at export/import and trade facilitation.
– Governance, corruption, and contract enforcement.
3
Introduction
• In recent years, intuition is not enough.
• Gravity models have become a complex business: back to
microfoundations!
– Different microfoundations imply different estimation techniques.
– Use of sectorally disaggregated data, and broad country samples,
brings out new issues for theory and empirics.
• To do good applied/policy research, it is important to be on
top of the latest developments in the literature.
4
The traditional gravity model
5
• Concepts and stylized facts of the gravity
approach
• Example of applications
• Identifying (“trade potentials”) using gravity
approach
6
Gravity force in Physics
MiM j
Fij G
d ij2
7
Gravity Analogy
Gravity force equation Intuitive gravity for trade
Yi Y j
X ij C
tij
MiM j
Fij G X ij exports (or trade) from i to j ,
2
D ij
C contstant,
Gravity force between two
objects depends on their Y economic mass ( GDP),
masses and inversely t trade costs between two countries
proportional to the square of
distance between them. distance, adjacency, .., " policy factors".
Export (or trade) between two countries
depends on their economic masses and
negatively related to trade costs between
them.
8
Trade and combined GDP
gen ln_gdp_both = ln(gdp_exp*gdp_imp)
twoway (scatter ln_trade ln_gdp_both) (lfit ln_trade ln_gdp_both)
Based on AP export data 2013 provided in WITS 9
Trade and distance
twoway (scatter ln_trade ln_distance) (lfit ln_trade ln_distance)
Based on AP export data 2013 provided in WITS
10
What is the gravity model?
• Gravity model is a very popular econometric model in
international trade
• The name came from its utilizing the gravitational force
concept as an analogy to explain the volume of bilateral trade
flows
– Proposed by Tinbergen (1962)
• Initially, it was not based on theoretical model, but just
intuition only
• Later on, a range of rigorous theoretical foundation has been
given.
– The most well-known benchmark so far is Anderson and van Wincoop
(2003).
11
Intuitive gravity model of trade:
Yi Y j
X ij C
tij
• Larger countries trade more than smaller ones
• Trade costs between two trade partners reduce trade
between them.
Empirical equation for basic gravity model:
ln X ij b0 b1 ln(Yi ) b2 ln(Y j ) b3 ln(t ij ) eij
b1 , b2 0; b3 0
A 1% change in Yi is associated with a b1 % chage in X ij .
12
Proxies for trade costs
• Distance
• Adjacency
• Common language
• Colonial links
• Common currency
• Island, landlocked
• Institutions, infrastructures, migration flows,..
• Bilateral tariff barriers
13
Why is it so popular?
• Intuitively appealing
• Fits with some important stylized facts
• Easily to use real data to explain trade flows with respect to
policy factors.
• Estimation using OLS
14
Applications of gravity models
• Analysis of elasticities of trade volumes
- Regional Trade Agreements (RTA), "natural regionalism" (Frankel & Wei,
1993, Baier & Bergstrand 2005)
- WTO membership
- Impact of NTBs on trade (Fontagné et al. 2005)
- Cost of the border (Mac Callum, Anderson & van Wincoop 2003)
- Impact of conflicts on trade
- FDI & trade: complements or substitute (Eaton & Tamura, 1994; Fontagné,
2000)
- Effect of single currency on trade (Rose, 2000)
- Trade patterns: inter and intra-industry trade (Fontagné, Freudenberg &
Péridy, 1998)
- Diasporas (community of immigrants)
- Internet
Applications of gravity models
• Analyse predicted trade flows and observe
differences between predicted and observed flows
(analysis of residuals)
- Trade potentials of economies in transition (out-of sample
predictions)
- Identify the natural markets and markets with an untapped
trade potential
- Predicted values are used in some cases as an input for CGE
modeling (Kuiper and van Tongeren, 2006)
- Use of confidence intervals in addition to predicted values,
in order to take into account the residual variance
Examples of Applications
• Effects of regional integration on trade
Do RTAs boost trade between members?
ln X ij b0 b1 ln(Yi ) b2 ln(Y j ) b3 ln(tij ) b4 (dummyRTAij ) eij
Do RTAs reduce exports from non - members?
ln X ij b0 b1 ln(Yi ) b2 ln(Y j ) b3 ln(tij ) ...
b4 (dummy BothInRTAij ) b5 (dummy OneInRTAij ) eij
• By using dummy variables, gravity models provide a crude measure of RTA
impact on trade but cannot distinguish the precise mechanisms.
• Both b4 > 0 and b5>0 implies trade creating RTA.
• Only b4> 0 while b5<0 implies trade diverting RTA
See, World Bank (2005) for survey.
17
• Two important limitations related to using gravity models for estimating
the impact of a RTA:
1. RTAs may be endogenous variables (ie. the causal link between the
formation of a RTA and trade flows). This endogeneity affects gravity-
based estimates.
2. Recent literature is replete with models in which regional integration
agreements are formed in the pursuit of other, non-trade goals (see, for
instance, Limao, 2006) or in which they have “non-traditional” gains (see
Ethier, 1998).
– South–South agreements have been rather more successful in non-trade dimensions
like the management of common resources than in the dimension of pure trade-
liberalization.
• The analysis of RTAs should avoid limiting itself to measuring trade
diversion and creation, although these are important issues for the
welfare of member countries.
18
Examples of Applications
• Effects of institutional weakness on trade
How does corruption affect trade?
ln X ij b0 b1 ln(Yi ) b2 ln(Y j ) b3 ln(tij ) b4 ln(corruptioni ) eij
Anderson and Marcouiller use a 58-country gravity model and corruption data
from the World Economic Forum to show that:
• Institutional weaknesses, generally corruption and lack of contract
enforceability, have a significant negative impact on trade.
• If Latin America increased measured institutional quality to the same level
as the EU, their trade would increase by about 30%: about the same as
with a major tariff cut.
19
Examples of Applications
• Effects of trade facilitation on trade
How much can trade facilitation boost bilateral trade?
ln X ij b0 b1 ln(Yi ) b2 ln(Y j ) b3 ln( d ij ) b4 ln(timeiX ) eij
• Djankov Freund & Pham (2010) use a gravity model with Doing Business
data on border crossing times (98 countries) to show that:
– Slower border crossing times can significantly reduce bilateral trade:
One extra day reduces exports by 1%.
– Time-critical agricultural and manufactured goods are particularly
sensitive to border crossing times:
• Agriculture: Fresh fruits and vegetables.
• Manufactures: Electronic goods; parts and components.
• Landlocked countries are particularly sensitive to border crossing times:
One extra day reduces exports by as much as 4%.
20
Examples of Applications for landlocked countries
• Effects of being landlocked countries on trade
How much does being lancdlocked reduce trade?
ln X ij b0 b1 ln(Yi ) b2 ln(Y j ) b3 ln(d ij ) b4 ln(landlockedij ) ... eij
Raballand (2003) applies a gravity approach to (46 countries of which 18 are
landlocked) show that being landlocked country would reduce trade by 80%
(when measured by a dummy variable).
• Determinants of trade costs
An example of trade-cost regressions, (See, Moïsé and Sorescu, 2013, etc.)
Without
landlocked
With landlocked
21