Asia/Pacific Regional Trade Agreements:
An Empirical Study
By
Bhavish Jugurnatha and Mark Stewarta* and Robert Brooksb
a RMIT University, School of Economics, Finance and Marketing, Melbourne
3000, Australia
b Department of Econometrics and Business Statistics, Monash University,
Caulfield East 3145, Australia
* Corresponding author.
Dr. Mark Stewart
RMIT University
Building 108, Level 12, 239 Bourke Street, Melbourne
GPO Box 2476V, Melbourne VIC 3001, AUSTRALIA
Tel + 61 3 9925 5879 Mobile +61 419 514 521 Fax 61 3 9925 5986
Email
[email protected]
Asia/Pacific Regional Trade Agreements:
An Empirical Study
Abstract
At the same time as the General Agreement on Tariffs and Trade (GATT) and
the World Trade Organization (WTO) have been encouraging trade liberalized,
there has been a proliferation of Regional Trade Agreements (RTAs). These
RTAs also aim to reduce trade barriers, but they do so it in a preferential
way. There is continued debate as to whether such RTAs are an effective way
of achieving free trade, or if increased trade among members causes less
trade with non-member countries? If RTAs increase total trade, this is
known as 'trade creation', whereas if the extra trade occurs at the expense
of non-members, this is called 'trade diversion'. Trade creation implies
improved welfare, whereas 'trade diversion' may adversely affect welfare.
This paper examines five different RTAs using a gravity model to see if
they have been trade creating or trade diverting. Annual data from 26
countries covering five RTAs in the Asia and Pacific region for the years
1980 to 2000 was used.
The results show that the effects of the different RTAs varied remarkably.
The Association of South East Asian Nations (ASEAN) and the Australian and
New Zealand Closer Economic Relations (CER) fostered greater trade with
trading partners and with the rest of the world. While the Asian Pacific
Economic Cooperation (APEC), the Southern Cone Common Market (MERCOSUR) and
the North American Free Trade Association (NAFTA) tended to be trade
diverting, that is, they expanded intra-bloc trade at the expense of trade
with others.
JEL classification: F1; F14; F15.
Keywords: Intra-Regional Trade, Gravity Models, ASEAN, CER, APEC, MERCOSUR,
NAFTA, Trade Creation and Trade Diversion.
________________________
1 Introduction
This paper addresses the question of whether Regional Trade Agreements
(RTAs) enhance welfare. This is examined by using a gravity model to
analyse the effect the Association of South East Asian Nations (ASEAN),
Australian and New Zealand Closer Economic Relations (CER), the Asian
Pacific Economic Cooperation (APEC), the Southern Cone Common Market
(MERCOSUR) and the North American Free Trade Association (NAFTA) have had
on the trade of both members and non-members of these RTAs.
'Trade creation' occurs when the introduction of an RTA allows an importing
country to purchase products at lower cost than was previously the case. In
contrast, 'trade diversion' is the substitution of a more costly source of
supply within an RTA for a less costly source outside. As the introduction
of an RTA will generally have both 'trade creation' and 'trade diversion'
effects, it is the net affect that needs to be assessed when deciding
whether an RTA hinders or enhances welfare. This paper tests for the
existence of 'trade creation' and 'trade diversion' as a result of RTAs by
using dummy variables within the context of a gravity model.
The rest of the paper is structured as follows. Section 2 describes the
analysis of RTAs. Section 3 provides a description of the data used, as
well as explaining the estimation procedures. Section 4 reports and
discusses the empirical findings, and finally section 5 provides some
concluding remarks.
2 Regional Trade Associations.
Initially economists saw Regional Trade Agreements (RTAs) as welfare
enhancing, as they were a step toward free trade. That is, as long as an
RTA did not increase trade barriers to non-members they were thought to
improve welfare. However, Viner's 1950 paper changed this idea by noting
that RTAs lead to both 'trade creation' and 'trade diversion'. Trade
creation occurs when the establishment of an RTA allows an importing
country to purchase products at lower cost than was previously the case.
Clearly this benefits both the importing country and the world as a whole.
