Transcript Martinez Zarzoso (RT Trade and Poverty)x

Distributional Effects of Trade
Policy
Inmaculada Martínez-Zarzoso
University of Göttingen (Germany)
and University Jaume I (Spain)
Outline
•
•
•
•
•
•
Introduction
Theories
Measurement and methods
Globalization versus hiperglobalization
Multilateral versus bilateral trade liberalization
Conclusions
I. Martínez-Zarzoso
Introduction
• A country‘s exposure to international trade affects
not only income but also the distribution of
resources
• Distributional changes, in what direction?
– Developing countries  labor abundant  better off
with more trade? Evidendence indicates otherwise
• Increase in inequality in the last 2-3 decades in
developing countries, why?
• Mechanisms: more exposure  impact on relative
prices  impact on wages and comsumer prices,
but who is affected and by how much?
I. Martínez-Zarzoso
Theories
• 1. Trade liberalization and economic growth link
• 2. General equilibrium approach (Porto, 2006):
trade policy affects household welfare by
changing the relative prices of goods, this change
affects in turn consumption and income.
The model, with some added ingredients can be
used to simulate trade policy changes
These ingredients are:
– Estimates of changes in traded good prices
– Estimates of wage-price elasticities
– Response of non-traded good to traded good prices
I. Martínez-Zarzoso
Measurement and methods
• Measuring trade liberalization
• Direct measures:
– Trade policy barriers from UNCTAD‘s TRAINS data
since 1989  tariff and non-tariff barriers  main
problem is lack of reporting.
– Free Trade Agreements (cross-country studies)
– Behind the border trade barriers from WB
• Indirect measures: Trade volumes (exports and
imports), but not only determined by trade policy
and transport cost, also endogenous variables
play a role (wages, incomes...)
I. Martínez-Zarzoso
Measurement and methods
• Measuring inequality, what to choose? skill premium,
wage inequality, income inequality (Gini), consumption
inequality?
• Income versus consumption (expenditure) data to
measure inequality (to construct Gini)
• Expenditure data from Living Standards Measurement
Surveys (LSMS), WB
• Why consumption is better?
– Captures better life-time well-being
– Income often underreported by households
– Trade policy changes the purchaising power of current
incomes
I. Martínez-Zarzoso
Methods
• Porto (2006) GE Approach
• Differential exposure approach, Goldberg and
Pavcnik (2005): within a country industries and
regions are differentially exposed to trade policies:
– Assumes labor is not perfectly mobile across industries
– Requires much weaker identification, but partial
equilibrium.
• Endogeneity of trade policy: double causality
issues, when are liberalization episodes
exogenous?
I. Martínez-Zarzoso
Globalization versus hiperglobalization
• Why after the mid-1990s things have been
different? The exposure of developing
countries to international markets has
increased substantially:
–
–
–
–
Decrease of trade protection
Increase share of imports and exports in GDP
Magnitude of capital flows: FDI
Exchange rate fluctuations
I. Martínez-Zarzoso
70
60
50
40
50
net income inequality
40
30
20
30
20
10
50
100
Exports (goods services, % GDP)
0
150
50
100
150
Trade (% of GDP)
50
40
20
30
30
40
net income inequality
60
70
50
0
0
•
50
100
150
Exports (goods services, % GDP)
200
250
0
50
100
Trade (% of GDP)
150
Scatter plot before (top) and after (bottom) 1995 for high (left) and low (right) income countries, 1960-2009
200
Multilateral versus bilateral trade
liberalization
• Bilateralization of world trade: proliferation of
regional/bilateral trade agreements in the 2000s
• Is this trend helping the poor redistributing
income?
• Most of them 3rd generation agreements
including:
–
–
–
–
Labor provisions: against “race-to-the-bottom”
Environmental provisions
Trade in services
Economic cooperation
I. Martínez-Zarzoso
Evolution of Regional Trade Agreements in the world, 1948-2012
Source: http://www.wto.org/english/tratop_e/region_e/regfac_e.htm
I. Martínez-Zarzoso
Evolution of Regional Trade Agreements with labour provisions, in the
world, 1997-2012
30
25
20
RTA_labour
15
RTA_labort_Ac
10
5
0
1996
1998
2000
2002
2004
2006
2008
2010
I. Martínez-Zarzoso
2012
2014
Conclusions
• Evidence suggests an increase in globalization and
inequality in developing countries in the recent
past, is this causal?
• Scarce support for the premise that trade
openness in developing countries would favour
the poor
• Adjustment occurs mainly within industries,why?
• Perhaps labour market liberalization also needed
to facilitate sectoral relocation of workers
• Short-run versus long-run effects, what time
frame to use?
I. Martínez-Zarzoso
New directions
• Use narrower measures of inequality
• Change focus from the country to the firm
• Heterogeneity of firms, plants, products and
workers is important
• Use of highly disaggregated data: for plants,
product lines, workers characteristics
• Narrower connection between theory and
empirics
I. Martínez-Zarzoso