Trade led Growth in Times of Crisis Asia Pacific Trade Economists Conference 2 3 November 2009, Bangkok Session 12 Factors Contributing to Export Performance in the Aftermath of Global Economic Crisis Clovis Freire & Mia Mikic Asia Pacific Research and Training Network on Trade www.artnetontrade.org
ARTNet 5th Anniversary Conference Trade-Led Growth in Times of Crisis Factors Contributing to Export Performance in the Aftermath of Global Economic Crisis Clovis Freire & Mia Mikic ESCAP freire@un.org mikic@un.org
After a period of six years of steady export growth, exports were in a free fall contraction much deeper than in any other recent crisis episode 50% 40% Year-on on-year monthly percentage change of exports (by origin) 30% 20% 10% 0% M1 1990 M9 1990 M5 1991 M1 1992 M9 1992 M5 1993 M1 1994 M9 1994 M5 1995 M1 1996 M9 1996 M5 1997 M1 1998 M9 1998 M5 1999 M1 2000 M9 2000 M5 2001 M1 2002 M9 2002 M5 2003 M1 2004 M9 2004 M5 2005 M1 2006 M9 2006 M5 2007 M1 2008 M9 2008-10% -20% -30% -40% 2 World Advanced Economies Emerging and Developing Economies Developing Asia
Contraction of exports was deeper for exports directed to advanced economies. However, even the exports to nonadvanced economies were affected Year-on on-year monthly percentage change of exports (by destination) 3
Which factors contribute to export performance in times of economic crisis? Exploring a simple model Total effect on demand for exports = income effect + substitution effect + e Fall in consumption owing to income reduction Change in the pattern of consumption & demand for cheaper varieties or products increases Sophistication of exports - Hausmann, Hwang and Rodrik (2007) 4
Income effect Country A Basket of exports A is more sophisticated than Result: Exports of Country A are more vulnerable than exports of Country B to reduction of demand for exports EXPY index - Hausmann, Hwang and Rodrik (2007) PRODY EXPY j j = = ( x j j l x X jl j jk ( x / X j ) Y j / X ) jk PRODY j l Country B Basket of exports B Y GDP per Capita xjk value of exports of product k of country j Xj Total exports of country j 5
Substitution effect Country A Basket of exports A GDP per capita A is as sophisticated as is lower than Country B Basket of exports B GDP per capita B Result: Country A is able to produce at a lower price than Country B and it may be less affected by the drop in demand for exports Index of advantage of countries competing in terms of export prices: EXPY / GDP per capita 6
Data Year-on-year monthly percentage change of exports EXPY IMF Direction of Trade Statistics (DOTS) Online HS 1992 (2004 2007) COMTRADE GDP per capita (2005 PPP-adjusted) - World Development Indicators (WDI) (2004 2007) Methodology PRODY - Consistent sample of 128 countries - HS1992 period 2004-06 EXPY EXPY/GDP per Capita 144 countries year 2007 7
log GDP per Capita 2007(PPP 2005) 6 8 10 12 Above the line Less sophisticated than expected given the GDP pc COM SUR MAC NOR SGP HKG IRL US A NLD AUT AUS FI DEU N CAN DNK BEL ISL JPN FRA GBR SWE CHE GRC ITA ESP ISR CZE CYP NZL SVN MLT TTO SAU KOR PRT ATG EST SVK HUN SYC LTU LVA HRV POL CHL MYSBWAMEX RUS ARG KNA TUR PANMUS CRI URY BGR LBN KAZSRB BLR ROU LCA BRA ZAF MKDCOL GRD PER VCT AZE DZA THA BLZ ALB ECUTUN UKR BIH JAM ARM EGY SLV MDV NAM CHN FJIGTM GEO JOR BOL HND MAR PRY SWZ TON VUT IDN MNG PHL GUY CPV NIC IND PAK VNM MDA YEM KGZ MRT STP CIV SEN KEN GHA ZMB GMBTZA BGD MLI UGA RWA MDG MWI ETH TGO MOZ NER BDI LUX Below the line More sophisticated than expected given the GDP pc 0 5000 10000 15000 20000 25000 EXPY 2007 R-squared = 0.5808 log (EXPY/GDP pc) =.0002188 * EXPY2007 + 6.455011 8
Empirical analysis Y = b + b EXPY + b log EXPYGDPr atio + u ij 0 j 1 j Where: Yij is year-on-year percentage change of total exports of a country i in a particular month j i 2 j i i j Results *** p<0.01, ** p<0.05, * p<0.1 What it means? The more sophisticated basket of exports higher impact on the export performance The more sophisticated than it is expected given their level of GDP pc lower impact on export performance When? Income effect Substitution effect 9
Does the direction of exports matter? ij 6 = b0 + b1 EXPYi + b2 log EXPYGD Pr atioi + k = 1 Y b dummymonth + u Where: Yij is year-on-year percentage change of total exports of a country i in a particular month j 3 k k i Results 1) Income and substitution effects are significant to explain variation of exports to advanced and non-advanced economies. 2) Model fits better exports to nonadvanced economies 10
Does the origin of exports matter? 6 ijm = b0 + b1 EXPYi + b2 log EXPYGDPr atioi + k= 1 Y b dummymonth + b dummydestination + u Where: Yij is year-on-year percentage change of total exports of a country i in a particular month j to area m Groups of countries 3k k 4 i a) World Bank income group classification Low, Lower middle, Upper middle and High income countries b) Countries with special needs Least Developed Countries (LDC), Landlocked Developing Countries (LLDC), and Small Island Developing States (SIDS) c) Geographical groupings -Asia-Pacific, Asia-Pacific developing economies, Asia-Pacific developing economies excluding China, Latin America, Caribbean, and Sub-Saharan Africa d) Asia-Pacific subregions -South-East Asia (SEA), East and North-East Asia (ENEA), South and South-West Asia (SSWA), Pacific Islands, and North and Central Asia excluding Russian Federation (CIS Asia + Georgia). 11
World Bank income group classification Significant to Low, Lower-middle, Upper-middle but not to High income countries 12
Countries with special needs Significant to Least Developed Countries (LDC) & Landlocked Developing Countries (LLDC) but not to Small Island Developing States (SIDS) 13
Geographical groupings Significant to Asia-Pacific, Asia-Pacific developing economies, Asia-Pacific developing economies excluding China and Sub-Saharan Africa but not to Latin America & Caribbean. 14
Asia-Pacific subregions Significant to South-East Asia (SEA), East and North-East Asia (ENEA), South and South-West Asia (SSWA) & Pacific Islands but not to CIS Asia 15
Conclusions and policy considerations In times of reduced demand, the pattern of year-on-year percentage change of a country s exports can be partially attributed to the reduction in the consumption of sophisticated products and the substitution of consumption towards lower-priced products Impact was larger on traditional exporters of sophisticated products ; countries that despite lower GDP per capita managed to export equally or similarly sophisticated products have been less affected Result implies an important policy consideration which is intuitive but not often pursued: increasing product sophistication increases economies resilience in terms of export performance in times of economic downturn 16
Your feedback is much appreciated Clovis Freire & Mia Mikic ESCAP freire@un.org mikic@un.org 17