In contrast, 'trade diversion' is the substitution of a more costly source
of supply within the RTA for a less costly source outside, and this would
negatively affect welfare. As an RTA will generally have both 'trade
creation' and 'trade diversion' effects, it is the net affect that needs to
be assessed when deciding whether it enhances welfare.
Johnson (1960) developed a partial equilibrium diagram that explained the
economic effects of 'trade division' and 'trade creation'[i]. The diagram
can be used to show that the affect of an RTA will be the sum of several
effects, and that in markets where trade is diverted countries may be
better or worse off. More recently computable general equilibrium models
have been used, and the results using these techniques confirm Johnson's
conclusions. Lloyd and MacLaren (2004) have an excellent survey article
that summaries the developments in the theoretical analysis of RTAs, while
Low (2003) discusses some of the practical issues relating to RTAs.[ii]
3 Methodology
3.1 Gravity Models
The most common empirical tool used to estimate the effects of RTAs is a
gravity model. A gravity model involves regressing trade on a series of
explanatory variables, then using dummy variables to ascertain whether this
relationship is affected by the existence of RTAs. Beckerman (1956),
Anderson (1979), Bergstrand (1989), Oguledo and Macphee (1994), Frankel and
Wei (1995), Kreinin and Plummer (1998), Krueger (1999), Cernat (2001),
Haveman, Nair-Reichert and Thursby (2003), Adams, Dee and Gali (2003),
Filippini and Molini (2003) and Tang (2005) are just some of the studies
which have used gravity models to estimate the 'trade creation' and 'trade
diversion' effects of various RTAs, including NAFTA, MERCOSUR, CER and
ASEAN.
Generally, studies on regional trading blocs find that trade volume is
directly related to the economic and physical size (GDP, population, land
area) of the countries involved, as well as transaction costs which are
usually proxied by such things as distance and cultural similarities (a
common language is often used for this). This paper examines these factors,
as well as adding the cost variables of the exchange rate and taxes to the
list.
Tinbergen (1962) and Linnermann (1966) provide initial specifications for
the gravity model and use it to look at the determinants of trade flows,
while Aitken (1973) was one of the first to apply this approach to
analysing RTAs. Others to have done this include Bayoumi and Eichengreen
(1997) and Frankel (1997), both of whom examined the effect RTAs had on non-
members as well as members. That is, these papers tried to separate the
'trade creation' and 'trade diversion' effects of RTAs.
3.2 Dependent Variables[iii]
Imports: Although gravity models typically employ total trade (imports plus
exports) as the dependent variable, this paper focuses on imports as they
more closely proxy the effects of domestic trade barriers.
3.3 Independent Variables[iv]
Gross Domestic Product, GDP: The model includes two GDP variables; GDPi is
for the importing country and GDPj is for the exporting country. As income
and output in each country increase, there would be both greater demand for
goods and services as well as increased production, therefore positive
relationships between both of these variables and imports to country i
would be expected.
Population, POP: Again there are two population variables, POPi for the
importing country and POPj for the exporter. It is anticipated that
countries with larger populations will both import and export more.
However, as was suggested by Aitken (1973), the larger is a country's
population the larger will be the ratio of the domestic market to the
foreign market, hence the smaller would be the potential export supply.
However, Bergstrand (1989) pointed out an inconsistency in this argument. A
larger population would allow for economies of scale, which may increase
the price competitiveness of the export country's production, thereby
leading to higher exports. Therefore, the sign on the coefficient of the
population of the exporting country (POPj) may be indeterminate, while the
sign for the importing country (POPi) is expected to be positive.
Distance, DIST: The physical distance between trading countries is a proxy
for transport costs. It is expected that transport costs would be
negatively correlated with trade. DISTij is the geographical distance
between the capital cities of the importing country i and the exporting
country j.
Surface Area, AREA: This is a country's total land area, including areas
under inland bodies of water and some coastal waterways. AREAi is the
surface area of the importing country, while AREAj is for the exporting
country. Generally, it is expected that larger countries will both export
and import more. However, it is possible that relative size may also be
important for comparative advantage reasons. This being the case the sign
on the AREA coefficients may also be indeterminate.
