Christian Helmers

1
Trade and Conflict A Sectoral Perspective

• Christian Helmers
• Wolfson College, Oxford University
• Oxford, UK
• Jean-Michel Pasteels
• International Trade Centre (UNCTAD/WTO)
• Geneva, Switzerland
• Jurgen Brauer
• Augusta State University
• Augusta, GA, USA
• June 2006
• 10th Annual Defense Economics Conference
• Thessaloniki, Greece

2
Overview

• Motivation and recent literature
• Theory and estimation strategy
• Results and interpretation

3
Motivation

• New try of getting at the conflict gt trade
  relationship
• uses panel data, 1992-1999 (rather than
  cross-section data)
• uses sector-level data (rather than aggregated
  data)
• uses a graduated conflict variable (rather than
  binary)
• takes account of possible endogeneity and
  simultaneity issues

4
Prior literature

• Conflict and trade literature
• Polachek morning session
• Glick and Taylor (NBER Working Paper 2005)
• Martin, Mayer, and Thoenig (CEPR Discussion Paper
  2005)
• Gravity equation literature
• Anderson and van Wincoop (2003)
• Helpman, et al. (2004)
• Santos Silva and Tenreyro (2004)
• Baier and Bergstrand (2005)

5
Prior literature

• Glick and Taylor (NBER Working Paper, 2005)
•  Conventional wisdom in economic history
  suggests that conflict between countries can be
  enormously disruptive of economic activity,
  especially international trade. Yet nothing is
  known empirically about these effects in large
  samples. We study the effects of war on bilateral
  trade for almost all countries with available
  data extending back to 1870. Using the gravity
  model, we estimate the contemporaneous and lagged
  effects of wars on the trade of belligerent
  nations and neutrals, controlling for other
  determinants of trade. We find large and
  persistent impacts of wars on trade, and hence on
  national and global economic welfare. A rough
  accounting indicates that such costs might be of
  the same order of magnitude as the direct costs
  of war, such as lost human capital, as
  illustrated by case studies of World War I and
  World War II. 

6
Prior literature

• Martin, Mayer, and Thoenig (CEPR Discussion
  Paper, 2005)
• We show that the intuition that trade promotes
  peace is only partially true even in a model
  where trade is beneficial to all, war reduces
  trade and leaders take into account the costs of
  war. When war can occur because of the presence
  of asymmetric information, the probability of
  escalation is indeed lower for countries that
  trade more bilaterally because of the opportunity
  cost associated with the loss of trade gains.
  However, countries more open to global trade have
  a higher probability of war because multilateral
  trade openness decreases bilateral dependence to
  any given country. we test our predictions on a
  large dataset of military conflicts in the period
  1948-2001. We find strong evidence for the
  contrasting effects of bilateral and multilateral
  trade. Our empirical results also confirm our
  theoretical prediction that multilateral trade
  openness increases the probability of war
  between proximate countries. This may explain why
  military conflicts have become more localized and
  less global over time.

7
Theoretical baseline
Basic formulation of gravity equation in a
stochastic logarithmic form (Anderson and van
Wincoop, 2003)
where i - the exporting country j - the
importing country Xij - trade from country i to
country j dij - distance between i and j tij -
trade costs (e.g., tariffs) Yi, Yj - country is
and js GDP respectively - country is
and js price indices s - elasticity of
substitution
8
Description of panel data

• Country sample 112 exporters 126 importers
• Time period 1992-1999 (8 years)
• Sector-level data 27 mfg sectors (3-digit level
  ISIC-rev2)
• Variables
• dependent variable
• trade data 1992-1999
• independent variables
• bilateral conflict (HIIK)
• distance
• market access measure (tariff)
• manufacturing production data (27 sectors)
• common language variable
• common border dummy
• colony dummies
• other specific country-pair dummies
• other variables

9
Sector definition
ISIC code ISIC code 300 - Total
manufacturing 353 - Petroleum refineries 311 -
Food products 355 - Rubber products 313 -
Beverages 356 - Plastic products 314 -
Tobacco 361 - Pottery, china, earthenware 321
- Textiles 362 - Glass and glass products 322
- Wearing apparel, except footwear 369 - Other
non-metallic mineral products 323 - Leather
products 371 - Iron and steel 324 - Footwear,
except rubber or plastic 372 - Non-ferrous
metals 331 - Wood products, except furniture 381
- Fabricated metal products 332 - Furniture,
except metal 382 - Machinery, except
electrical 341 - Paper and paper products 383 -
Machinery, electric 342 - Printing and
publishing 384 - Transport equipment 351 -
Industrial chemicals 385 - Professional
scientific equipment 352 - Other chemicals 390
- Other manufactured products
10
Estimation strategy

• Estimate POLS (pooled OLS) (conflict gt trade)
• Endogeneity possible
• random effects vs. fixed effects model Hausman
  test
• Simultaneity possible (conflict gt trade gt
  conflict)
• logit/probit random vs. fixed effects
  Hausman-type test
• System of simultaneous equations (3SLS)

11
Baseline regression pooled OLS
where i exporting country j importing
country k sector t year X trade from
country i to country j in sector k in year t D
distance between i and j Tariff bilateral market
access measure (for trade from i to j) by
sector/year Border i and j are neighboring
countries (1) or not (0) Language bilateral
measure of common language Conflict bilateral
measure of conflict between i and j in year
t Production joint value of production of sector
k in year t Price price index of exporting
country i in sector k in year t EXrate exchange
rate between i and j in year t Colony country i
former colony of country j ComCol country-pair
sharing same colonizer d year dummies (1994 is
the base)
12
Baseline regression pooled OLS
13
Baseline regression pooled OLS

• Country-pair effects not taken into account in
  the POLS
• For instance, specific country-pairs with above
  median
• conflict intensity and above median trading
  volume
• EGYSDN, RUSGEO, PHL-CHN, IRL-GBR, and PAK-IND
• gt endogeneity problem

14
Endogeneityfixed vs. random effects

• The results seem fine but about reversed
  causality?

15
Simultaneity logit/probit fixed vs. random
effects
Note Conflict transformed into binary variable
0 gt 0 gt0 gt 1 robust standard errors

• Simultaneity problem unlike Glick and Taylor
  (2005)

16
Simultaneous equations (3SLS)

• Structural form equations

• Reduced form equation A

• Estimated trade equation B

where
17
3SLS results
Estimated trade equation B results for
conflict equation A not shown
Insert here results
18
Sector-level results for estimated trade equation
B
19
Conclusion

• Major questions
• 3SLS does not include fixed effects (due to
  computational limitations)
• what instruments to use for conflict in a trade
  equation? Lagged values?
• how to interpret the sector-specific estimates?
• what about primary and tertiary sectors?
• how to interpret changes in conflict variable?