Exchange Rate, EXR: EXRi is the exchange rate of the importing country,
while EXRj is for the exporter. This paper uses a similar approach to that
of Soloaga and Winters (2001) when measuring this variable. EXR is defined
as the local currency value of one $US multiplied by the US GDP deflator
divided by the GDP deflator of the country in question. For country i an
increase in EXRi would indicate either a depreciation of the local currency
or a fall that country's relative prices, as such a negative coefficient
would be expected. For country j a positive coefficient would be expected.
Taxes on goods and services, TAX: TAXi is the International Monetary Fund
trade tax index for the importing country, while TAXj is the equivalent
index for the exporting country. As all taxes (other than lump sum taxes)
are distorting, a negative relationship would be expected for both of these
variables and imports to country i.
Language, LANG: As stated by Linnermann (1966) a common language between
two countries will influence the volume of trade. Therefore LANGij was
assigned a value of one if the two trading countries had the same official
language and zero otherwise. The hypothesis is that countries will trade
more when they have a common language, so a positive coefficient is
expected.
3.4 Model Specification
Equation 1 shows how this paper uses the above variables in the basic
gravity model.
logIMPORTijt = α0 + α1logGDPit + α2logGDPjt+ α3logPOPit + α4logPOPjt+
α5logDISTij + α6logAREAi + α7logAREAj+ α8logEXRit + α9log EXRjt
+ α10TAXit + α11TAXjt + α12LANGij + µijt
(equation 1)
A log specification was used as this has typically given the best results
and i is the importing country, while j is the exporting country.
To examine the effects of RTAs within the framework of this equation, dummy
variables were then added. When others have used this technique a consensus
has emerged that RTAs are generally trade creating. For example, Aitken
(1973), Bergstrand (1985) and Thursby and Thursby (1987) showed that the
European trading blocs increased trade during the 1960s and 1970s. Later,
work by Frankel and Wei (1995) and Frankel (1997) found evidence of 'trade
creation' in Asian and in North American trading blocs. While a recent
paper by Rose (2000) also found that regional trade arrangements, in
general, were trade creating. However, Hassan (2001) found that the South
Asian Association for Regional Cooperation (SAARC)[v] countries as a whole
traded less with the outside world than would be expected.
To examine the effects of RTAs the basic gravity model outlined in equation
1 was extended using dummy variables. Table 1 shows the list of included
RTAs, their members and starting years. The dummy variables took the value
of one if a country was a member of an RTA and zero otherwise, regardless
of the membership status of the trading partners. Equation 2 was used for
this.
(equation 2)
RTAki is unity when the importing country i is a member of the trading bloc
k and zero otherwise. A positive coefficient on this variable (α14 > 0)
implies that countries that are members of RTAs will import more than an
equivalent country that is not a member, and this would indicate that this
RTA is trade creating.
The second dummy variable, RTAkj takes the value of one if the exporting
country j belongs to RTA k and zero otherwise. If α15 > 0, this means that
countries import more from other countries that are members of RTAk, and
again this implies that RTAs are trade creating. That is, positive
coefficients on α14 and α15 imply 'trade creation', whereas if these
coefficients are negative, this means that trade is lower, therefore the
formation of trade blocs may be trade diverting.
If the dummy variable which is the product of RTAki and RTAkj is unity,
this means that both the importing country i and the exporting country j
are members of the same RTA k. Therefore, if α13 > 0 this means that
countries import more from countries that are members of the same RTA. An
indication that an RTA is trade diverting would be that α14 and α15 are
significant and negative, while α13 is significant and positive.
4 Estimation Procedure and Results
The data used in this paper covers a period of 20 years from 1980 to 2000
and includes 26 countries from five RTAs (ASEAN, CER, APEC, the Southern
Cone Common Market or MERCOSUR and NAFTA).[vi]
The model was estimated for four sub-periods; 1980 to 1985, 1985 to 1990,
1990 to 1995 and 1995 to 2000. The pooled data was used to estimate a
single regression equation, allowing for the coefficients to be different
in the four group periods. The model was also estimated for the entire
period using the full panel.[vii] The results for all of these equations
are presented in Tables 2 and 3.
4.1 The Basic Gravity Model Results.
Table 2 reports the basic gravity model results when estimated using panel
data for the four groups of observations as well as for the entire sample
of 1980 to 2000. In general, these equations fit the data well, indicating
that the proposed explanatory variables were significantly related to
bilateral trade. The coefficients of determination (R2) range between 79
per cent and 81 per cent. The F-test (p-value) results show that
collectively the models were highly significant. These results are in line
with the usual gravity model findings from other papers.
Table 2 shows that the coefficients on GDPi and GDPj are all significant
and positively signed, as are most of the population coefficients. That is,
rich more populace countries tend to trade more. Also, as expected the
coefficients on the distance variable, DISTij were all negative and
significant. This suggests that transport costs (proxied by geographical
distance) play an important role in determining the volume of trade between
countries.
The coefficients on AREAi and AREAj were all negatively signed and were
almost always significant. Frankel (1997) also found this to be the case,
and he suggested it was because large countries have more natural resources
and tend to trade less with others. The coefficients on LANGij were all
positive and significant. This implies that a common language and therefore
'cultural similarities' results in trade contacts being easier to make.
When it came to the exchange rate, recall that an increase in EXRi or EXRj
infers a depreciation of the real exchange rate. In line with expectations
the coefficients associated with the importing country, EXRi are all
significant and negative, while the EXRj coefficients are all significant
and positive.
The coefficients on TAXi (the importing countries trade tax index) were
mostly negative and significant for the later years 1990 to 1995, 1995 to
2000 as well as for the whole period 1980 to 2000, while they were
insignificant for the early years. This implies that tax regimes have
become more important in determining trade flows since the 1990s.
Similarly, the coefficients on the exporting country's tax rate TAXj were
negative and significant for the years 1980 to 1985, 1990 to 1995 and for
1980 to 2000. These results generally imply that higher taxes reduce trade,
which is in line with expectations. These tax related trade issues are
discussed in detail by Whalley (2002).
4.2 The Gravity Model Including RTAs.
To address the main questions in this paper, the analysis focuses on the
estimated coefficients associated with the 'trade creation' and 'trade
diversion' effects of RTAs. Table 3 reports the extended gravity model
results, including regional dummy variables, carried out for the four
groups of observations; 1980 to 1985, 1985 to 1990, 1990 to 1995 and 1995
to 2000, as well as for the full panel from 1980 to 2000.[viii] Again, in
general the regression equations fit the data well indicating that the
proposed explanatory variables are significantly related to bilateral
trade. The coefficients of determination (R2) range between 81 per cent and
86 per cent and the F-test (p-value) results show that collectively the
models are highly significant and explain a large portion of the variation
in the data.
4.2.1 ASEAN.
Concentrating on ASEAN, the cross sectional results in table 3 show that
the estimated coefficients on ASEANi were all were positively signed and
statistically significant. This means that ASEAN countries import more than
would be the case if they were not members of that RTA. Similarly, the
ASEANj coefficients are positive and significant, which means that ASEAN
counties also tend to export more. For the ASEAN intra-bloc trade variable
(ASEANij) the coefficients were mostly negative and significant. The
implication is that ASEAN countries have no preference for trade with other
ASEAN members; in fact these results imply that they trade less with each
other. A possible explanation for this may be that the ASEAN countries have
similar characteristics so comparative advantage could see them looking
elsewhere for countries to trade with.
Notwithstanding the sign of the ASEANij coefficient, all of these results
seem to be exactly what would be expected from such strongly outward
oriented economies that make up the ASEAN RTA, as such the conclusion is
that ASEAN is trade creating.
These conclusions are in line with Elliot and Ikemoto (2003) and Tang
(2005), as they both also used gravity models to show that the formation of
ASEAN and the 1992 signing of the AFTA (ASEAN Free Trade Agreement) did not
cause 'trade diversion'.
4.2.2 CER.
The results in table 3 indicate that the effects of CER were less
conclusive than was the case for ASEAN. The CERi coefficients were all
positive, but were only significant for the period 1985 to 1990, as well as
for the entire panel of 1980 to 2000. All of the export coefficients (CERj)
were positively signed and significant. The coefficients for the intra-bloc
variable CERij were also all positive, but were only significant for the
last two periods, as well as for the entire panel. The inference seems to
be that CER had a general 'trade creation' effect, but it has also caused
Australian and New Zealand's preference for trade with each other to
broadly increase. This is analogous to the findings of Frankel (1997) and
Tang (2005) who also found the intra-bloc trade effects of the CER to be
positive and significant. Although, like the current paper, Tang also found
that CER was associated with Australia and New Zealand increasing their
trade with the rest of the world.
4.2.3 APEC.
The estimates for APEC indicate that this RTA is trade diverting, rather
than trade creating. The coefficients on APECi and APECj are mostly
negative and significant, while all of the coefficients for the intra-APEC
trade variable (APECij) are positively signed and significant. This
indicates a strong intra-bloc effect with trade flowing more intensely
among economies that are members of APEC than with the rest of the world.
This suggests that APEC has not been achieving its goal of open
regionalism, and is in fact 'trade diverting', which implies reduced
welfare.
This finding for APEC is in line with the estimates of Frankel (1997) who
attributed the large positive coefficients on the intra-bloc variables to
the fact that a large share of total world trade is accounted for by APEC
member economies. Given that there has not been substantial trade reform
among APEC, it is likely that the size effects may dominate, and as such
this is likely to be a good explanation.
4.2.4 MERCOSUR.
The results here show that MERCOSUR has been clearly trade diverting. All
of the import and export coefficients (MERCOSURi and MERCOSURj) are
negative and significant, while all of the intra-RTA coefficients
MERCOSURij are positive and significant. This shows that since its
inception in 1991 MERCOSUR has increased trade among its members at the
expense of trade with the rest of the world, thereby reducing welfare. This
result supports the findings of Yeats (1998) and Clarete, Edmond and
Wallack (2003) who attribute the rise of intra-bloc MERCOSUR trade to the
introduction of discriminatory tariffs against non-members. Yeats further
noted that the intra-regional trade and export growth among members was
concentrated in products that were not competitive outside of the region.
4.2.5 NAFTA.
These results suggest some form of export diversion, as the NAFTj
coefficients are significant and negative, while the intra-region variable
NAFTij coefficient is positive and significant for 1995 to 2000.[ix] None
of the other coefficients are significant. These results are generally in
line with the study by Fukao, Okubo, and Stern (2003) who identified
NAFTA's 'trade diversion' effects for textiles and apparel imports into the
US, as Mexican products are substituted for Asian goods. In general,
however, it can be argued that these results show that NAFTA has had little
effect during the period covered in this study, which is in line with
Kruger's 1999 conclusion that as yet NAFTA has not had much effect on trade
patterns. More recently Tang (2005) also concluded that the formation of
NAFTA has had no significant effect on trade flows.
5 Conclusion
This study uses a gravity model to examine bilateral trade involving five
trading blocs, with data from 26 countries from 1980 to 2000. The estimated
coefficients from the basic gravity model show that GDP, population,
distance between trading partners, as well as cultural similarity (a common
language) and physical area explain much cross country trade.
The study also uses some price variables, namely the real exchange rate and
taxes and finds that the empirical results line up with prior expectations.
That is, real exchange rate movements had the expected affects, as
depreciations encouraged exports and discouraged imports. With regard to
taxes, the results suggest that taxation decreases bilateral trade.
The regression estimates for the effects of the different RTAs varied
remarkably. ASEAN and CER were found to foster greater trade worldwide and
so were welfare enhancing. However, although APEC, MERCOSUR and NAFTA
tended to expand intra-bloc trade, to some extent this was at the expense
of their trade with the rest of the world, which implies 'trade diversion'
and a loss of welfare.
References
Adams, R., P. Dee, J. Gali and G. McGuire. 2003. Trade and Investment
Effects of Preferential Trading Arrangements – Old and New Evidence.
Productivity Commission Staff Working Paper, Canberra.
Aitken, N. 1973. The Effect of the EEC and EFTA on European Trade: A
Temporal Cross-Section Analysis. American Economic Review, 63, 881 to
892.
Anderson, J. 1979. A Theoretical Foundation for the Gravity Equation.
American Economic Review, 69, 106 to 115.
Anderson, J. and D. Mercouillier, 1999. Trade Insecurity and Home Bias: An
Empirical Investigation. National Bureau of Economic Research Working
Paper No. 7000.
Bayoumi, T. and B. Eichengreen, 1997. Is Regionalism Simply a Diversion:
Evidence from the Evolution of the EC and EFTA. Chicago: University of
Chicago Press.
Beckerman, W. 1956. Distance and the Pattern of Intra-European Trade. The
Review of Economics and Statistics, 28, 31 to 40.
Bergstrand, J. H. 1985. The Gravity Equation in International Trade: Some
Microeconomic Foundations and Empirical Evidence. Review of Economics
and Statistics, 67, 474 to 481.
Bergstrand, J. H. 1989. The Generalised Gravity Equation, Monopolistic
Competition, and the Factor Proportions Theory in International Trade.
Review of Economics and Statistics, 71, 143 to 153.
Bhagwati, J. and A. Panagariya. 1996. Preferential Trading Areas and
Multilateralism- Strangers, Friends or Foes? In J. Bhagwati and A.
Panagariya (Eds.), The Economics of Preferential Trade Agreements.
Washington: American Enterprise Institute and AEI Press.
Cernat, L. 2001. Assessing Regional Trade Arrangements: Are South-South
RTAs More Trade Diverting? Policy Issues in International Trade and
Commodities Study Series. No. 16 UNCTAD.
Clarete, R., C. Edmond and J. S. Wallack. 2003. Asian Regionalism and its
Effects on Trade in the 1980s and 1990s. Journal of Asian Economics,
14, 91 to 129.
Department of Foreign Affairs and Trade (DFAT). 2004. The APEC Region Trade
and Investment. Retrieved 16.03.2004, from
www.dfat.gov.au/trade/ftas.html
Elliott, R. J. R. and K. Ikemoto. 2004. AFTA and the Asian Crisis: Help or
Hindrances to ASEAN Intra-Regional Trade? Asian Economic Journal, 18,
1 to 23.
Filippini, C. and V. Molini. 2003. The determinants of East Asian trade
flows: a gravity equation approach. Journal of Asian Economics, 14, 695 to
711.
Fitzpatrick, G. and M. Modlin. 1986. Direct Line Distances (International
Edition). The Scarecrow Press, Inc.
Frankel, J. 1997. Regional Trading Blocs in the World Economic System.
Washington: Institute for International Economics.
Frankel, J. and S. Wei. 1995. Emerging Currency Blocks. Berlin: Springer.
Fukao, K., T. Okubo and R. M. Stern. 2003. An Econometric Analysis of Trade
Diversion under NAFTA. The North American Journal of Economics and
Finance, 14, 3 to 24.
George Philip Limited. 1994. Philip's World Atlas and Gazetteer.
Havemann, J. 2002. International Trade Data. Retrieved 20.08, 2004, from
http://www.macalester.edu/research/economics/PAGE/HAVEMAN/Trade.Resourc
es/TradeData.html
Havemann, J., U. Nair-Reichert and J. Thursby. 2003. How Effective are
Trade Barriers? An Empirical Analysis of Trade Reduction, Diversion
and Compression. Review of Economics and Statistics, 85, 480 to 485.
Hassan, K. 2001. Is SAARC a Viable Economic Block? Evidence from Gravity
Model. Journal of Asian Economics, 12, 263 to 290.
International Monetary Fund. (Various issues). Direction of Trade
Statistics. Washington: IMF Publication Services.
Johnson, H. G. 1960. The Economic Theory of Customs Unions. Pakistan
Economic Review, 10, 14 to 30.
Kreinin, M. and M. Plummer. 1998. Ex Post Estimates of the Effects of the
European Single Market Program on the Exports of Developing Countries.
ASEAN Economic Bulletin, 15, 206 to 214.
Krueger, A. 1999. Trade Creation and Trade Diversion under NAFTA. National
Bureau of Economic Research Working Paper, No. 7429.
Linnermann, H. 1966. An Econometric Study in International Trade Flows.
Amsterdam: North Holland.
Lloyd, P. J. and D. MacLaren. 2004. Gains and Losses from Regional Trading
Agreements: A Survey. The Economic Record, 80, 445 to 467.
Low, L. 2003. Multilateralism, Regionalism, Bilateral and Cross-Regional
Free Trade Arrangements: All Paved with Good Intentions for ASEAN? Asian
Economic Journal, 17, 65 to 86.
Oguledo, V. and C. Macphee. 1994. Gravity Models: A Reformulation and an
Application to Discriminatory Trade Arrangements. Applied Economics,
26, 107 to 120.
Pacific Economic Cooperation Council for APEC. 1999. Assessing APEC
Individual Action Plans and Their Contribution to APEC's Goals,
Singapore. Unpublished manuscript.
Pomfret, R. 1997. The Economics of Regional Trading Arrangements. Clarendon
Press, Oxford.
Rose, A. 2000. One Money, One Market? The Effects of Common Currencies on
International Trade. Economic Policy, 15, 7 to 46.
Schiff, M. and A. Winters. 2003. Regional Integration and Development.
Oxford University Press, Washington, D. C.
Soloaga, I. and L. Winters. 2001. Regionalism in the Nineties: What Effect
on Trade? The North American Journal of Economics and Finance, 12, 1
to 29.
Tang, D. 2005. Effects of the Regional Trading Arrangements on Trade:
Evidence from the NAFTA, ANZCER and ASEAN Countries 1989 - 2000.
Journal of International Trade and Economic Development, 14, 241 to
265.
Thursby, J. and M. Thursby. 1987. Bilateral Trade Flows, the Linder
Hypothesis and Exchange Risk. Review of Economics and Statistics, 69,
488 to 495.
Tinbergen, J. 1962. Shaping the World Economy: Suggestions for an
International Trade Policy. New York.
Whalley, J. 2002. Taxes and Trade. Working Paper Series, The University of
Western Ontario.
World Bank. 2001. World Development Indicators (WDI) Online database.
Retrieved 28.04.2004, from
http://devdata.worldbank.org.mate.lib.unimelb.edu.au/dataonline/
Viner, J. 1950. The Customs Union Issue New York. Carnegie Endowment for
International Peace.
Yeats, A. 1998. Does MERCOSUR's Trade Performance Raise Concerns about the
Effects of Regional Trade Agreements? The World Bank Economic Review,
12, 1 to 28.
Table 1: List of Regional Trade Agreements (RTAs).
"PTA "Date "Member Countries "
" "of " "
" "Entry " "
"ASEAN "1967 "Indonesia, Malaysia, Philippines, "
"(Association of South " "Singapore and Thailand "
"East Asian Nations) " " "
" " " "
"APEC "1989 "Australia, Canada, Chile, China, Hong "
"(Asia-Pacific Economic " "Kong, China, Indonesia, Japan, Korea, "
"Cooperation) " "Malaysia, Mexico, New Zealand, Papua "
" " "New Guinea, Peru, Philippines, Russia,"
" " "Singapore, Thailand, United States, "
" " "Vietnam "
"CER "1983 "Australia, New Zealand "
"(Australia-New Zealand, " " "
"Closer Economic Relations" " "
"Trade Agreement) " " "
" " " "
"MERCOSUR "1991 "Argentina, Brazil, Paraguay, Uruguay "
"(Southern Cone Common " " "
"Market) " " "
" " " "
"NAFTA "1994 "Canada, Mexico, United States "
"(North American Free " " "
"Trade Agreement) " " "
" " " "
Table 2: Basic Gravity Model Estimates - Pooled Regression Analysis
"Dependent Variable: IMPORT " " " " "
"Method: Pooled Least Squares " " " " "
"Variable "Expected"1980-198"1985-199"1990-199"1995-200"1980-200"
" "signs "5 "0 "5 "0 "0 "
"C " "-12.6956"-13.4453"-12.8364"-13.4753"-13.1686"
" " "(0.0000)"(0.0000)"(0.0000)"(0.0000)"(0.0000)"
" " "*** "*** "*** "*** "*** "
"GDPi "+ "0.9141 "0.9520 "0.8901 "0.9994 "0.9490 "
" " "(0.0000)"(0.0000)"(0.0000)"(0.0000)"(0.0000)"
" " "*** "*** "*** "*** "*** "
"GDPj "+ "1.0748 "1.0385 "0.9582 "0.9791 "0.9968 "
" " "(0.0000)"(0.0000)"(0.0000)"(0.0000)"(0.0000)"
" " "*** "*** "*** "*** "*** "
"POPi "+ "0.1323 "0.0324 "0.0847 "0.0052 "0.0642 "
" " "(0.0011)"(0.3555)"(0.0007)"(0.8383)"(0.0001)"
" " "*** " "*** " "*** "
"POPj "? "-0.1892 "-0.1777 "-0.0223 "0.0632 "-0.0412 "
" " "(0.0000)"(0.0000)"(0.3481)"(0.0071)"(0.0071)"
" " "*** "*** " "*** "*** "
"DISTij "- "-1.2489 "-1.2133 "-1.1087 "-1.1233 "-1.1703 "
" " "(0.0000)"(0.0000)"(0.0000)"(0.0000)"(0.0000)"
" " "*** "*** "*** "*** "*** "
"AREAi "- ? "-0.0449 "-0.0365 "-0.1432 "-0.1956 "-0.1472 "
" " "(0.1868)"(0.2520)"(0.0000)"(0.0000)"(0.0000)"
" " " " "*** "*** "*** "
"AREAj "- ? "-0.0633 "-0.0191 "-0.1185 "-0.1791 "-0.1290 "
" " "(0.0203)"(0.4556)"(0.0000)"(0.0000)"(0.0000)"
" " "** " "*** "*** "*** "
"LANGij "+ "0.1632 "0.1531 "0.1518 "0.0774 "0.1483 "
" " "(0.0000)"(0.0000)"(0.0000)"(0.0032)"(0.0000)"
" " "*** "*** "*** "*** "*** "
"EXRi "- "-4.5973 "-2.3241 "-1.0401 "-1.0573 "-1.5067 "
" " "(0.0550)"(0.0000)"(0.0011)"(0.0005)"(0.0000)"
" " "* "*** "*** "*** "*** "
"EXRj "+ "0.0485 "0.0569 "0.0525 "0.0268 "0.0444 "
" " "(0.0000)"(0.0000)"(0.0000)"(0.0078)"(0.0000)"
" " "*** "*** "*** "** "*** "
"TAXi "- "-0.0431 "0.0139 "-0.0741 "-0.1438 "-0.0756 "
" " "(0.4623)"(0.7756)"(0.0653)"(0.0012)"(0.0021)"
" " " " "* "** "*** "
"TAXj "- "-0.1825 "-0.0358 "-0.0640 "-0.0590 "-0.0895 "
" " "(0.0003)"(0.4086)"(0.0663)"(0.1390)"(0.0001)"
" " "*** " "* " "*** "
"R-squared " "0.7973 "0.8139 "0.8138 "0.8265 "0.8129 "
"Adjusted " "0.7958 "0.8125 "0.8127 "0.8254 "0.8125 "
"R-squared " " " " " " "
"Prob(F-statis" "0.0000 "0.0000 "0.0000 "0.0000 "0.0000 "
"tic) " " " " " " "
"Cross-section" "293 "333 "456 "439 "486 "
"s " " " " " " "
"DW stat " "0.1932 "0.2086 "0.1764 "0.1357 "0.1757 "
"Panel " "1604 "1661 "1981 "1897 "6179 "
"observations " " " " " " "
*p
